MAXIMIZING POSITIVE SYNERGIES - World Health Organization

Table of Contents
Key Acronyms
Introduction The maximizing positive synergies project 1
and the purpose of this document
Burundi Building a health system together with Global 8
Health Initiatives, in the aftermath of war
Johann Cailhol, Thuba Mathole, Annie Parsons,
David Sanders, Denise Kandondo, Innocent Ndayiragije and
Théodore Niyongabo
Cameroon Evaluation of the national programme for access 20
to antiretroviral therapy
Boyer S., Eboko F., Camara M., Abé C., Owona Nguini M.E.,
Koulla-Shiro S., Moatti J-P.
Cameroon Impact of Global Health Initiatives on primary level health 31
care facilities: the case of Kumba and Limbe health districts
Peter M. Ndumbe, Julius Atashili
Central African Impact of the Global Health Initiatives 40
Republic on the health system: perceptions of informants
Julius Atashili, Marie-Claire Okomo, Emilia Lyonga, Nayana Dhavan,
Nikita Carney, Erin Sullivan, Peter Ndumbe
China Impact of the Global Fund HIV/AIDS programmes on 50
coordination and coverage of financial assistance schemes
for people living with HIV/AIDS and their families
Zhang Xiulan, Pierre Miège and Zhang Yurong
Georgia System-wide effects of the Global Fund on Georgia’s 59
health care systems
Ketevan Chkhatarashvili, George Gotsadze, Natia Rukhadze
Ghana Interactions between health systems and Global Fund-supported 65
TB and HIV programmes
Sai Pothapregada, Rifat Atun
Haiti Maximizing positive synergies between Global Health 74
Initiatives and the health system
LC Ivers, JG Jerome, E Sullivan, JR Talbot, N Dhavan, M StLouis,
W Lambert, J Rhatigan, JS Mukherjee
India Exploring how disease-specific programmes in TB and 88
HIV/AIDS interact with health systems
Krishna Dipankar Rao, Mathew Sunil George, Aarushi Bhatnagar,
Babita Rajkumari, Maulik Chokshi, Preeti Kumar and Indrajit Hazarika

Kenya The impact of Global Health Initiatives on the health system 97
Joseph Rhatigan, Erin Sullivan, Kileken ole-MoiYoi, George Kimathi,
Nayana Dhavan, Ephantus Kabiru
Kyrgyzstan Tracking global HIV/AIDS initiatives and their impact 107
on the health system
Gulgun Murzalieva, Julia Aleshkina, Arnol Samiev, Neil Spicer, Gill Walt
Malawi Global Health Initiatives and delivery of health care: 117
the case of the Global Fund
Victor Mwapasa and John Kadzandira
Pakistan Impact of the Global Polio Eradication Initiative on the 124
health system
Sania Nishtar
Peru Effects of the implementation of Global Fund-supported 134
HIV/AIDS projects on health systems, civil society and
affected communities, 2004-2007
Carlos F. Cáceres, Maziel Girón, Clara Sandoval, Roberto López,
José Pajuelo, Rocío Valverde, Patricia Vásquez, Alfonso Silva-Santisteban,
Ana Maria Rosasco
Rwanda Impact of Global Health Initiatives on the health system: 144
a mixed methods analysis
Mukherjee JS, Jerome JG, Sullivan E, May MA, Mayfield A, Lambert W,
Dhavan N, Carney N, Rhatigan J, Ivers LC
Senegal Effects of Global Health Initiatives on the health system 160
Papa Salif Sow, Fatou Francesca Mbow, Aliou Diallo,
Demba Dione, Marième Ba
South Africa The effects of Global Health Initiative funding for HIV/AIDS 168
on the health system
Thomas Bisika, Eric Buch, Thubelihle Mathole, Annie Parsons, David Sanders
Uganda Expanding targeted services into primary health care 181
Dr. Bernard Michael Etukoit, Mr. Richard Wanyama
Ukraine Effects of the Global Fund on the health system 192
Tetyana Semigina
United Republic Strengthening the health system to address 201
of Tanzania the burden of HIV/AIDS and TB
Thyra de Jongh, Rifat Atun
Zambia Global HIV/AIDS initiatives and health system capacity to 209
cope with the scale-up of HIV services
Phillimon Ndubani, Joseph Simbaya, Aisling Walsh, Ruairí Brugha

Policy analysis of the impact of Global Health Initiatives on health systems: 218
policies and lessons learned from donor programmes
Anne Rossier Markus, Seble Frehywot, Amie Heap, Alan Greenberg
Conclusion Directions for future research on positive synergies 236
Annex 01 Data sources for tables in country case studies 240
Annex 02 Contributing Institutions 242
The named contributors alone are responsible for the views expressed in this publication.

Key Acronyms 1
AIDS (SIDA) Acquired immunodeficiency syndrome
ANC Antenatal care
ART Antiretroviral therapy
ARVs Antiretrovirals
CBO Community-based organization
CCM Country coordinating mechanism
CDC U.S. Center for Disease control
CHW Community health worker
CSO Civil society organization
DFID U.K. Department for International Development
DOTS Directly Observed Treatment, short-course
FBO Faith-based organization
GAVI Global Alliance for Vaccines and Immunization
GHI Global health initiative
Global Fund Global Fund to fight AIDS, Tuberculosis and Malaria
GDP Gross domestic product
HIPC Initiative Heavily Indebted Poor Countries Initiative
HIV Human immunodeficiency virus
HMIS Health management information system
IDU Injecting drug user
ILO International Labour Organization
M&E Monitoring and evaluation
MDG Millennium Development Goal
MDR-TB Multidrug-resistant tuberculosis
MOH Ministry of Health
MPS Maximizing Positive Synergies
NACP National AIDS Control Programme
NGO Nongovernmental organization
ODA Official Development Assistance
1 All other abbreviations are spelled out in full in individual country case studies.

OIs Opportunistic infections
OVC Orphans and vulnerable children
PEI Polio Eradication Initiative
PEPFAR U.S. President's Emergency Plan for AIDS Relief
PHC Primary Health Care
PLWHA People living with HIV/AIDS
PMTCT Prevention of mother-to-child transmission
SWAp Sector Wide Strategic Approach
SYSRA Systemic Rapid Assessment Toolkit
TB Tuberculosis
UN United Nations
UNAIDS Joint United Nations Programme on HIV/AIDS
UNDP United Nations Development Programme
UNICEF United Nations Children's Fund
USAID United States Agency for International Development
VCT Voluntary counselling and testing
WHO World Health Organization
WHO NHA WHO National Health Accounts
WHO SIS WHO Statistical Information System (WHOSIS)
WHO WMR WHO World Malaria Report
World Bank MAP Multi-Country HIV/AIDS Program

Introduction: The Maximizing Positive Synergies
project and the purpose of this document
Over the last decade, Global Health Initiatives (GHIs) have mobilized substantial new resources for
health action in many low- and middle-income countries. The expansion of key services,
particularly the provision of HIV/AIDS treatment, has been striking, and millions of people have
benefited. But the scale-up of selected services by GHIs has placed new demands on national
health systems, revealed weaknesses in those systems, and rekindled debates on how countries
can best combine disease-specific programmes with broader agendas to improve the health of
their people.
Addressing knowledge gaps—at a critical time
In July 2009, policymakers of the G8 nations gathering in Italy will review policy and funding
priorities for global health at a critical moment. As the 2015 target date of the Millennium
Development Goals (MDGs) approaches, the global economic crisis threatens to slow recent health
gains and inflict its harshest effects on the world’s poorest countries and people – the very
communities that the MDGs were designed to help. Yet the opportunity exists to accelerate health
progress by sustaining ambitious global investments in health and ensuring that resources are
directed strategically for maximum impact. Understanding interactions between GHIs and health
systems is crucial to reach this objective.
Until recently, little solid evidence was available to guide policymakers and programme
implementers seeking practical, proven ways to shape interactions between disease-specific
programmes and health systems so as to achieve the greatest gains in health. In May 2008, the
World Health Organization (WHO) launched a process to generate evidence-informed guidance on
this issue for GHIs, countries, and global health partners. The “Maximizing Positive Synergies”
(MPS) project has engaged stakeholders in a collaborative effort to build new knowledge on how
GHI-supported programmes are impacting national health systems, and to harness this evidence
for policy and implementation.
This draft document presents an initial compilation of findings from MPS research partners who
have analysed GHI-health systems interactions in more than 20 countries. It offers a new body of
data to inform policy and guide strategic action by national authorities, GHIs and implementers.
Research partners are continuing to analyse their data; the results summarized here are
provisional. As analysis, validation and comparison proceed on the wealth of data gathered, MPS
country studies will provide a robust evidence base to support WHO’s normative guidance on
disease-specific programmes and health systems. In addition, the initial phase of MPS learning
summarized here has clarified directions for future research that will continue to generate fresh
evidence for policy.
Structure of this report
The introductory section of this document summarizes the main research questions, conceptual
framework and methodologies used in MPS country-level research. It describes how countries
were selected for inclusion in the MPS studies and indicates limitations of the work to date. The
1

core section of the document then presents 21 country case studies on GHI-health systems
interaction, along with a comparative analysis of the internal structures and policies of four major
GHIs: the GAVI Alliance (formerly the Global Alliance for Vaccines and Immunization), the Global
Fund to Fight AIDS, TB and Malaria (Global Fund), the United States President’s Emergency Plan for
AIDS Relief (PEPFAR) and the World Bank’s Multi-Country AIDS Programme (World Bank MAP). The
conclusion of the report outlines priorities for an ongoing agenda of research on GHIs and health
systems.
Context and long-range goals
Recent public health history has been marked by: (a) recurrent tensions between vertical and
horizontal models of disease control [1,2]; (b) ongoing debates about whether health should be
understood primarily as a national concern or, instead, as a global responsibility [3,4]. The rise of
GHIs has underscored the importance of these discussions, but also the need to frame problems
and solutions in new ways [5-9].
The MPS project represents an initial step towards a pragmatic, empirical resolution of these
entrenched debates. In the project’s opening phase, MPS researchers have collected data to
provide decision-makers with a clearer picture of how disease-focused programmes and national
health systems are influencing each other in country contexts. Subsequent work will analyse the
causal factors that lie behind the observed interactions, enabling the progressive delineation of
policy options and delivery strategies that can maximize health gains by aligning the distinctive
strengths of GHIs and national systems.
Along with the content of its findings, MPS has emphasized a process for generating, validating,
disseminating and operationalizing knowledge about health systems. As a worldwide
collaborative approach to health systems learning, the MPS research alliance builds upon and
advances the strategies pioneered in recent years by innovative regional and global research
networks, alliances and observatories 2 . As such, the MPS effort is yielding lessons that may
ultimately be useful to countries at all levels of income and at all stages in the development of
their health systems.
MPS research questions and conceptual framework
Maximizing Positive Synergies has pursued the following overarching research question:
“How can GHIs and national health systems optimize their interactions to capitalize on
positive synergies and minimize negative impacts thereby achieving their common goal of
improving health outcomes?” [10]
2 For example, the GHIN Network, www.ghinet.org, the Health Systems Knowledge Network of the WHO Commission on
Social Determinants of Health, (http://www.who.int/social_determinants/themes/healthsystems/en/index.html, and the
European Observatory on Health Systems, (http://www.euro.who.int/observatory).).
2

Specific sub-research questions pursued by MPS research partners included the following:
• How are major GHIs interacting with health systems at national and sub-national levels?
How are these interactions affecting the main components or “building blocks” of
countries’ health systems?
• How are they influencing processes such as scale-up of services; coordination of services;
and harmonization of donor priorities and activities?
• In selected local settings, is GHI support translating into impacts at the facility level? What
initial lessons emerge for improving policy and service delivery?
• Are the major GHIs interacting with health systems in similar ways, or can significant
differences among them be observed?
• How are major GHIs engaging civil society and communities? What is the role of civil
society and community organizations in strengthening synergies between GHIs and health
systems?
What specific areas of interplay between GHIs and health systems appear to be most
critical/strategic in accelerating action to improve health outcomes?
In formulating their research strategies and analyzing results, MPS researchers have adopted a
conceptual framework informed by WHO’s “building blocks” approach to health systems
strengthening. In an ongoing effort to clarify the problems, needs, expected outcomes, and key
variables in health systems performance, the WHO building blocks stress common elements that
recur in every health system and must work in concert if services are to be delivered effectively. In
their original formulation the building blocks included service delivery; health workforce;
information; medical products, vaccines and technologies; financing; and stewardship—meaning
leadership, governance and the fulfilment by officials and other professionals of their
responsibilities as guardians of the right to health [11]. To this list of fundamental systems
components, MPS researchers have added community and civil society participation as a further
critical dimension of health action.
In an iterative process, MPS researchers elaborated the WHO building blocks into a conceptual
framework that recognizes how the functioning of the building blocks is impacted by contextual
factors [5-7], as well as how systems components interact with and “feed back” upon each other
(Fig. 1). This framework reflects an emerging approach to health systems research in which the
focus has shifted from evaluating the efficacy and cost-efficiency of isolated biomedical
interventions to understanding how complex systems function to yield optimal health results. MPS
researchers readily acknowledge the incompleteness of the framework and its provisional
character. The framework has provided a fresh “way in” to the description and analysis of GHIhealth
systems interactions at country level, not a definitive answer. Nonetheless, the framework
has proved its value as a tool for shaping fruitful research questions and organizing results to
facilitate practical learning for policy and implementation.
3

Figure 1: Analytic framework for MPS research
Methodology
MPS researchers have mobilized a range of methodologies in conducting their investigations. Each
country case summary in the core section of this report includes a description of the specific
methods used. Here, we summarize some of the broad methodological issues that have arisen
across the MPS research effort as a whole.
Country selection
In selecting countries for inclusion as sites of MPS case studies, an initial sampling matrix was
developed, in consultation with partners, based on the following domains:
• Geographical spread
• Epidemiological profile (HIV, TB, malaria prevalence)
• Significant GHI investment as a percentage of GDP
• Perceived success or failure of GHI/health systems interaction
• Existing connection with partner institutions (given time constraints)
Given the particular prominence of HIV/AIDS as a global health challenge and a target of GHI
action, project leaders were concerned to include both high HIV prevalence countries and
countries facing concentrated epidemics in different regions.
4

To generate a list of country case study options, countries were filtered through the above criteria
and a final potential case study country pool was generated in consultation with WHO and civil
society. Partner institutions then selected case studies that corresponded to their current or future
research interests. In light of time constraints, project coordinators gave special consideration to
countries where research compatible with MPS goals was already underway at the time MPS was
formally launched. The academic consortium relied strongly on voluntary expressions of interest
from country-level researchers engaged in international networks and consortia.
Methods used by MPS research partners
The research previously undertaken and currently ongoing within MPS features multiple
components and distinct but related methodologies. The main MPS research products include:
• a systematic literature review
• mixed-method country case studies and facility-level studies
• cross-cutting comparative analyses of country studies
• multi-country studies and policy analysis
• cross-country quantitative analysis
• participant observation research led by civil society.
This document is a compilation of country case studies and facility-level studies. In many instances,
researchers are continuing to analyse their data and to refine their results as of this writing. The
summaries included in this report capture the state of findings and analysis as of mid-May 2009.
Ongoing analysis of the data obtained may yield additional insights..
Understanding complex systems: the strength of mixed methods
In recent years, new multidisciplinary approaches to health systems analysis have begun to
emerge, and innovative researchers have applied mixed-methods approaches to analyse the
health systems impacts of GHIs (see studies at www.ghinet.org).
MPS research builds on this pioneering work. It adopts the premise that multiple research
methods must be used in conjunction with each other to understand complex social systems. As
such, MPS contributes to the broader task of building a new field of multidisciplinary health
systems analysis. Many of the country-level case studies included in this report have adopted
mixed methods, collecting and analysing both quantitative and qualitative data, on the premise
that using quantitative and qualitative approaches in combination provides a better
understanding of research problems than either approach alone [12]. A multi-method approach
compensates for the weaknesses inherent in both quantitative and qualitative research and also
harnesses their complementary strengths [13, 14]. Mixed methods are best suited to build up a
nuanced picture of how complex entities like health systems and GHI programmes interact, and to
develop credible hypotheses about the factors that have favoured successful outcomes. In
particular, mixed methods enable investigators to capture non-quantifiable contextual influences
(for instance, historical and political factors) whose impact may elude more narrowly framed
5

studies. MPS researchers have linked quantitative and qualitative methods for systematic
triangulation of findings to strengthen results, provide richer evidence for conclusions, and build a
more complete knowledge base that can inform policy and practice.
Limitations of MPS research to date
Country researchers contributing to MPS have worked to an aggressive timeline to produce initial
results that can inform WHO normative guidance and policy debates within a critical window of
political opportunity. As a consequence, analysis of the rich qualitative and quantitative data
produced by MPS researchers is still ongoing. The findings presented in this report are preliminary.
A notable limitation of the accelerated initial phase of MPS research is that it has focused heavily
on PEPFAR and the Global Fund, with relatively little data collected on other GHIs. In subsequent
phases of investigation, this imbalance will be corrected. The challenging MPS timeline has also
meant that there has been limited opportunity to undertake cross-country analysis of findings—
though important initial steps in this direction have occurred. Systematic cross-country
comparative analysis will be a central component of the next phase of work on positive synergies.
The case study model used by MPS researchers has advantages but also limitations. In the research
by MPS collaborators, the case study approach has been constrained by factors impacting case
selection: access to countries, resources, and available time. Case studies are less suited to proving
causal pathways than to building and refining hypotheses about them based on informants’
experience and relevant documentation. In many countries, “longitudinal” case studies are
anticipated; however, the case summaries presented in this document generally provide a picture
of the situation at a particular point in time. They are most usefully construed as the initial phase of
a long-term research collaboration. In general, MPS research to date has fulfilled a critical
exploratory function. Providing rich descriptive data on GHI-health systems interactions in country
contexts, it has traced directions for a subsequent phase of research that will formulate and test
additional explanatory hypotheses to better understand the causal factors that enable or hinder
positive synergies.
6

References
[1] Brown TM, Cueto M, Fee E. The World Health Organization and the transition from
"international" to "global" public health. American Journal of Public Health, 2006, January;96(1):62-
72.
[2] Uplekar M, Raviglione MC. The "vertical-horizontal" debates: time for the pendulum to rest (in
peace)? Bulletin of the World Health Organization, 2007, May;85(5):413-4.
[3] Maciocco D. From Alma Ata to the Global Fund: The History of International Health Policy.
Social Medicine, 2008, 3(1):36-48
(http://www.nivel.nl/pdf/From%20Alma%20Ata%20to%20the%20Global%20Fund.pdf).
[4] Ooms G, Van Damme W, Baker BK, Zeitz P, Schrecker T. The “diagonal” approach to Global Fund
financing: a cure for the broader malaise of health systems? Global Health, 2008, 4:6.
[5] Atun R, Menabde N. Health systems and systems thinking. In: Coker R, Atun R, McKee M, editors.
Health systems and the challenge of communicable diseases: experiences from Europe and Latin
America. Maidenhead, Open University Press, 2008, p. 121-40.
[6] Atun R, Ohiri K, Adeyi O. Integration of Health Systems and Priority Health, Nutrition and
Population Interventions. Washington, DC, World Bank, 2008.
[7] Atun RA, McKee M, Drobniewski F, Coker R. Analysis of how the health systems context shapes
responses to the control of human immunodeficiency virus: case-studies from the Russian
Federation. Bulletin of the World Health Organization, 2005, Oct;83(10):730-8.
[8] Brugha R. Global Health Initiatives and Public Health Policy. In: Heggenhougen K, Quah S,
editors. International Encyclopedia of Public Health. San Diego, Academic Press, 2008, p. 72-81.
[9] Lawn JE, Rohde J, Rifkin S, Were M, Paul VK, Chopra M. Alma-Ata 30 years on: revolutionary,
relevant, and time to revitalise. Lancet, 2008, Sep 13;372(9642):917-27.
[10] World Health Organization. Report on the 3rd expert consultation on maximizing positive
synergies between health systems and Global Health Initiatives, WHO, Geneva, 2-3 October 2008.
Geneva, WHO, 2008.
[11] World Health Organization. Everybody’s business: strengthening health systems to improve
health outcomes: WHO’s framework for action. Geneva, WHO, 2007.
[12] Creswell JW. Qualitative inquiry and research design: choosing among five approaches. 2nd
ed. Thousand Oaks, CA, Sage Publications, 2007.
[13] Adamson J. Combined qualitative and quantitative designs. In: Bowling A, Ebrahim S, editors.
Handbook of health research methods: Investigation, measurement and analysis. Berkshire, Open
University Press, 2005, p. 230-45.
[14] O'Cathain A, Thomas K. Combining qualitative and quantitative methods. In: Pope C, Mays N,
editors. Qualitative Research in Health Care. Oxford, Blackwell, 2006, p. 102-11.
7

Burundi: Building a health system together with
Global Health Initiatives, in the aftermath of
war
Abstract
Johann Cailhol, 3, 4 Thuba Mathole, 3 Annie Parsons, 3 David Sanders, 3
Denise Kandondo, 2 Innocent Ndayiragije 2 and Théodore Niyongabo 2,5
Burundi’s health system was destroyed by civil war and has subsequently faced severe economic
constraints, creating an unreliable public health system. GHI activities in Burundi started in 2002,
initially relying on nongovernmental organizations (NGOs) and civil society organizations (CSOs).
This study used qualitative and quantitative methods to gather data on GHI activities in Burundi,
focusing on the health workforce.
The study found that GHIs and the Ministry of Health (MOH) have created parallel supervision,
monitoring and evaluation (M&E), and supply systems. This has led to distinct groups of patients,
health care providers and programmes managers, and a loss of efficiency. GHI funding is not
regularly channelled through the MOH: both the MOH and the Ministry of AIDS provide HIVrelated
care. This prevents the integration of HIV/AIDS into primary health services.
There is a high turnover of health care workers, lack of clear human resources policies and
insufficient salaries in public facilities. Both government and donors have attempted to address
such concerns, but joint efforts remain essential. In essence, improving the efficiency of GHI
funding requires health system strengthening. The pooling of donor funds for such initiatives is an
essential first step. Special attention should be given to health providers, as the core of the health
system.
Background
Burundi is a small land-locked country in central Africa, bordered to the north by Rwanda, to the
east and the south by the United Republic of Tanzania and to the west by the Democratic Republic
of Congo and Lake Tanganyika. It is home to 8.5 million people, 90% of them living in rural areas
[1]. The country averaged a 3.9% population growth rate from 2001 to 2007 [1] of which 46% is less
than 15 years of age [2]. Burundi’s inflation rate is high (23.5% in 2008) [1] and economic growth is
low (3.6% real GDP growth in 2007) [1]. It ranked 172 nd out of 179 countries on the UN Human
Development Index in 2006 [3]. After independence in 1962, the country experienced cyclical
crises attributed to ethnic tensions. The last civil war lasted 13 years and ended in 2005, though a
3 National Center for HIV Reference, Bujumbura, Burundi
4 University of Western Cape
5 University of Burundi, Bujumbura, Burundi
8

ebel organization operated until 2008 despite a cease-fire signed in 2006. Burundi is a fragile state
in a post-crisis situation, where ethnic issues remain problematic. The constitution was voted in by
the population in 2005 and includes a balance of ethnic groups in public employment.
Table 1 Basic Socioeconomic, Demographic and Health Indicators (*)
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 8,496 2007 World Bank
Geographic Size (sq. km) 25,680 2007 World Bank
GDP per capita, PPP (constant
2005 international $)
321.75 2007 World Bank
Gini index 33.27 2006 World Bank
Government expenditure on
health (% of general
government expenditure)
Per capita government
expenditure on health at
average exchange rate (current
US$)
5.3 2007 WHO NHA
2 2007 WHO NHA
Physician density (per 10,000)

at between US$ 14.5 and US$ 18.5 per capita [4]. Official external aid to the health sector was US$
60 million, of which US$ 20 million went to the HIV/AIDS sector and US$ 40 million to other sectors
[4]. For the same year, total government expenditure for health was US$ 8 million [4]. Use of
heavily indebted poor countries (HIPC) funds allowed national health expenditures for health to
rise from less than US$ 5 million in 2005 to 33 million in 2009 [5]. The current public health system
features free health care for women in labour and children under 5 years since May 2006, as part of
the government’s national objective of reducing maternal and child mortality.
Since 2003, Burundi has been approved for US$ 87 million in funding from the Global Fund for
HIV/AIDS, TB and malaria efforts, of which US$ 62 million has been disbursed. Burundi’s application
to round 8 of the Global Fund for HIV-related efforts has also just been approved [6]. Health care in
the public sector for HIV-infected patients receiving antiretrovirals (ARVs) is provided free of
charge, but this is not the case for people living with HIV/AIDS (PLWA) who do not receive ARVs.
One CSO in Burundi currently provides performance-based financing (PBF) for health care
structures to deliver free care to patients not on ARVs. Anti-tuberculosis drugs and TB-related
activities are currently provided free of charge in public health facilities. Malaria treatment is free
as of May 2009, but the diagnosis of malaria is not.
The Health System in Burundi
Burundi is in a transition stage between emergency aid and development aid, after a 13-year civil
war that ended in 2005. Its institutional capacity in general and in the health sector in particular is
very weak; the civil conflict destroyed most of the health-care infrastructure, resulting in high rates
of human resources brain drain and the deterioration of health systems. The conflict had a similar
impact on the economic situation of the country such that by 2007, the gross domestic product
(GDP) was only $100 per capita [7].
Until 2005, basic health needs in Burundi were supplied by NGOs specializing in humanitarian aid.
These NGOs are being progressively replaced by other NGOs with expertise in long-term
development, which increasingly receive GHI funding and direct assistance. The traditionally
predominant role of NGOs in the country, together with a lack of leadership by government, has
resulted in a number of uncoordinated activities in the health sector.
The main sources of funding for Burundi’s health system are now GHIs such as the Global Fund,
GAVI, MAP, PEPFAR and other donors. By 2006, 76% of total health expenditure was externally
financed [7]. The Global Fund contributes the largest amount, at 37% of the budget, and the rest is
from other donors [8]. PEPFAR supports some international NGOs providing HIV services in
Burundi.
The Ministry of Health (MOH) in Burundi is organized on four levels: a national level, divided into
programmes and cross-sectoral administration, a provincial level, a district level (currently being
developed) and a facility level through the primary health care (PHC) centres. Activities including
TB, malaria and immunization, are conducted as part of an integrated minimum services package
at facility level and coordinated by provincial or district offices. Since 2002, most of the public
facilities, from PHC centres to hospitals, are financially independent.
In 2004, under pressure from UNAIDS and the WHO, a separate Ministry for AIDS (MoA) was
launched. The MoA was directly linked to the presidency, to ensure political visibility in the fight
against AIDS and to direct multi-sectoral interventions. The MoA was transformed in 2007 into a
vice-ministry of the MOH in an attempt to integrate activities into the MOH, though a new
separation from the MOH occurred in January 2009 for political reasons.
10

The National AIDS Council (NAC) was created in 2002 following World Bank guidelines, to manage
HIV-related funds and coordinate multi-sectoral HIV activities. As a result, each national ministry
has an HIV unit that is in charge of HIV-related activities.
Objectives and Methodology
This study aimed to assess the effects of GHIs on the health system in a recent post-conflict
country, with a special focus on the human resources for health (HRH) landscape. Mixed methods
of research were used, with the data presented in this paper reflecting mainly the results of
national level research and some preliminary results from provincial level research. At national
level, 26 participants were interviewed, including senior officers in four key ministries (health,
AIDS, finance, civil service), NAC, GHI coordination, World Bank, WHO, UNAIDS, and NGO, bilateral
or multilateral donors. Documents and reports provided additional sources of data. Further work
was conducted at provincial and facility levels, in two provinces, one rural (Kirundo, one of the
poorest provinces among those severely affected by HIV) and one urban (Bujumbura-urban). Data
collection is currently ongoing in one additional province. Within each province, four facilities
were included: one public hospital, one GHIs-funded NGO, one public and one private primary
health care centre. Facility-level work consisted of a qualitative component of 21 semi-structured
interviews with key informants, focus group discussions with health care workers and
observations, as well as quantitative analysis of closed questionnaires and recorded data. Initial
research findings were presented at a workshop for key stakeholders in Bujumbura in May 2009.
This paper will focus on the health workforce. This appears to be one of the major bottlenecks of
the health system in Burundi. We used national-level interviews and some inputs from provincial
and facility-level data. Remaining themes and data will be further developed and presented in the
final report.
Results
Human Resources
Burundi’s political instability had significant negative impacts on HRH. No regular national census
on total HRH numbers exists, with figures only reflecting the public sector. Government employees
at all levels often also hold positions in the private sector. The number of HRH hired by NGOs,
donors, or in the private sector, has never been assessed. In 2008, the number of health workers in
Burundi’s public sector was insufficient by international norms. An average of 0.2 physicians
served every 10 000 inhabitants, with 1.4 nurses for every 3000 inhabitants, and 68 pharmacists
throughout the entire country. 80% of general physicians and 50% of nurses work in urban areas
although 91% of the population lives in rural areas [3]. The capital has a slightly higher rate of
physicians, with 1.4 per 10 000 inhabitants. A study conducted by the International Organization
for Migration (IOM) revealed that 150 physicians left the country between 1993 and 2002, because
of conflicts, low salary levels and poor working conditions [3]. The distribution of nurses varies
according to the level of ongoing insecurity and the unequal level of external donor support to
any given province.
The training of new health workers cannot replace those leaving the public sector. Staff
development is an area of concern in Burundi as it has only one public school of medicine with an
annual rate of graduation of around 50 physicians. Trainings in specializations are offered through
11

collaborations with European schools of medicine, but few trained specialists return to work in the
country due to poor working conditions. The government is aware of the insufficient number of
nurses and has decided to increase the number of students in nursing schools. Eight hundred
clinical staff (nurses, laboratory technicians and nurse assistants) graduate each year from six
nurses’ schools, of which two are private, and one unique laboratory technicians’ school. However,
75% of the nurses graduate with the lowest degree (of two years training after junior high school).
Pharmacists are exclusively trained in foreign countries.
The annual number of newly qualified clinical staff available for the public sector is routinely
depleted by trainings abroad and staff preference for the private or donor-funded sectors.
Physicians in Burundi are also often expected to fulfil management roles, to the detriment of
clinical work. The Global Fund has begun to contribute to the general academic training of
physicians, as well as funding a post-graduate degree in HIV care. It plans to expand these
contributions to the public nurses’ schools. To cope with an insufficient number of HRH, especially
in PHC centres in rural areas, support workers (with only primary school qualification and
additional health training) are recruited and assigned a role in immunization, family planning, or
antenatal consultations.
GHIs have in general contributed to multidisciplinary HRH trainings, but in sectors specific to GHIs’
focal diseases [9-11]. Global Fund malaria-funded activities, for example, include the training of
support workers in primary health care centres, so as to increase the number of health workers
able to diagnose malaria. However, the GAVI Health System Strengthening (HSS) programme has
expanded its trainings beyond its focus on health conditions to health system strengthening and
other areas such as managing obstetrical surgery emergencies. The majority of trainings focus on
technical areas of concern, and there are a few more general trainings on non-clinical topics such
as financial or human resources (HR) management.
The high number of trainings and the small number of health care providers in Burundi suggest
that every health care provider may have received numerous trainings, but this information is not
recorded. Lists or databases of those who receive training are not systematically kept by facilities
or training organizations, the MOH or the MoA. This is compounded by the high turnover rate of
HRH in the public sector. Added to the absence of a database on trained HRH, this can lead to one
health worker being trained several times.
Programme performance indicators rely on quantitative evaluations; qualitative evaluations of the
trainings are non-existent. Neither the content of each module is assessed, nor an evaluation
conducted of the quality of the service provided post training. Furthermore, the organization of
training sessions is not coordinated. Planning occurs only at national level and it is on an ad hoc
basis, with little response to identified needs at district or facility level. This is partly due to the
financial incentive of arranging a training session, as national level organizers also earn per-diems.
Financing
All public sector human resources in Burundi are coordinated by the Ministry of Civil Service. Salary
levels are decided by the Ministry of Civil Service and Ministry of Finance, and based on the highest
degree obtained as well as years of service. However, remuneration differs considerably between
sectors: for example, some civil servants in the financial sector received significant increases in
salaries and allowances in 2006 [12].
Overall, HIV-related services in the public sector that are not part of the MOH receive the most
funding for HRH through salaries, per diems and incentives. The MOH itself as well as other key
12

areas are poorly funded. GHIs, excluding GAVI, committed an average of US$ 20 million per year to
Burundi over the period 2002 to 2007, whereas government expenditure for the same period was
an average of US$ 8 million per year. More than 50% of GHI funds were earmarked for HIV-related
activities. HIV-related funds are channelled through the NAC and not through the MOH. NAC
hence contracts directly with public and private health facilities to implement activities, by-passing
the MOH. HIV is thus considered as distinct from the health sector by the general population,
patients, care-providers, programme managers and by national MOH staff.
Rural facilities rely heavily on nominally trained support workers to deliver services. These workers
are paid from each individual facility’s surplus, at a monthly salary of approximately US$ 4 to US$
24 per month and are not recognized by the MOH as civil servants. Annual salaries in the public
sector (including allowances but before taxation) range from US$ 62 to a maximum of US$ 152 for
a specialist physician. Living costs for a family of five in the capital Bujumbura are approximately
US$ 240 a month for flat rental and food. This excludes school fees, transportation, health care
costs, etc. As a result, most public sector health workers are concurrently employed in the public
sector or by NGOs. There has been little recent improvement in the general working conditions of
the health workforce. A national strike for living wages by public sector nurses and physicians has
been ongoing since December 2008.
Short-course trainings sponsored by GHIs are considered ‘disguised salaries’ and are often actively
organized to help public sector HRH financially survive by compensating for their extremely low
salaries. Participants in short-course trainings receive financial compensation in the general range
of US$ 3 to US$ 20 per diem.
This per-diem system is the result of insufficient salaries in public sector and generalized poverty
after the war. The pressures of poverty make it difficult to objectively choose who participates in
such ‘remunerated’ training. Inequities between per diems according to the funding source also
induced competition between staff on who attended which training sessions. At points
participants have gone on strike to demand higher per diems.
MAP1 has contributed to the salaries of NAC staff at national and provincial levels, to ensure the
operation of NAC, but this does not apply to facility health workers. Global Fund funds for HIV have
also contributed to the hiring of health mediators, a new HRH category without any clinical
degree, as well as that of other non-medical HRH in both the public and private sectors. These
workers are paid higher salaries than nurses in the public sector. Global Fund HIV-related funds are
also used to hire HRH for local NGOs, the effect of an agreement with NAC.
MAP funding for salary top-ups is in general restricted to NAC at national and provincial level. As
the result of fears around sustainability, staff in MOH facilities cannot receive such support. The
Global Fund does not provide incentives at facility level for TB or malaria-related activities, as these
conditions are considered part of the MOH’s integrated service delivery. However, the Global Fund
does allow the ‘topping-up’ of salaries for some national and provincial level TB-related MOH staff
and at national level for MOH staff working on malaria. As a result, a ten-fold difference in salary
can exist between people sharing the same responsibilities within the national HIV programme
and the national TB programme, depending on the employer (MAP or NAC or MOH). GAVI,
through its HSS project with the MOH, does provide incentives to its management unit at national
level and to ambulance drivers hired as part of the referral system in four provinces. Performancebased
financing (PBF) is planned for implementation in the near future at service delivery level.
Public structures have an agreement with NAC and the national HIV programme that each facility
staff receives a sum equal to 10% of that facility’s monthly expenditure, to compensate for any
profit lost as the result of not charging for antiretroviral therapy (ART)-related services. This ‘10%’
initiative is specific to Burundi. However, for instance in a general hospital, the benefit would
13

epresent only US$ 1.75 per month per worker if equally distributed. Salaries at HIV-focused NGOs
in Bujumbura are thus approximately four-fold higher than for staff with similar degrees in the
public sector and, moreover, are exempted from revenue taxes. The increase of activities at facilitylevel
funded by GHIs has also led to a parallel increase in workloads for public sector employees,
but with no related salary increases. 70% of hospitalized patients in the public sector are HIVpositive.
As a result, health workers would rather work for GHIs than in the public sector.
Inequities in salaries within the public sector have also impacted on supervision. A manager may
earn less than his employee or have to supervise a team of people with the same degrees and
identical tasks, but who have different contracts and hence earn different salaries. Discrimination
can also occur within a team.
Monitoring and Evaluation
Burundi’s national health information system (HIS) is not considered reliable, as information is
often late or incomplete. As a result, most programmes bypass it in reporting to donors and use
donors’ own reporting and data collection systems. The Global Fund’s separate budget for the
training, allocation and implementation of HIV and malaria monitoring and evaluation represents
more than US$ 8 million [13, 14]. This has led to situations in which one facility with many
programmes will have numerous reporting mechanisms and requirements, often duplicating each
other. The lack of common indicators hence increases staff workloads.
At facility level it was also found that the number of required reporting activities also prevented
staff supervision. Furthermore, supervision is not coordinated: district or provincial health offices,
for instance, will collect data independently of provincial HIV committees. Staff working more than
one job or overburdened due to insufficient staff numbers were also less likely to complete routine
tasks such as data entry, supervision and evaluation.
Table 2 Global Health Initiative Investments (*)
(*) Full data sources for all indicators are provided in Annex 1.
Global Fund
Round 1 HIV/AIDS 8,657,000 8,657,000
Round 2 Malaria 39,089,883 29,072,765
Round 4 TB 3,381,665 2,554,489
Round 5 HIV/AIDS 32,353,173 20,355,559
Round 7 TB 4,018,177 1,745,914
Round 8 HIV/AIDS 36,789,591 0
TOTAL: 124,289,489 62,385,727
14

PEPFAR*
Year Amount Allocated (in US$)
2006 2,117,000
2007 2,680,000
2008 4,031,650
TOTAL: 8,828,650
*Not a PEPFAR focus country; above sums represent total allocations to PEPFAR
country programmes from bilateral U.S. sources including USAID, Department of
Health and Human Services, Department of Labor, and Department of Defense.
GAVI
Disease Priority Amount Approved (in US$)
Hepatitis B vaccine: 466,000
Pentavalent vaccine 37,461,000
Tetravalent vaccine 4,574,000
Vaccine introduction grant 100,000
Injection Safety 420,000
Immunization services support 2,658,500
Health systems strengthening 8,252,000
TOTAL: 53,930,521
World Bank MAP
Project Title FY Approved /
Closing Date
Commitment
(in US$)
Multisectoral HIV/AIDS Control and Orphans Project 2002/2008 36,000,000
Burundi Second Multisectoral HIV/AIDS Project 2008/2011 15,000,000
TOTAL: 51,000,000
Medical Products, Vaccines and Technologies
In contrast to other countries, in Burundi it does not appear that GHIs have strengthened the
supply system. Maintenance of equipment purchased by GHIs is not effective, with frequent
breakdowns. Frequent HIV-related pharmaceutical shortages occur (namely ARVs and
opportunistic infection medications), as well as for reagents (HIV-testing, CD4 count, viral load)
purchased by NAC with MAP1 and Global Fund funds. For instance, in 2006 cotrimoxazole was not
available for almost a year and HIV-testing reagents have not been available since October 2008.
Local stakeholders attribute these shortages to insufficient reporting at facility level and a complex
procurement and administrative system. The supply system in Burundi is very complex with 28
different sources of funding and 26 different supply mechanisms [15].
15

Service Delivery
The decentralization of GHI-related care, particularly HIV-related services, has been clearly
hampered by the health system’s weakness in rural areas. Insufficient HRH numbers, and the
presidential decree of 2006 that children under five and women in labour receive free health care
access has led to clear cases of overwork in primary health care centres [16]. (General antenatal
services are not free.) The implementation of new activities, such as HIV-related services, has been
subsequently challenging. Prevention of mother to child transmission (PMTCT) coverage amongst
HIV-infected pregnant women was only 7.8% in 2008 and less than 20% of facilities offered PMTCT
services. Health workers do not consider existing GHI-related activities, like immunization, malaria
and TB, as time-consuming as HIV-related activities. The lack of coordination means that any
problems are exacerbated, as the response will often not reflect need.
Leadership and Governance
A National Committee for Aid Coordination (NCAC) was created in 2005 to coordinate all external
aid to Burundi. The NCAC supervises 13 sector groupings, of which one is concerned with HIV and
another with health, reflecting the parallel system of addressing HIV and health at ministry level.
The two most efficient groupings are the health and education groups, with recent efforts in the
health group concentrated on improving partnerships between government and technical and
financial partners.
Since 2007, Burundi has also been involved in piloting the International Health Partnership
Initiative (IHP+). However, the disbursement of IHP+ funds is very low (eight percent of US$
800,000 over two years) due to the heavy demands of administrative procedures and lack of
ownership by the government [17]. Political decisions, such as the removal of senior national MOH
officers to meet constitutionally mandated ethnic requirements, have at times disrupted the ability
of the MOH and other bodies to meet targets.
GHI funding has given CSOs and NGOs the capacity to quickly and effectively implement activities.
The post-crisis nature of Burundi has meant that this has often involved bypassing official state
structures. However, CSOs and NGOs cannot substitute for national systems and have now begun
to reach their limits. Efforts to ensure that government and non-government services work
together have recently increased, but are often rejected by the public sector due to salary
differences.
Discussion
GHIs in Burundi have improved access to specific disease-focused drugs and examinations by
providing free access and increasing availability, though shortages and maintenance issues exist.
The selective building of capacity through NGOs, CSOs and disease-specific programmes in the
public sector has increased the distortion of public health sector service provision. Service
fragmentation, such as the management of TB and HIV by two different programmes, has meant
that clinical staff struggle to effectively treat patients. The lack of coordination between ministries
as well as in supervision, training, or M&E has led to wasted funds. GHIs efficiency has also been
affected by instability and low remuneration of HRH.
16

GHI funding has had positive impacts on the financing of HRH, with selected HR receiving
incentives and decent salaries, in turn providing for capacity building. There has been a selective
increase in some structures’ HRH numbers, such as for HIV-related programmes or NGOs. However,
this has contributed to the distortion of an already weakened system by fuelling brain drain from
the public sector. For instance, between 2006 and 2008 a local GHI-funded NGO increased its
clinical staff from 5 to 14, whereas in a rural public hospital over the same period this number
decreased from 26 to 23.
However, government and donors are aware of the issues arising from weak institutional capacity.
There are cooperative efforts to strengthen the health system and to overcome the key bottleneck
of workforce financing and capacity. Though GHIs entered Burundi to address emergency and
conflict-related health issues, there is growing recognition that effective health care requires
building the government’s capacity to address long-term primary health care needs. The
government of Burundi continues to struggle with the ramifications of the long civil war, and basic
issues of capacity building often hamper its work.
Burundi is taking part in new initiatives such as the GAVI HSS programme, developed at a global
level to address issues of health system strengthening. The MOH created a HR division in 2006 and
is moving towards an HR planning policy and reform of clinical training. Performance-based
financing and the harmonization of indicators are ongoing processes. In an effort to increase
ownership, part of MAP’s second round of funding will be channelled from NAC to the MOH.
The study has resulted in certain recommendations that could assist both government and GHIs in
strengthening Burundi’s health systems.
Both short-term and long-term strategies are available to GHIs and the government of Burundi.
Short-term strategies could improve the ability of government and GHIs to provide
comprehensive disease-focused programmes, by ensuring that better resourced facilities are not
as reliant on the channelling of patients from poorer resourced health centres.
Long-term health system strengthening will require joint efforts by GHIs and government in the
building of basic capacity at national, provincial and facility levels. The two key areas of HRH and
health information systems are fundamental to these processes.
Conclusions
HSS should be prioritized by all GHIs in Burundi, with funding pooled as part of a common HSS
fund that is in turn used to support increases in all public sector HRH salaries, ensure coordinated
supervision, build one effective national supply and maintenance system and in strengthening the
national HIS. Such a common fund would support HRH trainings on two different levels, by
reinforcing the supply and quality of national academic HRH degrees and creating an in-service
training structure. Funds saved by each GHI through such pooling processes would then be used
to ensure equity and quality of service delivery. For instance, malaria diagnosis could be provided
free-of-charge and all patients diagnosed with HIV-infection could receive free care. Innovative
health coverage for the whole population could be developed. The need for GHIs to be flexible
and adapt themselves to each country is exacerbated in Burundi, where a weakened health sector
received a sudden and significant amount of funds, without HSS planning or strong technical
support from donors. Technical assistance, as well as strengthened evaluation by GHI or donorfunded
programmes, would assist local stakeholders in increasing efficiency.
17

Long term HSS will require government’s commitment to increasing HRH salaries in the public
sector, through joint efforts with GHIs and other donors or the implementation of innovative
financial policies. Such a salary increase would only be effective when supported by the
implementation of a strong HRH policy that ensures increased conscientiousness and leads to
improved management and evaluation. Health services management should also be devolved to
trained non-clinical staff, instead of expecting trained physicians to act as managers.
An overhaul of the present structuring and coordination of trainings is also urgently required, as
both a short-term and long-term measure, in order to reduce inequity in who is trained and to cut
dependency on trainings as salary supplementation. The organization of trainings should reflect
health system needs (for example, focusing on more general issues such as financial or HR
management instead of purely technical concerns). The systematic collection of such data through
consultation with all levels of leadership will also help increase the accountability of government’s
structures to its population.
However, government will struggle to implement such initiatives if political uncertainty at national
level continues to lead to unpredictable changes of management level staff. The duplication of
activities by the MoA and the MOH, as well as the channelling of funds through the NAC, stifle
clear leadership and ownership. Sustainable and equitable health care financing is also necessary,
with a short-term focus needed on increasing government health expenditure to at least 15% of all
expenditure.
The immediate concern of both GHIs and government should be HRH, as each part of the health
system relies on the availability and retention of trained health workers. As one participant noted,
“human resources are the real bottleneck, if they are not supported, good results cannot be
achieved” [18].
18

References
[1] The World Bank Group. Burundi Data-at-a-Glance, 2008
(http://devdata.worldbank.org/AAG/bdi_aag.pdf; accessed 19 February 2009).
[2]World Health Organization. WHO Statistical Information System (WHOSIS).
[3] United Nations Development Programme. Human Development Reports: Burundi
(http://hdrstats.undp.org/countries/data_sheets/cty_ds_BDI.html; accessed 19 February 2009).
[4] Rapport sur le financement du secteur de la santé. Ministère de la santé et la Banque Mondiale.
Burundi, Novembre 2007.
[5] Mission d’études des coûts et du financement du système de santé au Burundi. Commission
Européenne. Burundi, February 2009.
[6] The Global Fund. Burundi and the Global Fund.
(http://www.theglobalfund.org/programs/portfolio/?countryID=BRN; accessed 6 March 2009).
[7] Technical report on financing mechanisms of health development plan: Burundi. WHO, 2005.
[8] Partners in impact: Results report. Global Fund to Fight AIDS, Tuberculosis and Malaria, 2007.
[9] GF-HIV grant performance report, grant number BRN-506-G04-H. GFATM, October 2008.
[10] GF-TB grant performance report, grant number BRN-708-G06-T. GFATM, March 2009.
[11] GF-malaria grant performance report, grant number BRN-202-G02-M-00. GFATM, August 2007.
[12] Les disparités des statuts pécuniaires dans l’administration publique au 31 décembre 2007.
Cour des comptes, Burundi, June 2008.
[13] Plan stratégique de lutte contre la malaria 2008-2012. Ministère de la santé, OMS. Burundi, July
2007.
[14] Plan stratégique de lutte contre le VIH/SIDA. Présidence, Ministère de la lutte contre le
VIH/SIDA. CNLS, 2007-2011. Burundi, December 2006.
[15] Circuits d’approvisionnements des produits pharmaceutiques au Burundi. Ministère de la
santé, OMS. Burundi, July 2007.
[16] Rapport des effets de la mesure de subvention pour les enfants de moins de 5 ans et pour les
accouchements sur les structures et la qualité des soins. Observatoire de l’action
gouvernementale. Burundi, February 2009.
[17] The Inter-Regional Country Health Sector Teams meeting, Lusaka, Zambia, February 28 -
March 1, 2008. IHP+, Taking Stock Report: Burundi, 2008.
[18]National-level interview number 16, MPS Burundi study. 2009.
19

Cameroon: Evaluation of the National Programme for
Access to Antiretroviral Therapy
Abstract
Boyer S. 1,2 , Eboko F. 1 , Camara M. 3 , Abé C. 4 , Owona Nguini M.E. 5 , Koulla-Shiro S. 6,7 , Moatti J-P. 1
An evaluation of the national antiretroviral treatment (ART) programme in Cameroon gave us the
opportunity to assess its impact on quality of care, equity, and unsafe sex.
The cross-sectional survey (“EVAL”- ANRS 12 116) was conducted in 2006-2007 among a sample of
3151 HIV-positive adults attending 27 HIV-services at the three levels of health care
decentralization (central/provincial/district). Multivariate two-level analyses were conducted to
assess the impact of HIV-care decentralisation on quality of care, equity and unsafe sex.
The main results showed that quality of care in district HIV services was as good as in central and
provincial HIV services and even better for some outcomes, such as ART adherence and mental
health related quality of life (HRQL). Some structural factors limiting quality of care have also been
identified such as the lack of qualified human resources and difficulties with the supply of
antiretroviral drugs (ARVs). Prevailing inequities in access to care were found to be linked both to
socio-economic and structural factors. Regarding prevention behaviours, access to ART was
associated with a lower risk of unsafe sex.
Our results confirmed the feasibility of HIV care decentralization. However, long term sustainability
urgently requires better integration of this HIV-targeted programme into comprehensive health
care reform of financing mechanisms, human resources management and drug procurement
systems.
1 INSERM/IRD/Aix-Marseille Universities Research Unit 912 (Economic & Social Sciences, Health Systems & Societies)
2 Centre for Disease Control of South-Eastern France (ORS-PACA)
3 Centre for Economics (CEPN), University of Paris 13
4 Socio-anthropological Research Institute (IRSA), Catholic University of Central African States, Yaoundé
5 Paul Ango Ela Fondation for Promotion of Geopolitics in Central Africa (FPAE) & University of Yaoundé 2
6 Ministry of Public Health – Division of Health Operations Research, Yaoundé
7 Faculty of Medicine & Biomedical Sciences, University of Yaoundé 1
This study was supported by the French National Agency for AIDS Research (ANRS).
20

Background
Cameroon is a central African country bordered by the Federal Republic of Nigeria to the west; the
Republic of Chad to the northeast; the Central African Republic to the east; and the Republic of
Equatorial Guinea, the Gabonese Republic, and the Republic of the Congo to the south. The
country is home to 18.5 million people and averaged a 2.2% annual population growth between
2001 and 2007 [1]. With a gross domestic product (GDP) per capita of 2005 (PPP constant 2005
international dollars) [2], Cameroon ranked 150 th out of 179 countries on the UN Human
Development Index in 2006 [3].
The first AIDS case in Cameroon was diagnosed in 1985 [4]. Today, the country suffers from a
generalized HIV epidemic characterized by high disparities between genders, provinces and rural
versus urban areas: the HIV prevalence rate reaches 5.5% in the adult population (15 to 49 years of
age) and between 10.7% and 11.9% among adult women in the most affected areas, which include
the provinces of the North-West, South-West and East and the capital city of Yaoundé [5]. By the
end of 2007, the number of people living with HIV/AIDS (PLWHA) was estimated to be 540 000 [6].
In 2006, Cameroon had an infant mortality rate of 87 per 1000 live births, and a maternal mortality
rate of 1000 per 100 000 live births [7]. WHO estimated that incidence of smear-positive pulmonary
TB in Cameroon was 83 per 100,000. Of all new cases, 15% occurred in HIV-positive individuals, and
1.7% were multi-drug resistant (MDR-TB) [8].
Administratively, Cameroon is divided into ten semi-autonomous provinces (“regions”), each
headed by a presidentially-appointed governor. These provinces are then further sub-divided into
divisions, sub-divisions, and finally districts. The health system in Cameroon is organized at three
levels: a central level including national administrative units in the Ministry of Health, as well as
referral hospitals, responsible for setting policy; an intermediary level, including regional
delegations and regional programme coordinators, in charge of technical support to district-level
programmes; and a peripheral level, composed of health areas grouped into health districts, which
is considered as the operational level for the provision of primary health care services.
Out-of-pocket expenses made up more than two-thirds of total health expenditure and external
resources 7.1% in 2006 [7]. Government health expenditures represented 1.46% of GDP in 2005 [2].
With an average of 2 physicians and 16 nurses per 10,000 inhabitants, the country has been
classified as by the World Health Organization as one of the 57 countries in the world having a
critical crisis in its health workforce [9]. Moreover, human resources for health are largely
concentrated in the urban areas: in 2004, 69.94% of the country’s physicians worked in cities, but
only 53.42% of the population lived in cities [10].
Official development assistance to Cameroon was $1.68 billion in 2006 [2]. Debt relief under the
Heavily Indebted Poor Countries initiative in 2006 brought Cameroon’s total external debt down
from US$ 7.2 billion in 2005 to US$ 3.2 billion in 2006 (or 17.7% of GDP) [1,2].
21

Table 1 Basic Socioeconomic, Demographic and Health Indicators (*)
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 18,533 2007 World Bank
Geographic Size (sq. km) 465,400 2007 World Bank
GDP per capita, PPP (constant 2005
international $)
2,005.49 2007 World Bank
Gini index 44.56 2001 World Bank
Government expenditure on health (%
general government expenditure)
Per capita government expenditure on
health at average exchange rate (current
US$)
6.6 2007 WHO NHA
12
2007 WHO NHA
Physician Density (per 10,000) 2 2004 WHOSIS
Nursing and midwifery density (per
10,000)
Maternal mortality ratio (per 100,000 live
births)
16 2004 WHOSIS
1,000 2005 WHOSIS
DPT3 coverage (%) 82 2007 WHOSIS
Estimated adult HIV (15-49) prevalence
(%)
Estimated antiretroviral therapy coverage
(%)
5.1 (3.9-6.2) 2007 UNAIDS
25 (21-32) 2007 WHO/UNAIDS/UNICEF
Tuberculosis prevalence (per 100,000) 195 2007 WHO GTD
Estimated malaria deaths, all ages 21,146 2006 WHO WMR
Methodology
In 2001, Cameroon initiated one of the largest programmes for access to ART in Western and
Central Africa. It proceeded in two stages, beginning with the decentralization of HIV care from the
central level, in the economic and administrative capital cities of Yaoundé and Douala, to an
intermediate level in provincial hospitals (2001-2003), which was then followed by
decentralization to district hospitals, starting in 2005.
As of June 2008, ART delivery was based at 24 accredited treatment centres (ATCs) located in the
main hospitals of Douala (Littoral province) and Yaoundé (Centre province) and in each capital of
the eight other provinces. These ten provinces serve as mentors and reference centres for 108 HIV
management units (MUs) at district level. Overall, ART delivery facilities are available in 106 out of
the 174 districts. A simplified approach for ART management at district level has been developed
22

ased on WHO recommendations for scaling-up ART in resource-limited settings [11, 12]. This
approach allows ART initiation even when CD4 count is not available, using total lymphocytes
count and clinical stage of the disease.
Public subsidies were also devoted to ARV drugs, leading to a progressive reduction in out-ofpocket
payments by HIV-infected patients at the point of delivery, until the achievement of free
access to ART for all eligible patients in May 2007. Foreign aid, in particular from the Global Fund
contributes substantially to the financing of the national programme for access to ART.
In June 2008, 53 238 persons living with HIV/AIDS (PLWHA) – 58% of the estimated number of the
Cameroonians requiring ART according to the National AIDS Control Committee (See Table 1) –
were receiving treatment. This figure constitutes a significant scale-up from the 600 Cameroonians
on ART in 2001, and gives the country one of the highest rates of ART coverage in the world.
In 2006, the Ministry of Public Health of Cameroon requested an evaluation of the national ART
programme.
Researchers from the Universities of Yaoundé and the French Agency for AIDS Research (ANRS)
jointly carried out the evaluation which aimed to assess the programme’s performance according
to the level of HIV-care delivery in the following main areas: (i) early entry to care for HIV patients
and quality of care for ART treated patients, (ii) financial accessibility of HIV-services and equity in
access to treatment, (iii) prevention behaviours.
The evaluation was based on cross-validation of quantitative and qualitative surveys aimed at
collecting original data from patients, health care professionals and public health decision makers.
These data were compared with secondary analysis of existing sources (reports from the Ministry
of Public Health, National AIDS Control Committee (NACC), National Centre for the Supply of
Essential Drugs and Medical Consumables (CENAME), UN and bilateral organizations, the Global
Fund etc.).
A national, cross-sectional survey was conducted in Cameroon from September 2006 to March
2007. Participants were recruited in 27 hospitals delivering HIV care (eight “central” Accredited
Treatment Centres (ATCs) of the two main cities, six ATCs in the provincial capitals and 13 district
managing units (MUs) located in six provinces around the country. Those eligible to participate in
the survey were aged 21 years or older and had been diagnosed as HIV-positive for at least three
months. Participation was proposed to eligible patients who came for a consultation during the
survey period, according to a random selection procedure. At the time of the survey, ARV drugs
were not free; they were delivered to patients at the price of 3000 CFA (Communaute Financiere
Africaine) Francs, or about 6 United States dollars (US$), for Triomune, and 7000 CFA francs, or
about US$ 14, for other treatments (at US$ 1 = 492.6 FCFA).
Among the 3488 patients approached, 3170 (91%) agreed to participate in the survey, and 99% of
those patients filled out the questionnaire, leading to a total sample of 3151 respondents (global
response rate = 90%).
Quality of care according to the level of health care delivery was assessed in the subsample of
patients who had been ART-treated for at least six months by using four main criteria: (1) average
monthly gain in CD4 cells/mm 3 since initiation of treatment; (2) adherence to ART 6 ; (3) occurrence
6 Adherence to ART was measured using a validated list of questions concerning dose taking during the previous four
days, as well as with respect to the time schedule and occurrence of treatment interruptions during the previous four
weeks. These data enabled us to compute two scores for adherence, concerning the previous four days and the previous
four weeks, often used in previous cohort studies.
23

of treatment interruptions (>two days); and (4) physical and mental health-related quality of life
(HRQL) 7 .
Financial accessibility of HIV services was assessed in the population of ART-treated patients and
non-ART-treated patients using the concept of catastrophic health-related expenditures, which,
according to the WHO definition, could apply to households spending more than 20% of their
income in health care expenditures.
Finally, prevention behaviours (or “safe” sex) were assessed in the population having sexual
intercourse during the three months prior to the survey with a main partner of unknown HIV status
or seronegative status. “Safe” sexual behaviours were defined as the systematic use of condoms
with the main partner.
Table 2 Global Health Initiative Investments (*)
(*) Full information on data sources is provided in Annex 1.
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 3, HIV/AIDS 55,500,617 44,743,262
Round 3, Malaria 31,781,187 29,881,464
Round 3, TB 5,804,961 4,575,789
Round 4, HIV/AIDS 16,194,089 15,915,458
Round 5, HIV/AIDS 12,070, 127 3,662,574
Round 5, Malaria 14,310,624 4,248,021
TOTAL: 135,661,605 103,026,568
PEPFAR*
Year Amount Allocated (in US$)
2007 660,000
2008 2,017,677
TOTAL: 2,677,677
*Not a PEPFAR focus country; above sums represent total allocations to PEPFAR
country programmes from bilateral U.S. sources including USAID, Department of
Health and Human Services, Department of Labor, and Department of Defense.
7
Patients’ HRQL was assessed using the Medical Outcome Study Short-Form General Health Survey (MOS SF-12), which
has been validated in the specific context of HIV infection.
24

GAVI
Disease Priority Amount Approved (in US$)
Pentavalent vaccine 28,217,000
Tetravalent vaccine 9,776,000
Yellow fever vaccine 6,060,000
Vaccine introduction grant 358,500
Injection Safety 1,029,300
Immunization services support 9,407,120
Health systems strengthening 9,846,000
TOTAL: 64,692,469
World Bank MAP
Project Title FY Approved /
Closing Date
Commitment
(in US$)
Multisectoral HIV/AIDS Project 2001/2007 50,000,000
TOTAL: 50,000,000
Results
Quality of care according to the level of health care delivery and early entry to care
A relatively good quality of care in district HIV services
Mean CD4 gain was 15.5 cells/mm3 per month in ART-treated patients followed up for at least six
months in District MUs and Central ATCs and 12.0 cells/mm 3 per month for patients followed-up in
provincial ATCs (p48 hours) during the previous four weeks concerned
11.6% of the patients. No difference was observed between the different levels of health care
decentralization. Multivariate analyses showed that low income, high alcohol consumption and
high negative perception of ART toxicity (measured as number of perceived treatment side effects)
were independent predictors of ART interruption. Some structural factors were also identified as
associated with ART interruption: ART shortage, lack of counselling by social workers or PLWHA
associations, low medical facilities equipment and high degree of task delegation from doctors to
health care workers were predictive of ART interruptions.
Finally, results showed that physical HRQL was similar at all levels of health care delivery; whereas
mental health was significantly better at district level. Social and/or economic insecurity and bad
health (hospitalization during the previous six months, low body mass index, high perceived drug
25

side effects) were associated with lesser physical and mental HRQL. Several supply-side
characteristics were also associated with HRQL: the lack of counselling by social workers or PLWHA
associations had a negative impact on mental quality of life; and a low proportion of physicians in
the medical staff was associated with a lower physical quality of life.
Decentralization and earlier entry into HIV care
Fifteen percent of patients reported a delay of at least six months before their first consultation
after HIV diagnosis. However, the proportion of patients who reported a delay of more than six
months before entry into HIV care was lower in patients followed up at district level: only 9%
reported a delay of consultation (>six month(psix months before consulting) included the characteristics of the HIV service
(higher delay in provincial and central ATCs and in HIV services located in small or medium-size
hospitals) and a number of patient characteristics: gender and marital status, the circumstances of
the HIV diagnosis (test not performed in the hospital providing HIV care, test performed during a
voluntary screening campaign) and patient’s negative perception of ART toxicity.
Structural factors limiting health care quality at central, provincial and districts levels
Several supply-side characteristics have been identified as limiting health care quality. Lack of
physicians in the HIV medical staff has a negative impact on three criteria of health quality
evaluation (gain in CD4, treatment interruptions, and HRQL); lack of psychosocial support by social
worker has a negative impact on two criteria (treatment interruptions and mental HRQL).
A high degree of task shifting to nursing staff in the care of ART-treated patients, ARV shortage and
ARV delivery in an open space all increased the risk of ART interruption.
Finally, a lack of technical equipment was associated with low physical HRQL and a lack of
psychosocial support by experienced staff (social workers, community workers or associations of
PLWHA) was associated with low mental HRQL.
Financial accessibility of HIV-services and equity in access to care and treatment
At the time of the study, implemented before the adoption of free access to ART, out-of-pocket
payments to access HIV care were significant. Even after the May 2007 decision to provide free-ofcharge
ARVs, direct and indirect costs of HIV disease may still constitute barriers to HIV care access
and care efficiency.
Impact of user fees for ART on treatment effectiveness
As noted above, at the time of the survey, monthly ART prices were established for all HIV
treatment centres at 3000 FCFA, or about 6 United States dollars, for Triomune, and 7000 FCFA, or
about US$ 14, for other treatments. Analyses conducted among the sample of ART-treated
patients showed that 20% reported financial difficulties in purchasing their antiretroviral drugs
during the preceding three months. After adjustment for socioeconomic and clinical factors,
reports of financial difficulties in purchasing ART were significantly associated with lower
adherence to ART and with lower CD4 lymphocyte counts after six months of treatment.
26

Total health expenditures for HIV care and catastrophic health expenditures
In the study sample, health expenditures (other than for ART treatment) equalled on average US$
11.3 per month for treated patients and US$ 18.6 per month for untreated patients.
For treated patients, these expenditures consisted primarily of transport and medication (other
than ART treatment) costs; for untreated patients, expenditures were devoted for the most part to
medication, biological exams and hospitalization.
Without considering expenditures for ART, about a quarter of treated patients and a third of
untreated patients enrolled in the study still faced catastrophic health care expenditures.
Catastrophic expenditures were logically associated with patients’ income and more frequently
affected patients who were not officially classified as “indigent” and so did not qualify for an
exemption from user fees 8 . The risk of catastrophic expenditures was also higher for those who did
not live in the same city as the follow-up hospital; who consulted a physician outside of the
hospital; or who had been hospitalized during the previous six months.
However, the risk of facing catastrophic expenditures was lower for patients followed in HIV
services of the provincial or district levels compared to those followed in the ATCs of the central
level.
Direct and indirect costs of HIV disease
Besides health expenditures, PLWHA face supplementary costs, known as "indirect costs"
connected to the loss of professional activity and thus resources. These costs were greater for
untreated patients than for treated patients, as treated patients are more likely to have an
occupation (71% of treated patients had an occupation at the time of the survey versus 56% of
untreated patients, p

Prevention behaviours
Nearly half (48%) of the HIV patients were living in a couple (defined as marriage or free union and
sharing or not the same home) and 45% reported sexual activity during the previous 3 months.
Among these patients, 89% declared one main sexual partner and 11% several partners.
Regarding unsafe sexual behaviours with the main partner, among the 907 patients who had
sexual intercourse during the three months prior to the study with a main partner of seronegative
or unknown HIV status, 35% reported not systematically using condoms. Analysis showed that
patients receiving ART were two times less likely to report inconsistent condom use than nontreated
patients. This may be partly due to the fact that ART-treated patients had better
information on prevention strategies as messages of this type are essentially given at the time of
diagnosis and at initiation of therapy. These results have important public health implications in
terms of secondary prevention. They are confirmed in the female population that represents the
majority of PLWHA in Cameroon.
Discussion
Data suggest that the recovery of health due to ART has opened up new life perspectives for
PLWHA. The independent evaluation of the national ART programme confirmed the feasibility of
decentralisation in Cameroon: ART-delivery centres based in decentralized district hospitals
present similar and sometimes better performance than referral centres. Closer proximity of
district hospitals facilitates patients’ adherence to treatment, improvement of their psychological
well-being, and earlier access to treatment after HIV diagnosis. Quality of care issues and patients’
dissatisfaction seem to be more prominent at higher levels of the health delivery pyramid because
of the higher number of patients attending provincial and central hospitals and the higher
workloads of their health care staff. These results highlight the necessity to adapt medical
resources to the needs in order to anticipate possible bottlenecks in HIV departments which could
limit access to care at any level of the health delivery pyramid.
In addition, the evaluation revealed the presence of several barriers which can have an impact on
the quality of care:
• The difficulties of supply of ARV drugs and reagent for CD4 exams may result in shortages
of stock which may have a negative impact on treatment adherence.
• Lack of highly qualified medical staff, in particular physicians and social workers, is
associated with lower immunological improvements, lower adherence and a poorer
quality of life for patients. An insufficient number of physicians also limits access to care,
contributes to delays in starting up ARV treatment and undermines equity.
• Finally, lack of motivation among the medical staff (possibly associated with
precariousness of their employment status, weakness of salaries and high workload) can
also influence health care quality, although the impact of this phenomenon could not be
quantified.
This underlines the urgency to strengthen health services and to set up human resources
strategies adapted to the needs, in order to maximize the efficiency of decentralization. Efforts
28

must also involve a better distribution of tasks among staff, in particular in the medical follow-up
of patients, and a revalorization of medical and paramedical professions.
Despite the policy of free ARVs introduced in May 2007, financial difficulties in access to care have
not been solved. While decentralization of HIV care reduced the risk of catastrophic health care
expenditures for PLWHA, in the EVAL sample about a quarter of the ART-treated patients still face
catastrophic health care expenditures. The sustainability of the free ARV drugs policy at the point
of delivery and its potential extension to other components of HIV/AIDS services are needed more
than ever in Cameroon and will necessarily imply the improvement of tax-based funding, and/or
health insurance risk pooling mechanisms, as well as the allocation of a greater share of public
resources to the health system.
29

References
[1] The World Bank Group. Cameroon at a glance. Washington, DC, World Bank, 2008
(http://devdata.worldbank.org/AAG/cmr_aag.pdf; accessed 20 May 2009).
[2] The World Bank Group. World Development Indicators (WDI) Online.
[3] United Nations Development Programme. Human Development Reports: Cameroon. New York,
UNDP, 2008 (http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_CMR.html;
accessed 20 May 2009).
[4] World Health Organization. Cameroon Summary Country Profile for HIV/AIDS Treatment Scale-
Up. Geneva, WHO, 2005 ( http://www.who.int/hiv/HIVCP_CMR.pdf; accessed 20 May 2009).
[5] INS (Institut National de la Statistique) & ORC Macro. Demographic Health Survey in Cameroon
[in French]. Calverton, Maryland, USA, 2004.
[6] UNAIDS. 2008 Report on the global AIDS epidemic 2008.
[7] World Health Organization. WHO Statistical Information System (WHOSIS).
(http://www.who.int/whosis/en/).
[8] World Health Organization. Global Tuberculosis Database.
(http://www.who.int/globalatlas/dataQuery/default.asp).
[9] World Health Organization. The World Health Report 2006. Working together for health.
Geneva, WHO, 2006.
[10] World Health Organization. Global Atlas of the Health Workforce 2008. Geneva, WHO, 2008.
[11] Ministère de la Santé Publique, République du Cameroun. Guide national de prise en charge
des personnes vivant avec le VIH/SIDA niveau hôpital de district. Yaoundé, Ministère de la Santé
Publique, République du Cameroun, 2005.
[12] World Health Organization. Scaling up antiretroviral therapy in resource-limited settings:
treatment guidelines for a public health approach (2003 revision). Geneva, WHO, 2004.
30

Cameroon: Impact of Global Health Initiatives on
primary level health care facilities:
Abstract
The case of Kumba and Limbe Health Districts
Peter M. Ndumbe, Julius Atashili 9
In the past five years, The Republic of Cameroon has received support from at least two GHIs: the
Global Fund and GAVI. Although national data suggest an overall improvement of health service
delivery and outcomes with these initiatives, it is not clear what their impact has been on
peripheral primary level health facilities. We surveyed 18 health facilities in the Kumba and Limbe
health districts in the South-West region of Cameroon. Overall, data collection and storage in these
facilities was rudimentary, seriously hampering an exhaustive assessment of the impact of GHIs.
Nevertheless, the limited available data suggest modest increases in human resources and not
much improvement in the financing of these health facilities, the infrastructure, and selected
indicator services such as referral services, tuberculosis care services and radiology services. Most
peripheral facilities could identify specific immunization activities that were indirectly supported
by GAVI. The impact of the Global Fund was mainly experienced in the regional (provincial) and
district hospitals, in which there were improvements in the human resources, and facilities
involved in tuberculosis and/or HIV care. Notwithstanding, these services still appeared
insufficient, as district level data revealed a lower than expected vaccine coverage rate and low
antiretroviral adherence rates. The assessment of the impact of GHIs in this milieu and their
subsequent improvement will require a more rigorous monitoring and evaluation system at the
peripheral level.
Background
Cameroon is bordered by the Federal Republic of Nigeria to the west; the Republic of Chad to the
northeast; the Central African Republic to the east; and the Republic of Equatorial Guinea,
Gabonese Republic, and the Republic of the Congo to the south. Cameroon ranked 150th out of
179 countries on the UN Human Development Index in 2006 [1].
The health system in Cameroon is organized at three levels: a central level including national
administrative units in the Ministry of Health (MOH), along with referral hospitals, responsible for
setting policy; an intermediary level, including regional delegations and regional programme
coordinators, in charge of technical support to district-level programmes; and a peripheral level,
which is composed of health areas grouped into health districts and is considered the operational
level for the provision of primary health care services.
9
Faculty of Health Sciences, University of Buea / Centre for the Study and Control of Communicable Diseases, University
of Yaounde I
31

Out-of-pocket payments account for more than two-thirds of total health expenditure, and
external resources made up 7.1% of the total expenditure on health in 2006 [2]. Physicians are
largely concentrated in the urban areas. In 2004, 69.94% of the country’s physicians worked in
cities, but only 53.42% of the population lived in cities [3].
Official development assistance to Cameroon was US$ 1.68 billion in 2006 [4]. Debt relief under the
Heavily Indebted Poor Countries initiative in 2006 brought Cameroon’s total external debt down
from US$ 7.2 billion in 2005 to US$ 3.2 billion in 2006 (17.7% of GDP) [4,5].
Map of South West Region indicating Kumba and Limbe
Health Districts, Cameroon
Limbe
Kumba
Source: Orok JB : Etude des causes de non utilisation des services de vaccination dans un contexte de couverture vaccinale administrative faible dans le District de Santé de
Error! Bookmark not defined.
KUMBA (Cameroun). Mémoire de DIU, 2008
32
Map of Cameroon indicating
South West Region, Cameroon
Source: http://en.wikipedia.org/wiki/Image:Provinces_of_Cameroon_EN.svg

Objectives and methodology
Our objectives were to assess the impact of the Global Fund and GAVI in primary health care
facilities in the Limbe and Kumba Health Districts in Cameroon.
The study was exempt from ethical review as not involving human subjects and not involving the
use of identifiable patient information.
Trained research assistants visited 18 health facilities serving the populations of the Kumba and
Limbe Health Districts. In each health district, the health centres surveyed were conveniently
sampled to reflect the urban versus rural make-up of health areas as well as to include facilities run
by both the government and non-governmental organizations. We also surveyed the two district
hospitals (first level referral) and regional hospital (second level referral) serving these two districts.
The facility data abstraction form developed for Maximizing Positive Synergies (MPS) was used in
this study. In each facility we abstracted data on financing, personnel, infrastructure and health
outcomes, as much as available. Available personnel were also queried for non-recorded data,
particularly for preceding years.
We surveyed 18 facilities spread over 16 health areas in both health districts. The overall
population covered in both districts was estimated at 428 991 inhabitants. The majority (12) of
these facilities were government run, with others being run by religious institutions (4) and
parastatal companies (2).
Results
Financing
Budget lines for funds from either GAVI or the Global Fund were not reported by all facilities.
Funding for these GHIs was managed at the intermediate and central levels of the health system,
the peripheral facilities being focused on implementation. Because the funds from these GHIs are
pooled with funds from other sources (such as the government and other bilateral and multilateral
donors) in the central/intermediate levels before dissemination to peripheral levels, it was difficult
for managers at peripheral health facilities to estimate the proportion of funding received from
each GHI. Most managers, however, acknowledged receiving support in the form of free vaccines,
materials for safe injections and cold chain, first-line treatment for tuberculosis, malaria, HIV and
opportunistic infections, as well as reagents for laboratory diagnosis.
Amongst the facilities that could estimate the proportion of different funding sources, there was a
trend towards a lesser proportion of funding from both the MOH and out-of-pocket spending. The
number reporting full funding from the MOH reduced, while the number with less than 50% of
funding from the MOH increased. Concurrently, the number of facilities with less than 50% out-ofpocket
funding increased in the last five years.
Health Workforce
The greatest change in the health system appears to have been the increase in the number of
health care personnel. The total number of human resources increased in half the facilities
surveyed. This increase was mainly driven by both the number of doctors and pharmacy assistants
33

as well as the number of community and lay workers. Surprisingly, up to five facilities reported
reductions in the number of personnel. These were either facilities in rural areas (3) or urban
facilities that experienced changes in the focus of services provided (2). Despite this overall
increase, the number of health personnel was still low: 40 doctors and 218 nurses, corresponding
to one doctor for every 10 725 inhabitants and one nurse for every 1 968 inhabitants.
Service Delivery
Overall there was a slight improvement in the hospital facilities. The number and size of inpatient
beds, consultation rooms and operation rooms either stayed constant or increased. One facility
reported a decrease in the number of delivery ward beds but this simply resulted from converting
a delivery ward into an inpatient ward.
Referral services
There was little change in the referral services in the last five years. The number of facilities with
ambulances or facilities that accompanied patients from facility to facility remained static.
Tuberculosis care services
There was a trend towards a slight increase in the availability of tuberculosis services. The number
of facilities providing TB diagnosis services, TB treatment, Directly Observed Therapy, Short-course
(DOTS) and with access to multi-drug resistant TB (MDR-TB) treatment slightly increased.
Radiology services
There was no improvement in access to radiology services. About 80% of facilities report having
access to X-ray and ultrasound services either on site or within two hours of the facility.
GHI-related outcomes
The facility survey was limited in its ability to comprehensively quantify health outcomes related to
the Global Fund and GAVI. However, two independent studies assessing certain aspects of these
GHIs had recently been conducted in the Limbe Regional Hospital and the Kumba Health District.
A study of the impact of a new policy rendering free the provision of antiretrovirals (to AIDS
patients), a policy primarily supported with Global Fund funding [6], showed that while this policy
was followed by a slight (not statistically significant) increase in the total number of patients
enrolled for antiretroviral treatment in the Limbe Regional Hospital (from an average of 50 to 60
new patients monthly), it was associated with a significant increase in the adherence to treatment
from 27.6% to 37.4% (self-report of taking prescribed treatments at least 95% of the time).
34

Another survey of immunization coverage in the Kumba Health District showed that 10% of infants
had never received any vaccine while approximately 15% were incompletely immunised
according to the locally recommended immunisation schedule [7].
Discussion
We conducted this survey to appreciate the potential changes at the primary level of the health
system following the introduction of GHIs in Cameroon. Little reliable data were collected at each
facility to validly ascertain the recent changes in health outcomes in the facilities surveyed.
Nevertheless, the limited data available suggest modest increases in human resources and little
improvements in the financing of these health facilities, the infrastructure, and selected indicator
services such as the referral services, the tuberculosis care services and radiology services. Most
peripheral facilities could identify specific immunization activities that were indirectly supported
by GAVI. The impact of the Global Fund was mainly experienced in the regional (provincial) and
district hospitals in which there were improvements in the availability of human resources and
facilities involved in tuberculosis and/or HIV care.
While we describe some changes in the health system, these changes are not all attributable to
GHIs as multiple other initiatives could be implicated. For example with funding from the Highly
Indebted Poor Countries (HIPC) initiative, thousands of healthcare workers were recruited into the
national health workforce. Funding from bilateral sources such as the German co-operation (GTZ)
or the French assistance (through the “Contrat de Désendettement et de Développement”, C2D)
and international sources such as the African Development Bank and the World Bank also
improved the status of health facilities [8]. At best, the GHIs may have contributed to these
changes in the health system.
Deficiencies in data collection and storage at health facilities surveyed limited any exhaustive
assessment of the impact of GHIs. The improvement of health facilities in these districts will first
require a more rigorous monitoring and evaluation (M&E) of the processes and outcomes involved.
This may necessitate specific funding (either within the GHIs or through external mechanisms) to
set up M&E systems, including recruiting statisticians (or equipping health workers with basic
health statistics skills) and setting up electronic, rather than paper-based, data collection systems.
While the current health information system involved monthly reporting of data to intermediate
and central levels of the health system, there was no evidence of any meaningful use of these data
at peripheral facilities. Providing feedback to primary-level health care providers might go a long
way toward improving both the quality of the data collected and the use of this data to improve
health services at the primary level. The M&E need not be extensive: it could be based on a sentinel
system of representative and geographically diverse health districts and/or health facilities. The
accuracy of data collected could be guaranteed by setting up a prospective, rather than
retrospective, M&E system.
It will also be difficult to disentangle the specific effect on health systems of each GHI from that of
other funding sources, particularly at the primary level. Rather, it may be more realistic to assess
the overall impact of all funding sources. Future GHIs need to incorporate an assessment of
baseline indicators (including at primary level facilities) prior to their introduction, as an integral
part of the programme.
35

Table 1: Characteristics of facilities
Characteristics Frequency %
Type of facility Provincial Hospital 1 5.6
District hospital 2 11.1
Mission Hospital 3 16.7
Health centre 10 55.6
Health Post 1 5.6
Parastatal clinic 1 5.6
Level of facility Hospital 5 27.8
Health Centre 12 66.7
Health post 1 5.6
Managing authority Government 12 66.7
Mission 4 22.2
Parastatal 2 11.1
Acknowledge support from GFATM Yes 11 61.1
No 7 38.9
Acknowledge support from GAVI Yes 14 77.8
No 4 22.2
Table 2: Comparison of sources of funding and selected health services before and after GHI begun
Characteristics Before % After %
Source of funding
Ministry of health
0% 5 27.8 3 16.7
1-49% 1 5.6 6 33.3
50-99% 2 11.1 4 22.2
100% 4 22.2 2 11.1
Not reported
Out of pocket
6 33.3 3 16.7
0% 2 11.1 2 11.1
1-49% 3 16.7 5 27.8
50-99% 3 16.7 3 16.7
100% 0 0.0 0 0.0
Not reported
Referral services
10 55.6 8 44.4
Higher level facility within 2 hours by car 15 83.3 16 88.9
Radio/phone/internet connection to higher level facility 9 50.0 9 50.0
Ambulance 4 22.2 4 22.2
Transportation fee to facilitate referral 3 16.7 4 22.2
Referral form used from facility to facility 15 83.3 14 77.8
Referral accompanied from facility to facility 10 55.6 10 55.6
Referral form used from community to facility
Tuberculosis services
6 33.3 6 33.3
TB diagnosis 7 38.9 8 44.4
TB treatment 3 16.7 4 22.2
DOTS 2 11.1 3 16.7
Access to MDRTB treatment
Radiology services
1 5.6 2 11.1
X-ray on site 3 16.7 3 16.7
X-ray within 2 hours 13 72.2 12 66.7
Ultrasound on site 4 22.2 4 22.2
Ultrasound within 2 hours 13 72.2 12 66.7
36

Table 3: Estimated changes in health infrastructure between 2003 and 2008
Infrastructure Change N %
Inpatient ward beds Decrease 0 0.0
No change 13 72.2
Increase 3 16.7
Not reported 2 11.1
Inpatient ward size Decrease 0 0.0
No change 15 83.3
Increase 1 5.6
Not reported 2 11.1
Outpatient consultation rooms Decrease 0 0.0
No change 16 88.9
Increase 2 11.1
Outpatient clinic size Decrease 0 0.0
No change 15 83.3
Increase 3 16.7
Delivery ward beds Decrease 1 5.6
No change 15 83.3
Increase 0 0.0
Not reported 2 11.1
Delivery ward size Decrease 0 0.0
No change 16 88.9
Increase 0 0.0
Not reported 2 11.1
Operating room tables Decrease 0 0.0
No change 9 50.0
Increase 0 0.0
Not reported 9 50.0
Operating room size Decrease 0 0.0
No change 9 50.0
Increase 0 0.0
Not reported 9 50.0
Counseling rooms Decrease 0 0.0
No change 13 72.2
Increase 1 5.6
Not reported 4 22.2
Counseling rooms size Decrease 1 5.6
No change 12 66.7
Increase 1 5.6
Not reported 4 22.2
N: number of facilities
37

Table 4: Estimated changes in human resources between 2003 and 2008
Human resources Change N* %
Total Decrease 5 27.8
No change 4 22.2
Increase 9 50.0
Doctors Decrease 0 0.0
No change 10 55.6
Increase 8 44.4
Nurses and assistants Decrease 6 33.3
No change 6 33.3
Increase 6 33.3
Laboratory technicians and assistants Decrease 1 5.6
No change 12 66.7
Increase 5 27.8
Pharmacists and assistants Decrease 0 0.0
No change 12 66.7
Increase 6 33.3
Community and Lay workers Decrease 3 16.7
No change 8 44.4
Increase 7 38.9
Others Decrease 2 11.1
No change 14 77.8
Increase 2 11.1
N: number of facilities
38

References
[1] United Nations Development Programme. Human Development Reports: 2008 Statistical
Update. Cameroon. New York, UNDP, 2008.
(http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_CMR.html; accessed 19
May 2009).
[2] World Health Organization. WHO Statistical Information System (WHOSIS). Geneva, WHO
(http://www.who.int/whosis/en/, accessed 19 May 2008).
[3] World Health Organization. Global Atlas of the Health Workforce. Geneva, WHO, 2008.
[4] The World Bank Group. World Development Indicators (WDI) Online. Washington, DC, World
Bank.
(http://web.worldbank.org/WBSITE/EXTERNAL/DATASTATISTICS/0,,contentMDK:21725423~pageP
K:64133150~piPK:64133175~theSitePK:239419,00.html).
[5] The World Bank Group. Cameroon at a glance. Washington, DC, World Bank, 2008.
(http://devdata.worldbank.org/AAG/cmr_aag.pdf; accessed, 19 May 2009).
[6] Ebeng Depe RS. Evaluation de la gratuité des ARV au Cameroun: cas du centre de traitement
agréé de l’hôpital régional de Limbe. Mémoire présenté en vue de l’obtention d’un “Master en
Santé Publique”, University of Antwerp, Belgium, 2008.
[7] Orok JB. Etude des causes de non utilisation des services de vaccination dans un contexte de
couverture vaccinale administrative faible dans le District de Santé de KUMBA (Cameroun).
Mémoire pour l’obtention du Diplome Interuniversitaire, Ouidah, Benin, 2008.
[8] Government of Cameroon. Rapport de situation annuel 2006, GAVI. May 2007.
39

Central African Republic: Impact of Global Health
Initiatives health systems:
perceptions of informants
Julius Atashili 1,2 , Marie-Claire Okomo 2 , Emilia Lyonga 2 , Nayana Dhavan 3 , Nikita Carney 3 , Erin Sullivan, 3 Peter Ndumbe 1,2
Abstract
The Central African Republic (CAR) has received funds from GHIs targeting HIV, malaria, TB and
vaccine-preventable diseases. We interviewed 18 informants involved in these GHIs. Overall, GHIs
were thought to have positively impacted the governance, financing, monitoring and evaluation,
workforce, infrastructure and technology and community participation in the health system. There
was a perceived improvement in access to prevention and care of targeted diseases with multiple
treatments or vaccines being made free. GHIs also fostered decentralization of services, promoting
greater equity of access. Nevertheless, there was still room to improve the impact of GHIs.
Informants thought that GHIs should focus on reinforcing the health system, not just targeted
diseases. In the grant allocation procedures, consideration should be given to country needs and
not limited to the quality of proposals. More funding is needed for training, recruitment of human
resources and infrastructure. There was also a perceived need for better incentives and salaries for
health care workers and concern about the sustainability of the programs initiated with GHI
funding.
Background
The Central African Republic (CAR) is a landlocked country in Central Africa. Bordered by Chad in
the north, Sudan in the east, the Republic of the Congo and the Democratic Republic of the Congo
in the south, and Cameroon in the west, the CAR is home to 4.3 million people. Its population has
grown at an average rate of 1.7% since 2001 [1]. The CAR ranked 171 st out of 177 countries on the
UN Human Development Index in 2005[2].
Most of the population has limited access to basic health services. In 2008, 76% of the population
lived more than 10km from the nearest health centre [3]. Malaria remains the leading cause of
premature death in the CAR. The entire population is at risk of malaria, with an estimated 4268
malaria deaths in 2006 [4]. The HIV epidemic in the CAR is widespread and generalized and has
been steadily growing since the first AIDS case was reported in the early 1980s [5] In 2007, 160 000
people were living with HIV/AIDS [6]. The infant mortality rate in the CAR was 114 per 1000 live
births in 2006; the maternal mortality rate was 980 per 100 000 live births in 2005 [7]. WHO
estimated that incidence of smear-positive pulmonary TB in the CAR was 149 per 100 000 in 2006
[8]. Of all new cases, 18% occurred in HIV-positive individuals, and 1.1% were multi-drug resistant
(MDR-TB) [8].
40

Table 1 Basic Socioeconomic, Demographic, and Health Indicators*
*Full data sources for all indicators are provided in Annex 1
Indicator Value Year Source
Population (thousands) 4,343 2007 World Bank
Geographic Size (sq. km) 623,000 2007 World Bank
GDP per capita, PPP (constant 2005
international $)
673.66 2007 World Bank
Gini index 43.57 2003 World Bank
Government expenditure on health (% general
government expenditure)
10.893 2007 WHO NHA
Per capita government expenditure on health
(current US$)
6 2007 WHO NHA
Physician density (per 10,000)

Objectives and Methodology
In this paper we describe the impact of GHIs, primarily the Global Fund and GAVI, on the health
system as perceived by key informants in the Central African Republic.
We conducted a descriptive qualitative study in which we interviewed key health personnel at the
central level of the health system. With the assistance of WHO country representatives and
Ministry of Health (MOH) officials, 18 key informants involved in either the Global Fund or the GAVI
alliance were identified and interviewed in-person using a semi-structured questionnaire.
Interviews were conducted in French, translated to English and analysed using the software NVivo
8 (QSR International, Cambridge, MA). The transcripts were coded using the key themes reflected
in the WHO’s description of “building blocks” of a health system, health outcomes and crosscutting
themes.
Results
This paper is based on the descriptions and perceptions of key health personnel involved in the
administration of either the Global Fund or GAVI in the Central African Republic. These
descriptions are not represented as factual information, but rather, in recognition that the success
of GHIs depends, at least in part, on how they are perceived in the various countries in which they
are being implemented. As with every qualitative analysis, the description seeks to describe a
general common view presented by interviewees, not just anecdotal cases.
Leadership and Governance
The governance of GHI funding in the country was described as being in transition. As with many
other countries, there is a Country Coordinating Mechanism (CCM) involved in setting priorities
and channelling funds. In its initial applications the CAR designated the country office of the
United Nations Development Programme (UNDP) to be the primary beneficiary of funds from the
Global Fund. However these responsibilities are being gradually transferred to institutions under
the Ministry of Health (MOH) including the National AIDS Control Committee. There was a
perception that the original choice of UNDP as primary beneficiary may have resulted in confusion
on the respective roles of the UNDP and the MOH and its facilities – there was a feeling that UNDP
was taking over the role of the MOH and yet was not answerable to the local authorities.
There was a perception among informants that the procedures for applying for funds from the
Global Fund were complicated, demanding and not always easy to follow. Some informants felt in
particular that the choice of proposals to fund ought not to be based on the quality of proposals;
rather, that decisions ought to be based on the burden of diseases and the need for funds in each
country. One informant can be paraphrased as stating that “countries that lack financial resources
are also lacking in other things” implying a concurrent lack of human and technical resources to
develop competitive proposals.
On the other hand, rigid reporting requirements that accompany GHI-funding may have improved
capacity at the national level. Performance-based allocations policies by GHIs have been extended
to other domains of the MOH. However, reporting procedures were also considered to be
complicated, placing too much demand on scarce human resources better used in the delivery of
care than paperwork. This concern was particularly acute with secondary beneficiaries who lacked
42

the capacity for timely reporting. However, there appears to be a gradual transfer of management
capabilities from the national level to sub-national levels.
The influx of resources from the Global Fund is perceived to have accelerated the decentralization
of the governance of disease care, with non-governmental organizations (NGOs) playing a greater
role—particularly at peripheral levels in which there were no government-run facilities.
Nevertheless, the extent of decentralization was limited by human resources capacity. The role of
multilateral institutions, such WHO and the United Nations Children’s Fund (UNICEF), in the
overseeing management and the disbursement of funds was appreciated.
GHIs have also encouraged government partnerships with NGOs, both local and international, the
private sector and academia. Private clinics and organizations have taken more active roles in
HIV/TB/malaria care, slightly alleviating the burden on the government-run health facilities. The
private sector has also partnered with government in the distribution of bed nets, while the
national university’s Faculty of Medicine has been involved in training students on research topics
related to the target diseases.
Table 2 Global Health Initiative Investments*
*Full data sources for all indicators are provided in Annex 1
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 2, HIV/AIDS 24,904,652 23,056,692
Round 4, HIV/AIDS 15,126,131 8,495,262
Round 4, Malaria 16,663,897 12,671,585
Round 4, TB 4,569,039 3,111,176
Round 7, HIV/AIDS 15,799,899 1,845,868
Round 8, Malaria 13,324,208 0
TOTAL: 90,287,826 49,180,583
GAVI
Disease Priority Amount Approved (in US$)
Pentavalent vaccine 5,345,000
Yellow fever vaccine 1,604,000
Vaccine introduction grant 200,000
Injection safety 140,900
Immunization services support 1,611,360
Health systems strengthening 3,163,000
TOTAL: 12,064,226
43

Financing
There was a general appreciation of the funds being made available through various GHIs. The
process of channelling the funds was, however, considered to be excessively centralized.
According to informants, GAVI funding was pooled by the MOH with other funds targeting
immunization, in contrast to the Global Fund, which directed resources to a primary recipient and
sub-beneficiaries based on proposals.
Both positive and negative features were attributed to the GHI-funding processes. Performancebased
financing was perceived as important, allowing for accountability and less waste. It also
served as an incentive to improved performance, transparency and management. On the other
hand, delays between approval of funds and their disbursement apparently forced some activities
to be launched in the absence of funds or some activities to be delayed with a potential impact on
patient care. Restrictions placed on the use of funds–for example, cases in which funds could only
be used for the targeted diseases and not other, often more pressing, health problems–concerned
some informants. Furthermore, there was concern that human resources limitations were delaying
reporting, and might thus slow much-needed funding.
Most health-care expenditure in the CAR is in the form of out-of-pocket payments by patients.
While very far from ideal, this financial contribution by patients has meant that they had a say in
the operation of the system. A potential unintended consequence of the availability of free drugs
is that patient input in the choice of care could be markedly reduced or eliminated altogether.
Monitoring and Evaluation
The health information system is believed to have improved (with a greater and faster availability
of data) but much still needs to be done to computerize the system and train providers in the use
of information technologies. GAVI has improved surveillance systems, particularly for vaccinepreventable
diseases. There has also been some parallel reporting, as the indicators and
frequencies requested by the Global Fund differ from those used by the national health system.
Driven by reporting requirements of both GAVI and the Global Fund, providers and other health
personnel have been trained in data collection and are more aware of its importance.
Nevertheless, some informants felt that personnel responsible for data collection at the primary
level did not prioritize this task – they do not use the information locally, but simply send it to the
national level where the data are aggregated and sent on to funding organizations. Clinical staff
were also perceived to be focused on patient care and less inclined to spend time on data
collection. There is therefore a need for personnel dedicated to data collection and reporting.
Problems such as poorly understood formulae, language barriers or inaccurate translation have
also been encountered in the Global Fund reporting forms, underscoring the need for specific
training of personnel who collect primary data. In general the Global Fund reporting requirements
were considered more complex in comparison to the forms proposed by GAVI.
Health Workforce
GHIs had a noticeable positive impact on the size of the healthcare workforce in CAR. The capacity
(skills) and competencies of these personnel have also been reinforced through numerous
trainings and refresher courses offered to personnel in various parts of the health system, from
central level cadres to doctors, nurses, counsellors, laboratory technicians and community leaders.
44

Some of these personnel were supported with resources from the Global Fund, thus strengthening
the health system more broadly. Despite these improvements in the workforce, there was a
perception that salaries were still too low, and that more financial incentives were needed to
compensate for the increased patient load in health facilities.
While some informants believed that the general decentralization of services may have reduced
“brain drain” of skilled personnel from rural to urban areas, others thought that some health
workers had abandoned their ministry positions to work for GHIs, where compensation was better.
The freeze on public-sector recruitment imposed by IMF-backed structural adjustment programs
was perceived to be a problem: personnel recruited and trained to reinforce Global Fund activities
could not be absorbed into the public sector, despite obvious need. Some trained personnel were
reported to have left their posts because they could not secure a fixed contract with either the
MOH or the Global Fund. Strategies to support some of these temporary workers were being
considered, though precisely how that could be accomplished was a matter of debate.
Overall there was a perceived improvement in the health workforce following the advent of GHI
funding. Nevertheless, long patient wait times and over-burdened health services underscored
continuing challenges in this area.
Medical Products, Vaccines and Technologies
There was overwhelming agreement on the positive role played by GHIs in improving the
availability of drugs, health commodities, and equipment for laboratory diagnosis. GHI funding
was felt to have improved access to antiretroviral drugs (ARVs), antimalarials, bed nets, TB
treatment, treatment for opportunistic infections vaccines, sterile needles, syringes and infusion
supplies. Many laboratories, particularly in rural areas, now have the capacity to perform CD4
assays, as well as being equipped with microscopes for TB diagnosis, and reagents for other
biochemistry assays. Health facilities were also provided with equipment such as refrigerators and
motorcycles or transport vehicles for immunization campaigns. Some central-level units were also
believed to have received computers for monitoring & evaluation (M&E).
GHIs were felt to have improved the supply chain. In particular, Global Fund resources
strengthened the existing drug procurement infrastructure, while GAVI improved the cold-chain
supply for vaccines.
Technical support from international organizations, such as WHO, and from external consultants,
were also considered important in the development, validation and dissemination of local
management guidelines for the care of HIV/AIDS, TB and malaria.
Health Infrastructure
Several new facilities were constructed and existing ones renovated with GHI support. In some
cases voluntary counselling and testing (VCT) centres were erected next to existing hospitals. This
was particularly important in CAR, where much infrastructure has been destroyed by armed
conflict. The current level of infrastructure was still considered to be insufficient, with bed
shortages in many hospitals.
45

NGOs and Civil Society
The overall perception was that GHIs, the Global Fund in particular, substantially improved the
participation of NGOs in the health system in CAR. Local and international NGOs were represented
in the CCM. Some local NGOs also served as secondary beneficiaries of funds, while others have
been involved in HIV information and education campaigns, mosquito net distributions, homebased
HIV-care and even TB care. GAVI also supported NGOs involved in vaccination campaigns.
While NGOs are hampered by the lack of financial and human resources, there was a perception
that partnership with governmental institutions allowed for training in areas where their capacity
is presently inadequate. Informants believed that NGOs played an important role in the system
and that their participation was a plus for the system. There was however a concern that some
NGOs would not be likely to be sustainable in the absence of GHI funds.
Service Delivery
Overall there were perceived improvements in all aspects of health care delivery and these were
not limited to HIV/AIDS, TB and malaria or vaccine-preventable diseases. There was also improved
public awareness about the target diseases. Testing and counselling for HIV was fuelled by the
sense that an HIV diagnosis was not fatal, owing to increased access to antiretroviral treatment
(ART). With the availability of GHI funding the price of drugs and laboratory assays has been
markedly reduced. Malaria treatment and insecticide-treated bed nets are free for children and
pregnant women. TB treatment is also free and there have been fewer interruptions in drug
supply. Drugs are also more widely distributed nationwide. GHIs have thus made a significant
impact in the numbers of patients receiving services.
Under GAVI more vaccines have been made available and supplemental vaccination campaigns
have been organized with the aim of “catching-up” vaccinations in children who may have missed
some regularly scheduled vaccines because of armed conflict.
Key informants did not have hard data on the impact of the GHIs on health outcomes per se.
Nevertheless there was an impression that there was reduced mortality and prolonged survival in
patients with HIV. It was not known however whether HIV prevalence was decreasing, as the
surveys that have been conducted used different methodologies. It was not clear what impact may
have been made on the frequency of high-risk behaviour. Although data specific to malaria were
also lacking, at least one informant felt that an observed reduction in infant mortality could be
due, in part, to better malaria care.
GAVI was perceived to have improved vaccine coverage and also to have increased the number of
diseases targeted in the expanded immunization program. Vaccines against Haemophilus influenza
b and Hepatitis B Virus were expected to be added with support from GAVI.
Cross-Cutting Themes
Some themes discussed by the informants did not fit into any of the aforementioned categories or
included multiple categories. One such area was the coordination and integration of GHIs within
the health system. Informants thought that while some aspects of GHIs were well integrated in the
system, others were not. GHIs are well coordinated at the central level of the MOH. Drug supply,
training and M&E also appear to be well integrated. Funds from GAVI in particular have been
involved in strengthening the health system and there was a belief that GAVI was well integrated
46

as funds were pooled centrally, and some programme equipment (such as motorcycles and other
vehicles) has been used for other purposes. Nevertheless, there was an impression that
integration was not complete: the Global Fund was felt to recruit personnel and make funds
available only for target diseases. Some aspects of GHIs were parallel to existing systems,
essentially running as vertical programs. The monitoring, evaluation and reporting system of the
Global Fund and surveillance for polio were considered as examples of systems running parallel to
CAR’s National Health Information System.
With respect to equity and regional variation, there was an overwhelming sense that GHI funding
had made it possible for services (including preventive, screening/diagnosis and treatment) and
infrastructure to be decentralized from urban centres into remote rural areas, thus improving
access to care. The extent of decentralization was however perceived to be insufficient; rural
populations still needed more help.
Many informants had concerns about the long-term sustainability of services provided by GHIs.
What would happen to patients, infrastructure, personnel, and NGOs when funding was no longer
available?
Spillover Effects
Informants described a number of unplanned or unexpected effects of GHIs. On the positive side,
GAVI and the Global Fund strengthened overall health system infrastructure and improved
services for non -targeted diseases. On the other hand, negative effects included migration of
health care workers from non-GHI-funded units (or activities) to GHI-funded units (or activities). In
a few cases there was also a perception that free access to drugs could have negative
consequences, including a reduction in patient choice of treatment, undercutting of local
pharmaceutical companies (one informant mentioned a local infusion fluid manufacturer being
undercut by the provision of free infusion fluids from foreign sources), and undermining the costrecovery
system of financing in health facilities. For example, malaria is the most prevalent disease
in health facilities in the CAR. Prior to drugs and some laboratory assays being made free, a small
proportion of patient payments went to maintaining and running some health facilities. With
drugs being made free, the amount of money recovered could be substantially reduced, thus
impacting the financing of the health facilities.
Recommendations
Informants made numerous suggestions on how to improve the impact of GHIs on the health
system. Some informants thought that GHIs should focus on reinforcing the health system as a
whole, not just targeted diseases. In the grant allocation procedures, consideration should be
given to country needs and not simply the quality of proposals. Informants felt that technical
support should be made available to countries to assist in writing high-quality proposals. More
funding was also felt to be needed for training, recruitment of human resources and infrastructure.
Informants recommended efforts to improve reporting, archiving and computerization of the
health system. There was also a perceived need for better incentives and salaries for health care
workers. Finally, concern was expressed about the potential interruption of GHI funding; in the
event that discontinuation was being considered, informants stressed the importance of ensuring
that states can provide resources for the system to continue functioning at the same level.
47

Discussion
Interviews with key informants in the CAR suggest that they perceive GHIs to be particularly
important in the health system. GHIs are thought to have positively impacted the governance,
financing, monitoring and evaluation, workforce, infrastructure and technology, and community
participation in the health system. There was a perceived improvement in access to prevention
and care of targeted diseases, with multiple treatments or vaccines being made free. In addition to
these positive effects, there were a few negative spillover effects described; however, these were
largely overwhelmed by the positive effects of GHIs. Quantitative data will be needed to confirm
these perceptions. The recommendations by informants may be useful in maximizing the positive
synergies between GHIs and the CAR’s health system.
Acknowledgements
We thank health officials in the Central African Republic who provided their perceptions. We also
acknowledge writing and analysis assistance from the Global Health Delivery Project at the
Harvard School of Public Health. We also thank the WHO country representative for CAR and staff
for assistance in accessing study participants.
48

References
[1] The World Bank Group. Central African Republic Data-at-a-Glance 2008.
[2] United Nations Development Programme. Human Development Reports: Central African
Republic 2008.
[3] United Nations Office for the Coordination of Humanitarian Affairs. Central African Republic
Humanitarian Country Profile.
(http://www.irinnews.org/country.aspx?CountryCode=CFA&RegionCode=GL; Accessed 25
February 2009).
[4] WHO Global Malaria Programme. World Malaria Report 2008.
[5] World Health Organization. Central African Republic Summary Country Profile for HIV/AIDS
Treatment Scale-Up 2005.
[6] UNAIDS. 2008 Report on the Global AIDS Epidemic 2008.
[7] Core Health Indicators: Central African Republic. World Health Organization Information
System; 2008.
[8] WHO. Global Tuberculosis Control: Surveillance, Planning, Financing 2008.
[9] Humanitarian and Development Partnership Team Central African Republic. Health.
(http://hdptcar.net/blog/health/; Accessed 25 February 2009).
[10] The World Bank Group. World Development Indicators: Central African Republic; 2008.
[11] International Monetary Fund and International Development Association. Central African
Republic Joint World Bank/IMF Debt Sustainability Analysis 2008.
[12] GFATM. Central African Republic and the Global Fund.
(http://www.theglobalfund.org/programs/portfolio/?lang=e&countryID=CAF; Accessed 25
February 2009).
49

China: The Impact of the Global Fund HIV/AIDS
Programmes on Coordination and
Coverage of Financial Assistance
Schemes for People Living with HIV/AIDS
and their Families
Abstract:
Zhang Xiulan, Pierre Miège and Zhang Yurong 10
This study assesses the impact of the implementation of Global Fund financed programmes on
coordination between different institutions at the local level, as well as on the distribution of social
assistance schemes for people living with HIV/AIDS (PLWHA) and their families. It is based on a
large household survey (involving 1120 households and 4850 people) and in-depth interviews
with local public health leaders, conducted in 12 counties within four provinces (Anhui, Henan,
Yunnan and Guizhou) between October 2006 and April 2007. Analysis of these qualitative data
shows that Global Fund programmes contribute to improved coordination at the local level
because they are implemented through the leadership of the local Centers for Disease Control
(CDCs). The CDCs successfully coordinate the actions and programmes of the diversity of actors
involved in HIV/AIDS prevention and treatment. Household interviews reveal that counties
receiving (round 3 or 4) Global Fund monies benefit from improved access to treatment and better
coverage by the different financial support programmes. Consequently, PLWHA and their families
report lower economic pressure and better integration within the community. These findings
therefore point to better coordination between the Public Health Bureaux,, which register patients
and insure access to treatment, and Civil Affairs Bureaux in charge of the distribution of social
assistance to PLWHA.
Background
The world’s most populous country, China is home to 1.32 billion people and averaged a 0.6%
population growth rate between 2001 and 2007 [1]. The country ranked 81 st out of 177 countries
on the UN Human Development Index in 2005 [2]. The richest 10% held 34.9% of the country’s
income in 2004 [2]. China now has the world’s fastest-growing economy and is undergoing what
has been described as a ‘second industrial revolution.’ The economy averaged a real gross
domestic product (GDP) growth rate of 9.5% between 1997 and 2007. However, corresponding
progress on the country’s social indicators has lagged behind [1]. The economic disparity between
10School of Social Development and Public Policy, Beijing Normal University
Acknowledgement: The research for this study summary was funded by The Alliance for Health Policy and Systems
Research, based at the World Health Organization. The study summary was produced with the support of the GHIN
Network (www.ghinet.org).
50

urban China and the rural hinterlands, industrial expansion, and rising incomes, have accelerated
migration from the rural areas to the country’s eastern cities. Disease burden has largely shifted
from communicable to non-communicable diseases and injuries [3]. This transition, in
combination with decreasing birth rates and an increasingly elderly population, has created new
challenges for the health system. In addition, the outbreak of severe acute respiratory syndrome
(SARS) in southern China in 2003 cost an estimated US$ 11 billion globally and challenged the
Chinese government to strengthen its approach to public health [4].
Most low-income households in China face significant financial and other barriers to essential
health services and medicines. In 2004, an estimated 85% of the population lacked health
insurance and out-of-pocket payments constituted the majority of growing health expenditure [4].
In 2008, up to 30% of poor people in China stated that health is the single most important cause of
their poverty [5]. The new Rural Cooperative Medical Scheme (RCMS) aimed to achieve 100%
coverage by the end of 2008. However, benefits are limited to catastrophic illness and inpatient
medical services, pre-payment is frequently required, and reimbursement is low [5]. The public
sector’s share of overall health expenditures has steadily gone down in terms of real GDP [4]. A
large portion of the Chinese population seeks out traditional Chinese medicine (TCM), while the
government has promoted the development of a modern TCM industry [5]. China is one of the few
countries where doctors outnumber nurses, but many experienced health professions have
moved to hospitals or clinics in the cities [5].
Official Development Assistance (ODA) to China was US$ 1.25 billion in 2006 [6]. The country’s
total external debt was US$ 322.85 billion or 12.1% of GDP [1, 6]. Since 2003, the Global Fund has
approved US$ 482.5 million in funding for China and has disbursed US$ 313.1 million to date. Two
grants, one for HIV/AIDS and one for TB, have been approved, but are still awaiting funding [7].
Table 1 Basic Socioeconomic, Demographic and Health Indicators (*)
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 1,304,500 2007 World Bank
Geographic Size (sq. km) 9,327,488 2007 World Bank
GDP per capita, PPP (constant
2005 international $)
5,045.64 2007 World Bank
Gini index 46.9 2004 World Bank
Government expenditure on
health (% general government
expenditure)
10.3 2007 WHO NHA
Per capita government
expenditure on health at average
exchange rate (current US$)
51 2007 WHO NHA
Physician density (per 10,000) 14 2003 WHO SIS
Nursing and midwifery density
(per 10,000)
10 2003 WHO SIS
Maternal mortality ratio (per
100,000 live births)
45 2005 WHO SIS
DTP3 coverage (%) 93 2007 WHO SIS
Estimated adult HIV (15-49)
prevalence (%)
0.1 (

Table 2 Global Health Initiative Investments (*)
(*) Full data sources for all indicators are provided in Annex 1.
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 1, Malaria 6,406,659 6,242,698
Round 1, TB 91,118,721 66,174,166
Round 3, HIV/AIDS 302,919,984 92,033,746
Round 4, HIV/AIDS 63,742,277 56,475,863
Round 4, TB 56,140,000 51,096,518
Round 5, HIV/AIDS 28,902,073 19,934,233
Round 5, Malaria 38,522,396 26,117,213
Round 5, TB 49,453,178 28,039,132
Round 6, HIV/AIDS 5,812,876 5,164,504
Round 6, Malaria 7,047,932 6,686,758
Round 7, TB 5,313,263 1,104,382
Round 8, HIV/AIDS 19,916,552 0
Round 8, TB 28,561,014 0
TOTAL: 703,856,925 359,069,213
PEPFAR*
Year Amount Allocated (in US$)
2004 4,000,000
2005 7,250,000
2006 9,250,000
2007 9,750,000
2008 9,959,500
TOTAL: 40,209,500
*Not a PEPFAR focus country; above sums represent total allocations to PEPFAR
country programmes from bilateral U.S. sources including USAID, Department of
Health and Human Services, Department of Labor, and Department of Defense.
GAVI
Disease Priority Amount Approved (in US$)
Hepatitis B vaccine 21,953,000
Vaccine introduction grant 800,000
Injection Safety 15,925,729
TOTAL: 38,678,919
52

Context
In 2007, the number of PLWHA was estimated by a joint mission by the Chinese Ministry of Health
(MOH), UNAIDS and the WHO to be 700,000 (a range of 550,000-850,000) [8]. China is classed as
having a concentrated epidemic with a low HIV prevalence, (around 0.05 to 0.08%, depending on
data source) but high incidence rates among specific sub-populations. Indeed, 39% of reported
HIV-positive cases have been transmitted through needle exchange by injecting drug users (IDUs),
19% through blood sales, 18% through heterosexual transmissions, and 1% through homosexual
transmission. In total 70% of PLWHA are between 20-39 years old, and over 70% are male [9].
The epidemic mostly affects rural communities in the poorest provinces, especially in central
China, the south-west and the east. This patterning of the HIV/AIDS epidemic in China has required
the extension of social and financial assistance to these already economically vulnerable rural
populations [9].
China has benefited from Global Fund support in rounds 3,4,5,6 and 8, with US$ 421 million
approved and US$ 163 million disbursed (up to Round 6) for HIV/AIDS programmes. Most of the
funding has been disbursed to rural and relatively poor counties that have been identified as
needing special attention. Launched in 2003, mostly through the support of Global Fund Round 3,
the China Comprehensive Aids Response (CARES) project has supported HIV/AIDS activities in a
number of counties in the most affected provinces of China. Antiretroviral therapy (ART) has been
extended and scale-up has been impressive: in December 2004, 8500 people were receiving ART;
by October 2007, more than 31 000 people were being treated with ART including 805 children [9].
Different financial and economic support schemes have been set up, although these are not
funded by the Global Fund or other GHIs. The only programme designed specifically for PLWHA is
the “Four free services and one care” (si mian yi guan huai). This provides free treatment, free
voluntary counselling and testing (VCT), free prevention of mother-to-child transmission (PMTCT),
free schooling for AIDS orphans, and social relief for HIV patients. In addition, PLWHA and their
households are also eligible to join local level social assistance schemes managed by the Civil
Affairs Bureaux. These include the: “Five Guarantees” scheme, which provides a collective safetynet
for the rural elderly, people with disabilities, and orphans without family caregivers; the
Minimum Income Guarantee Scheme (Dibao); the Assistance for Extremely Poor Households
(Tekun); and the Medical Financial Assistance (MFA) [10].
Objectives and Methodology
The purpose of this study was to assess the extent to which the implementation of Global Fund
financed programmes has contributed to increased coordination between different institutions at
the local level, and whether this has led to improvements in access to social assistance schemes for
PLWHA and their families.
The survey took place between October 2006 and April 2007 in twelve counties within four
provinces: Anhui, Henan, Yunnan and Guizhou. Anhui and Henan, in central China, are two of the
provinces where most HIV infections were caused by the practice of blood selling. These provinces
were targeted with money from the third round Global Fund grant. In Yunnan and Guizhou, in the
south of the country, the main mode of HIV transmission is sharing needles and syringes amongst
IDUs. Round 4 funding was used to support HIV/AIDS activities in a number of locations within
these provinces.
53

In Yunnan and Henan, the epidemic is concentrated in a number of specific areas, whereas in
Anhui and Guizhou, PLWHA are relatively scattered, and therefore more difficult to reach. The
provincial governments of Anhui and Yunnan have been more responsive towards the epidemic
than Henan and Guizhou, and have integrated local groups and organizations in the design and
implementation of prevention and treatment programmes.
Table 3 shows the 12 counties within these four provinces that were surveyed. Many of these
counties have benefited from Global Fund monies as well as from a Pilot Programme developed
by the Ministry of Health (MOH). The latter was initiated in 2003 in 51 counties (extended to 127
counties in 2004), and aims to establish comprehensive HIV/AIDS prevention and control policies.
Funding is provided by the national government and is matched by funding from the three main
levels of local government (provincial, municipal and county levels). At the county level decisions
are taken by the local Public Health Bureaux, and the local Centre for Disease Control and
Prevention (CDC) leads on the programme implementation.
To study the changes in the way programmes were coordinated and implemented at the county
level, in-depth interviews were conducted with 25 local leaders in 12 counties: they included
heads of the local CDC and Civil Affairs Bureau, as well as hospital and clinic directors. A household
survey was also conducted: a questionnaire was distributed to 1120 households (representing a
total of 4850 people), and this aimed to elicit their current economic condition, experiences of
access to treatment, access to the different financial assistance schemes, and level of social
support from relatives and other community members.
Table 3: Global Fund and MOH Pilot Program in the Surveyed Counties of China
Province County Global Fund MOH Pilot
Programme
Henan Shangcai Round 3 Covered 368
Jingjiu Round 3 Covered 125
Anhui
Yingzhou
Funan
Round 3
Round 3
Covered
Covered
29
100
Linquan Round 3 Covered 57
Ruili Not covered Not covered 142
Yunnan Longchuan Round 4 Not covered 147
Yingjiang Not covered Covered 53
Zhijin Not covered Covered 45
Guizhou
Tongren
Duyun
Not covered
Round 4
Covered
Not covered
29
19
Nanming Not covered Covered 6
54
Number of
households
interviewed

Results
Leadership and Governance
Since the 1990s, multiple coordination mechanisms have been formed to address the HIV/AIDS
epidemic. The main state coordination mechanism is the State Council AIDS Working Committee
Office (SCAWCO), composed of 23 representatives from central ministries and some key provinces.
Some ministries, provinces and municipalities have also set up committees to coordinate HIV/AIDS
activities. Several attempts to improve coordination resulted in the establishment of the China
Country Coordination Mechanism for HIV/AIDS Prevention and Control (CCM) in 2002. This was
founded in order to meet the requirement for applying for Global Fund money. In 2003, the
Government decentralized the implementation of Global Fund programmes, allowing much
greater leadership at the county level.
In 2003, the Chinese government selected the local CDC to coordinate and supervise all the Global
Found programmes dedicated to HIV/AIDS. This study explores the effect that the change in
coordination had at the local level. The interviews with local actors leading on health care policy,
doctors, directors of methadone clinics and other facilities aimed at high-risk groups reported that
the restructuring and reorganization of the CDC had greatly improved coordination. They
indicated that meetings were regularly organized with the main leaders of the bureaux and
institutions concerned with HIV/AIDS-related programmes. These include: Civil Affairs Bureaux in
charge of distributing financial aid and welfare allowances; Public Security Bureaux which oversee
the work of rehabilitation centres; Public Health Bureaux that run and supervise the hospitals and
clinics; local private doctors; and Education Bureaux which develop information and prevention
programmes.
Given that most of the counties studied are relatively poor and transportation is difficult, the need
for good cooperation between all the institutions operating in the concerned areas is increased.
Interviewees reported that since the CDC had been established as the mechanism responsible for
coordination, there had been better exchanges between the different activities, policies and
programmes and noted that crucial information is now shared by the different organizations. For
example, the meetings enable the Bureaux that register new cases of infection (Public Health and
Public Security) to inform the Civil Affairs Bureaux – which distribute financial support.
Service Delivery
The only programme designed specifically for PLWHA is the “Four free services and one care” (si
mian yi guan huai). This provides free treatment, free VCT, free PMTCT, free schooling for AIDS
orphans, and social relief for HIV patients. In addition, PLWHA and their households are also
eligible to join local level social assistance schemes managed by the Civil Affairs Bureaux. These
include the: “Five Guarantees” scheme, which provides a collective safety-net for the rural elderly,
people with disabilities, and orphans without family caregivers; the Minimum Income Guarantee
Scheme (Dibao); the Assistance for Extremely Poor Households (Tekun); and the Medical Financial
Assistance (MFA).
The interviews conducted with heads of households confirm the positive effect of improved
coordination. Counties receiving (round 3 or 4) Global Fund monies, showed improved access to
treatment, better coverage by the different financial support programmes, and, consequently, a
lowering of economic pressure on the PLWHA and their families. In the survey, 63% of the PLWHA
(889 persons) received ART. In the counties benefiting from both the MOH Pilot Programme and
the Global Fund programmes, 80% of the PLWHA (700 persons) received treatment. In counties
55

funded only from the MOH Pilot Programme, 25% (38 persons) received treatment, and in the
counties funded only from the Global Fund, 36% of PLWHA (67 persons) received treatment. In the
county that did not receive any funding from the Global Fund or the MOH, 30% of PLWHA (51)
accessed ART.
Similarly, PLWHA in Global Fund and MOH Pilot Programme financed counties had a significantly
higher chance of receiving financial assistance: 71% of PLWHA’s families received financial
assistance from at least one scheme, compared to only 11% of families in counties benefiting from
the MOH Pilot Programme but not from Global Fund monies. These results illustrate a synergy
between the Global Fund-supported programme and the activities financed by the MOH: the MOH
Pilot Programme is expanding the availability of resources at the local level to support households
in financial need, but alone, it offers poor coordination and does not lead to a better distribution of
these resources.
In the counties included in Global Fund round 3 and round 4 programmes families reported higher
average income and less impact on their daily lives, as well as on family relations, marital relations,
and relationships with neighbours. Interestingly, and probably due to the improvement in the
financial conditions of the concerned families, in such counties PLWHA reported higher levels of
support from spouses and other family members.
Discussion
The results from the household questionnaires show that PLWHA living in counties that received
resources from the Global Fund have a significantly higher probability of having access to
antiretrovirals (ARV), as well as treatment for opportunistic diseases. PLWHA, and their families, are
also better covered by the different financial and welfare programmes, reducing the economic
impact of infection on those rural households. As these social assistance programmes are
managed by the local Civil Affairs Bureaux, these findings confirm that in the counties that have
received resources from the Global Fund, there has been improved coordination and better
information sharing between the institutions. The impact of extra financial resources has been
strengthened by the decentralized leadership and responsibilities given to the local CDCs, which
have reduced inefficiency in programme design and implementation, and helped to distribute
funds to the organizations and the households which needed them most.
These findings are similar in counties included in the MOH Pilot Programme and in those which do
not benefit from it: in these two groups of counties, Global Fund programmes lead to better
coordination and expanded access to treatment. More importantly, the MOH Pilot Programme
allocates crucial extra funding to local governments, but in itself, does not lead to increased
coverage of financial assistance or access to treatment. However, when a MOH Pilot Programme
county is also benefiting from Global Fund monies, new resources are more efficiently distributed
to PLWHA’s families, demonstrating the impact of Global Fund programmes on CDC coordination
and therefore on the overall implementation of economic and social support schemes.
Interviews with local leaders and families of PLWHA show that the Global Fund programmes have
positively contributed to the delivery of the government welfare and support programmes for
PLWHA. The improvements in the coverage of support programmes have a direct impact on the
income of these households, helping reduce tensions and problems within families, and increasing
the level of support from relatives and other members of the community.
Most interviewees reported the Global Fund grants were an indispensable supplement to
government efforts in the fight against the HIV/AIDS epidemic. They are perceived to have
56

induced changes in the way the local administrative institutions operate, and have been matched
by funds dedicated to financial support and welfare. Better access to treatment on the one hand
and expanded coverage of these financial aid programmes on the other, have improved the daily
life of PLWHA and their families, and contributed to lowering social and psychological pressures
from relatives and other members of the community.
However, some problems remain. In some counties, the meetings organized by the CDC have led
to an increase in the number of HIV-related activities of various institutional actors, and therefore
to a lowering in the efficiency of programmes’ implementation. In other counties, better
coordination has not persuaded all the local bureaux to participate in programmes, leaving most
of the burden to the Public Health Bureaux, which sometimes complain about an increased
workload.
Furthermore, the efforts of coordination must be extended horizontally and vertically. Interviews
with local leaders suggest that experiences from other counties are not shared, and thus best
practices do not serve as lessons for counties facing similar challenges. Inter-county exchanges
would be a valuable experience for lesson-sharing and assessing progress. Finally, there are still
some difficulties in programme implementation, revealing the necessity for improved
coordination systems between different levels (central, provincial, county, and village) in order to
distribute resources efficiently and to those communities most in need.
57

References
[1] China at a glance. World Bank, 2005
(http://www.worldbank.org.cn/English/Content/chn_aag02.pdf; accessed 20 March 2009).
[2] Human Development Reports: China 2008 Statistical Update. New York, United Nations
Development Programme, 2008
(http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_CHN.html;
accessed 20 March 2009).
[3] Country Cooperation Strategy at a glance: China. Geneva, World Health Organization, 2008
(http://www.who.int/countryfocus/cooperation_strategy/ccsbrief_china_en.pdf; accessed
20 March 2009).
[4] World Health Statistics 2009 (http://www.who.int/whosis/en/).
[5] World Development Indicators 2007. Washington DC, The World Bank, 2007 (Proprietary online
database: http://ddp-ext.worldbank.org.ezpprod1.hul.harvard.edu/ext/DDPQQ/member.do?method=getMembers;
accessed 20 March
2009).
[6] China National Health Account. Geneva, World Health Organization, 2009
(http://www.who.int/nha/country/chn.pdf; accessed 20 March 2009).
[7] Epidemiological Fact Sheet on HIV and AIDS: Core data on epidemiology and response, China
2008 Update. Geneva, UNAIDS/WHO Working Group on Global HIV/AIDS and STI Surveillance,
2008 (http://www.who.int/globalatlas/predefinedReports/EFS2008/full/EFS2008_CN.pdf;
accessed 20 March 2009).
[8] 2005 Update on the HIV/AIDS Epidemic and Response in China. Beijing, Ministry of Health,
People’s Republic of China, Joint United Nations Programme on HIV/AIDS, World Health
Organization, 2006 (http://data.unaids.org/publications/External-
Documents/rp_2005chinaestimation_25jan06_en.pdf; accessed 20 March 2009).
[9] UNGASS Country Progress Report, P.R. China, January 2006-December 2007. State Council AIDS
Working Committee Office, United Nations Theme Group on AIDS, 2008
(http://data.unaids.org/pub/Report/2008/china_2008_country_progress_report_en.pdf;
accessed 20 March 2009).
[10] Xu Y, Zhang X, Zhu X. Medical Finance Assistance in Rural China: Policy Design and
Implementation. Studies in HSO&P, 2008, 23:295-317.
58

Georgia: System-wide Effects of the Global Fund
on Georgia’s Health Care Systems
Abstract
Ketevan Chkhatarashvili, George Gotsadze, Natia Rukhadze ∗
This study assesses the effects of the Global Fund on the health system in Georgia, focusing on the
policy environment, public-private interactions, human resources and access to HIV/AIDS services.
The Global Fund is the largest GHI in Georgia and has disbursed over US$ 26 million over six
rounds of funding. These grants contributed to approximately 2.8% of total health expenditure in
the country.
Global Fund resources have led to more HIV/AIDS preventative, diagnostic, curative and care
services, and currently antiretroviral therapy (ART) is available to all people known to require
treatment. Grants have also been used to develop the capacity of government and
nongovernmental health providers and include addressing issues of stigma and marginalization.
Stigma, however, continues to be a barrier for HIV patients when they access general health
services.
Funding for HIV/AIDS has enabled the government to move national resources away from this area
and increase spending on other healthcare priorities. As a consequence, HIV/AIDS service
providers have become significantly dependant on GHI funding, threatening their sustainability
when the funding ends. In light of ongoing reforms towards complete privatization of primary
and hospital care, it is not clear how HIV/AIDS services that are supported by GHIs will be
integrated in Georgia’s private health care system.
Background
Located in the Caucasus region between Europe and Asia, Georgia is bordered by the Russian
Federation to the north, Azerbaijan to the east, Armenia to the south, and Turkey to the southwest.
Georgia ranked 93 rd out of 179 countries on the UN Human Development Index in 2006 [1]. The
country has experienced rapid growth in real GDP, but the war over South Ossetia (in 2008) caused
significant damage to the economy, and Georgia is now faced with rising poverty in the rural
areas, a lack of employment opportunities, and poor infant and maternal health [2].
There are a number of financial and geographic barriers to accessing health care in Georgia. Prior
to emerging as an independent state amidst the Soviet Union’s collapse in 1991, 4.5% of GDP was
spent on health. The estimated public health spending was around US$ 500 per capita. The fiscal
crisis of the transition in the early 1990s hit the health sector particularly hard, and by 1994,
government expenditure on health declined to around US$ 0.8 per capita (0.3% of GDP) [3]. From
∗
All authors are affiliated with the Curatio International Foundation. The research for this study summary was funded by
The Alliance for Health Policy and Systems Research, based at the World Health Organization. The study summary was
produced with the support of the GHIN Network (www.ghinet.org).
59

1994, private spending and emergency assistance from international donors became a major
source of health sector financing.
Despite structural and systematic changes such as moving from a costly primary care model to a
more affordable family medicine model; and introducing an insurance scheme, Georgia’s financing
for health care services is highly dependent on out-of-pocket payments. As a result of the high
out-of-pocket expenses as well as a scarcity of qualified providers in rural areas, Georgia has
relatively low service utilization rates. A 2002 survey conducted in the Georgian countryside found
that only 59.5% of those who face a health problem seek care, while 15.1% self-treat [4].
To address these problems, the Georgian government launched a Primary Health Care (PHC)
reform initiative in 2002 aimed at improving care in both rural and urban areas [5]. However, the
PHC reform has not received adequate financing. By 2007, most hospitals, dental clinics, and
pharmacies in Georgia had been privatized [6]. In 2007, public funds accounted for only 18% of
total health expenditure, compared to nearly 72% from out-of-pocket sources [7]. External
resources made up 6.7% of total health expenditure in 2006 [8].
Official Development Assistance (ODA) to Georgia was US$ 360.6 million in 2006 [9]. Its total
external debt was US$ 1.96 billion or 25.4% of GDP [9, 10].
Table 1 Basic Socioeconomic, Demographic and Health Indicators (*)
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 4,400 2007 World Bank
Geographic Size (sq. km) 69,700 2007 World Bank
GDP per capita, PPP (constant 2005 international $) 3,365 2007 World Bank
Gini index 40.4 2007 World Bank
Government expenditure on health (% of general
government expenditure)
4.2 2007 WHO NHA
Per capita government expenditure on health at
average exchange rate (current US$)
35 2007 WHO NHA
Physician density (per 10,000) 45 2007 WHO SIS
Nursing and midwifery density (per 10,000) 39 2007 WHO SIS
Maternal mortality ratio (per 100,000 live births) 66 2005 WHO SIS
DTP3 coverage (%) 98 2007 WHO SIS
Estimated adult HIV (15-49) prevalence (%) 0.1 [

Objectives and Methodology
The effects of Global Fund funding on Georgia’s health system were assessed in a two-phase study
conducted during 2004-2008. The study was part of the System Wide Effects of the Fund (SWEF)
Network, which participates in the Global HIV/AIDS Initiatives Network (GHIN), where researchers
are studying the effects of GHIs on countries’ HIV/AIDS programmes and health systems.
The studies were designed to assess the effects of the Global Fund on the policy environment, on
public-private interactions, on human resources, and on access to HIV/AIDS services. The findings
presented draw from a base-line survey implemented in 2004; the survey was carried out in 35
health facilities, in 26 districts, with a follow up study a year later [11, 12]. Data collected was both
quantitative and qualitative, based on structured questionnaires with service providers and semistructured
interviews with 36 key stakeholders. Study protocols were approved by the Bioethics
Committee of Georgia.
For each of the themes identified, the following tools were utilized by the research team, following
SWEF and GHIN protocols [13]:
Policy environment
Stakeholder interviews were conducted. In total, 24 interviews were completed between January
2006 and February 2007, with representatives of the Country Coordinating Mechanism (CCM), the
Ministry of Health (MoLHSA), members of the Parliamentary Committee on Health and Social
Issues, representatives of International and local NGOs, members of the donor community, and
managers at the National AIDS Center, the National Institute of Drug addiction, and with the
Principal Recipient.
Public-private mix
Structured interviews were conducted with managers at 10 NGOs, who are implementing the
Global Fund-financed HIV/AIDS programme in Georgia. These interviews were complemented by
on-site visits to projects and a review of documents.
Human resources
Thirty-five health facilities were selected in three different geographical locations, and 201 primary
health care providers interviewed. Sampling was based on incidence rates of the three target
diseases.
Access to HIV/AIDS and TB services
Exit interviews were conducted with a small sample of TB (n=19) and AIDS (n=20) patients.
Additionally, in-depth interviews were conducted among high risk group representatives (60
injecting drug users [IDUs] and 60 commercial sex workers [CSWs]).
Results
Leadership and Governance
Global Fund resources contributed to the establishment of CCMs, which over time have improved
their overall governance and functionality since the initiation of the Global Fund grant. Interviews
with key individuals revealed that multi-sectoral coordination has benefited from the process,
which respondents attributed to the leadership qualities of the CCM chair. The follow-up survey
revealed that CCM members became much more active and had developed a better
61

understanding of their roles and responsibilities since the time of the baseline survey, as well as a
better knowledge of the health system and ongoing reforms.
Financing
Other effects of Global Fund financing on the health system are mixed. For example, while support
from the Global Fund for TB services is in line with Georgian government priorities to integrate
vertical programmes at the primary care level, this is less true for HIV/AIDS and malaria
interventions, as both are vertical in nature. In light of ongoing reforms towards complete
privatization of primary and hospital care, it is not clear how vertical programmes will be
integrated into Georgia’s private health care system. Many stakeholders perceive Global Fund
monies as reinforcing vertical tendencies.
Table 2 Global Health Initiative Investments (*)
(*) Full data sources for all indicators are provided in Annex 1.
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 2, HIV/AIDS 32,855,709 12,111,223
Round 3, Malaria 806,300 806,300
Round 4, TB 5,536,965 4,245,476
Round 6, HIV/AIDS 6,130,724 2,763,821
Round 6, Malaria 1,587,960 1,587,960
Round 6, TB 9,314,136 9,314,136
TOTAL: 56,231,794 30,828,916
PEPFAR*
Year Amount Allocated (in US$)
2006 1,689,480
2007 1,520,000
2008 961,130
TOTAL: 4,170,610
*Not a PEPFAR focus country; above sums represent total allocations to PEPFAR
country programmes from bilateral U.S. sources including USAID, Department of
Health and Human Services, Department of Labor, and Department of Defense.
GAVI
Disease Priority Amount Approved (in US$)
Hepatitis B vaccine 705,000
Vaccine introduction grant 100,000
Injection Safety 65,600
Immunization services
135,500
support
Health systems strengthening 435,500
TOTAL: 1,441,398
62

Global Fund resources have had dual effects on public financing for health care. On one hand,
these funds have allowed the government to move national, fiscal resources away from HIV/AIDS,
TB and malaria and to increase spending levels on other healthcare priorities. For example, from
2001 to 2006 public expenditure on health grew on average by 23% annually, taking into account
inflation. However, allocations increased only marginally for TB and malaria and declined for
HIV/AIDS. Expectations that the government would increase allocations for target diseases have
not been met in Georgia; hence, service provision for target diseases is significantly dependant on
the Global Fund. This has raised concerns among stakeholders, as it could threaten the
sustainability of service provision after funding ends.
The issue has been further aggravated by increases in service availability, which have helped
preventive, curative and care services to reach more individuals, but have also significantly
increased recurrent cost requirements for HIV/AIDS, TB and malaria. If these diseases are not
controlled, recurrent cost requirements will grow and consequently will aggravate funding
shortages currently observed. Most key stakeholders interviewed expressed concerns regarding
the sustainability of services in the long-term. They were of the view that in the medium to longterm
it is unlikely that the government will be able to fully replace Global Fund monies with
internal fiscal resources. Policymakers at the national level were particularly concerned that the
Global Fund work with the government to develop gradual exit strategies spanning 10-15 years -
strategies which would take into account both changes in the epidemic and in the economic
situation.
In Georgia, Global Fund grants contributed about 2.8% of total health expenditure, which is
relatively low compared with sub-Saharan Africa.
Health Workforce
The study also shows that training funded by the Global Fund grants has helped develop the
capacity of health providers in both the private and public sectors, including addressing issues of
stigma and marginalization. Between the first phase of the study and the second, findings
suggested that health providers’ attitudes towards patients had become more positive: health
providers were less afraid of the target diseases and expressed greater readiness to render the
necessary care and treatment.
Medical Products, Vaccines, and Technologies
The impact of Global Fund financing was significant in generating positive results for individuals
suffering with target diseases: funds helped to supply necessary diagnostic tests and drugs, and as
a result, ART is available to all who require treatment.
Community/Civil Society
However, while patients with target diseases have benefited from free services, largely funded by
the Global Fund, their access to general health services remains limited, and interviewees said they
often face stigma and confidentiality problems when they use general health services.
The study suggests that Global Fund financing also played a significant role in creating social
networks of patients suffering from target diseases. This has facilitated people living with HIV/AIDS
(PLWHA) being able to meet and exchange information, better understand their health and social
problems and to become more open about their status.
63

References
[1] Human Development Reports: Georgia. New York, United Nations Development Programme,
2008 (http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_GEO.html, accessed
19 March 2009).
[2] Georgia – Summary of Joint Needs Assessment. Findings prepared for the donor conference,
Brussels, October 22, 2008. United Nations & World Bank, European Bank for Reconstruction and
Development, European Commission, European Investment Bank, International Finance
Corporation (http://siteresources.worldbank.org/INTGEORGIA/Resources/301645-
1224598099977/summary.pdf, accessed 19 Mar 2009)
[3] A. Telyukov MP, G. Gotsadze, and L. Jugeli. 2003. Situation Analysis for a New Strategy of
Technical Assistance in the Health Care Sector of Georgia. Bethesda, MD: The PHRplus Project, Abt
Associates, Inc.
[4] Gotsadze G, Zoidze A, Vasadze O. Reform strategies in Georgia and their impact on health care
provision in rural areas: evidence from a household survey. Social Science & Medicine. 2005;60:809-
821.
[5] Georgia Primary Health Care Development Project. Washington, DC: The World Bank, 2002
(http://web.worldbank.org/external/projects/main?pagePK=64283627&piPK=73230&theSitePK=4
0941&menuPK=228424&Projectid=P040555, accessed 3 Mar 2009)
[6] Country Cooperation Strategy at a glance: Georgia. Geneva, World Health Organization, 2007
(http://www.who.int/countryfocus/cooperation_strategy/ccsbrief_geo_en.pdf, accessed 19 Mar
2009).
[7] Georgia National Health Account. Geneva, World Health Organization, 2008
(http://www.who.int/nha/country/geo.pdf, accessed on 19 Mar 2009)
[8] WHO Statistical Information System (WHOSIS) [online database]. Geneva, World Bank. 19 Mar
2009.
[9] World Development Indicators (WDI) Online[online database]. Washington, DC: World Bank.
19 Mar 2009.
[10] Georgia at a glance. Washington, DC: The World Bank Group, 2008
(http://devdata.worldbank.org/AAG/geo_aag.pdf. , accessed 19 Mar 2009).
[11] Curatio International Foundation. Effects of GFATM on Georgia's Health System Development
2008 (www.curatiofoundation.org, accessed 19 February 2009).
[12] Curatio International Foundation. Final report of the System Wide Effects of Fund (SWEF)
Study in Georgia (www.curatiofoundation.org, accessed 19 February 2009).
[13] Bennett S, Fairbank A, 2003. The System-Wide Effects of The Global Fund To Fight AIDS,
Tuberculosis and Malaria: A Conceptual Framework. Technical Report No. 031. Bethesda, MD: The
Partners for Health Reformplus Project, Abt Associates Inc.
64

Ghana: Interactions between Health Systems and
Global Fund supported TB and HIV
programmes
Abstract:
Sai Pothapregada, Rifat Atun *
Ghana, a beneficiary of support from the Heavily Indebted Poor Country Initiative, aims to achieve
middle-income country status by 2015, supported by sustained economic growth and an
articulated health policy goal of ‘Creating Wealth Through Health’. The country has benefited from
large inflows of external resources accounting for nearly 22% of total health expenditure in 2006.
The Global Fund contributes to more than 85% of the National AIDS Control Programme budget
and 95% of funding for the Tuberculosis programme. This external support has been universally
acknowledged as critical for rapid scale-up of disease-specific interventions.
The existing mechanism for donor coordination and tracking earmarked funds under the Sector
Wide Strategic Approach (SWAp) facilitated seamless integration of Global Fund programmes into
existing structures, in line with the national health sector plan. The Global Fund projects have
contributed to strengthening service delivery, availability of anti-TB/antiretrovirals (ARVs) and
capacity building training for the workforce. These projects, however, do not often address other
systemic challenges like institutional density, workforce availability, and availability of essential
drugs. Aligning the Country Coordinating Mechanism (CCM) into the national accountability
framework, integrating parallel and often demanding reporting requirements, and exploring
alternatives to the competitive proposal-based funding mechanism to ensure predictable funding,
have all been identified as critical areas for deliberation and action.
Background
The Republic of Ghana is centrally located in West Africa and is bordered on the east by the
Togolese Republic, on the north and northwest by Burkina Faso, and on the west by the Republic
of Côte d’Ivoire. Ghana is home to 23.5 million people, and has experienced an average annual
population growth of 2.2% since 2001[1]. Since 1997, Ghana’s economy has sustained an average
real GDP growth of more than 5%, with a 6.3% growth in 2007 [1]. The government now explicitly
aims to achieve middle-income status by 2015 and is well ahead of schedule in achieving the key
poverty-related Millennium Development Goals (MDGs). However, Ghana has not shown similar
progress on its human development indicators – it ranked 142 out of 179 countries on the UN
Human Development Index in 2006 [2].
Health care is provided by a multitude of players. The Ministry of Health (MOH), which is
represented by the Ghana Health Service and teaching hospitals, owns approximately 49% of total
health facilities. The private sector owns approximately 21%. Christian Health Association of Ghana
(CHAG) institutions own 8% while private maternity homes own approximately 17%. .Distribution
of health facilities and staff favours the affluent regions in south Ghana. It is estimated that more
than 70% of the population rely on traditional medicine, even though this has not yet been
adequately integrated into the formal health sector [3].
65

In 2003, the government launched the National Health Insurance Scheme (NHIS). Under this
scheme, adults could receive unlimited medical treatment for an annual fee of US$ 8, and children
could be seen free of charge if both parents were registered. Progress with the scheme has been
slow. By the end of the first quarter of 2008, however, more than 55% of Ghanaians had bought
into it [4]. NHIS funds are now the major source of health funds, accounting for 31% of the health
budget in 2007, second only to 44% by the Government of Ghana [5]. Private expenditure on
health (63.5% of total health expenditure) is greater than public expenditure on health and has
increased by 4.5% since 2000 [6]. Out-of-pocket expenses make up the bulk of private expenditure.
In 2006, external resources constituted 22.4% of the total expenditure on health [6]
Official development assistance to Ghana was USD 1.18 billion in 2006. The country, however,
remains highly indebted [7]. In 2006, Ghana’s total external debt was USD 3.19 billion or 25.1% of
GDP [1],[7].
Table 1 Basic Socioeconomic, Demographic and Health Indicators (*)
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 23,462 2007 World Bank
Geographic Size (sq. km) 227,540 2007 World Bank
GDP per capita, PPP (constant 2005
international $)
1,260.21 2007 World Bank
Gini index 43.0 2006 World Bank
Government expenditure on health (% of
general government expenditure)
5.5 2007 WHO NHA
Per capita government expenditure on health at
average exchange rate (current US$)
14 2007 WHO NHA
Physician density (per 10,000) 2 2004 WHO SIS
Nursing and midwifery density (per 10,000) 9 2004 WHO SIS
Maternal mortality ratio (per 100,000 live births) 560 2005 WHO SIS
DTP3 coverage (%) 94 2007 WHO SIS
Estimated adult HIV (15-49) prevalence (%) 1.9 (1.7-2.2) 2007 UNAIDS
Estimated antiretroviral therapy coverage (%) 15 (13-19) 2007
Tuberculosis prevalence (per 100,000) 353 2007 WHO GTD
Estimated malaria deaths 25,075 2006 WHO WMR
66
WHO/UNAIDS/UNICEF

Table 2 Global Health Initiative Investments (*)
(*) Full data sources for all indicators are provided in Annex 1.
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 1, HIV/AIDS 14,170,222 14,170,222
Round 1, TB 5,687,055 5,685,493
Round 2, Malaria 98,613,734 8,849,491
Round 4, Malaria 38,887,781 38,887,781
Round 5, HIV/AIDS 97,098,678 49,218,985
Round 5, TB 31,471,784 22,556,665
Round 8, HIV/AIDS 51,498,200 0
Round 8, Malaria 39,639,118 0
TOTAL: 377,066,572 139,368,637
PEPFAR*
Year Amount Allocated (in US$)
2004 7,000,000
2005 7,304,300
2006 7,291,000
2007 6,630,000
2008 7,455,450
TOTAL: 35,680,750
*Not a PEPFAR focus country; above sums represent total allocations to PEPFAR
country programmes from bilateral U.S. sources including USAID, Department of
Health and Human Services, Department of Labor, and Department of Defense.
GAVI (in US$)
Disease Priority Amount Approved (in US$)
Pentavalent vaccine 106,564,000
Yellow fever vaccine 6,745,000
Vaccine introduction grant 100,000
Injection Safety 855,300
Immunization services support 3,676,300
Health systems strengthening 9,670,000
TOTAL: 127,611,488
World Bank MAP
Project Title FY Approved/Closing Date Commitment (in US$)
AIDS Response Project (GARFUND) 2001/2005 25,000,000
Multisectoral HIV/AIDS Program 2006/2011 20,000,000
67

Objectives and Methodology
The data collection for this case study included semi-structured interviews, field observations, and
examination of secondary data sources. Key informant interviews were conducted in a semistructured
fashion using a set of interview questions based on the Systemic Rapid Assessment
toolkit (SYSRA) and adapted to the analysis of integration of health interventions [8],[9]. Our
analysis focused on the HIV/AIDS and TB programmes at the central, regional and district levels.
We conducted a total of 40 individual and group interviews, selected purposively or by
snowballing to include a diverse range of implementers, policymakers, and partners. Two
conveniently selected districts in the neighbouring western region of the capital were visited, and
the regional and district health management staff, programme nodal officers and managers of
health facilities were interviewed.
The relevant portions of the interview were roughly coded to chart data to the health system
function. Secondary data sources (e.g. national strategy papers, annual reports, evaluation reports,
Global Fund proposals) were analysed in a similar thematic way. Primary and secondary data
sources were compared for internal validation of findings.
Results
Leadership and Governance
As part of the wider health sector reform process initiated in the mid 1990s, the government
undertook structural reorganization along the lines of a purchaser-provider split with the MOH as
the purchaser and regulator of service provision. Responsibility for service provision belongs to the
Ghana Health Services (GHS - the main service provider), along with the teaching
hospitals/specialized institutions, quasi government organizations, and the private sector
including non-governmental organizations (NGOs) and the traditional system.
As one of the beneficiaries of the enhanced HIPC Initiative, Ghana has benefited from large inflows
of external resources, accounting for nearly 30% of total health expenditure in 2003 and around
22% in 2006. Policymakers recognized the need to develop strategies for better donor
coordination in the context of an overall health sector policy set by the MOH, while protecting
national systems against the proliferation of parallel structures. To this end, the government
adopted SWAp, setting the stage for the development of a sector strategy that would embrace all
sources of funds and form the basis of planning and resource allocation by both the MOH and
donor partners. This initiated the medium-term strategic framework. The country is currently
implementing its third strategic framework (2007-2011), ‘Creating Wealth Through Health’, with
the goal of making Ghana a middle income country by 2015.
Common management arrangements; systems for partner coordination for the development and
implementation of an annual programme of work; and evaluation of progress against a set of
universally agreed benchmarks or targets, led to a gradual, systematic integration of donor
projects into a sector programme. At the centre of all these reforms was the strategic leadership
offered by the government, notably the MOH.
Against this background, Global Fund projects have been seamlessly integrated into the existing
systems, with Global Fund support viewed as a source of earmarked funds for AIDS/TB/Malaria, in
line with national strategic objectives. On the other hand, the creation of the CCM for grant
proposal preparation and monitoring of grant performance has been less successful, and is viewed
68

y many key informants as a duplication of existing structures for coordination among partners.
The CCM is widely regarded as a body with authority (for shaping and approval of country
proposals for Global Fund funding) but with no clearly identifiable accountability structures.
Financing
Global Fund support for HIV/AIDS accounted for 40% of total external support for the entire
HIV/AIDS sector in 2007 (budgetary estimates) and 57% in 2008 (budgetary estimates). In 2007 the
Global Fund supported more than 85% of the National AIDS Control Programme’s (NACP)
expenditure. Global Fund contributions towards the National TB Control Programme accounted
for more than 95% of its budget in 2006 (excluding salaries and Government of Ghana (GoG)
expenditure on the general health system).
The health sector has seen improvements in overall funding, primarily due to increments in the
Internally Generated Funds (IGF) from the NHIS, accounting for more than 30% of total budget in
2006. The donor pool fund has fallen from 12.7% in 2005 to 3.4% in 2007, as budget support was
brought into line with the Paris Declaration and the Ghana Harmonization and Alignment Plan.
This policy change has raised concerns that health care has to compete with other government
priorities. There is no evidence, however, to back up the argument that Global Fund support led to
a decline in GoG funding or that of other partners. As noted by a high-level official, in the absence
of Global Fund support it would not have been feasible to scale-up HIV/AIDS interventions, or to
initiate either public-private mix (PPM) schemes or the Enablers package for TB control.
With reference to institutional arrangements for HIV/AIDS financing, the Ghana AIDS Commission
(GAC) was created in 2001 as the supra-ministerial body responsible for coordinating a broad
based, multi-sectoral approach for HIV/AIDS in the country (National Strategic Framework II: 2006-
2010). The GAC oversees and coordinates funding for all HIV/AIDS interventions, including the
health sector response. Global Fund support is captured as direct funding to the implementing
agency (the MOH/GHS), as against the other pooled and earmarked funds received by the GAC. In
order to minimize duplication of activities and to optimize the use of scarce resources, several key
informants expressed a need for improved coordination between GAC and other implementing
partners, including NACP.
In contrast to the TB programme, where funds are routed through regional/district accounts, the
HIV/AIDS programme division manages most of the funds (80% of which go towards drugs,
commodities, and infrastructure/equipment for HIV/AIDS), with sub-recipients (including
NGOs/Faith-based organizations) accounting for 7% of total expenditure. This results in limited
flexibility at the regional and district level to plan or implement interventions or for M&E.
As mentioned above, the health system has structures in place for capturing expenditures by
source and by activity, including for Global Fund grants. However, the demand for quarterly
reporting and reporting by Service Delivery Areas (SDAs) places an added burden on the system.
This is further complicated by frequent changes to the ways in which activities are grouped into
SDAs and the nomenclature of the SDAs.
With the Global Fund as a major source of health sector funding, another crucial challenge
concerns the sustainability of activities in the event of an interruption of grants or the failure of
future proposal submissions. Such disruptions can severely jeopardize the national planning
process.
69

Health Workforce
Ghana is challenged by inadequate skill mix, insufficient numbers and improper distribution of its
health workforce. To improve workforce retention, distribution and motivation the government
has initiated several measures. These include additional duty allowance, consolidation of salary
structures, improved training opportunities and other fellowship programmes, as well as loans
towards the purchase of cars or for housing schemes. These measures have led to an increase in
payroll costs, which account for nearly 90% of the GoG health budget, and 47% of total health
budget.
Except for dedicated data managers at regional levels for the HIV/AIDS programme, and technical
officers at the national level, no additional human resources have been recruited under the
projects (accounting for less than 5% of the project costs). Global Fund support has contributed to
programme-related training of the workforce, fellowship trainings and participation in
international workshops/seminars, which improve provider confidence and quality of care.
However, combined with the training burden of national programmes, such activity by GHIs has
tended to overload the system: the NACP annual report observes that scale up of antiretroviral
therapy (ART) services was delayed by the non-availability of staff engaged in ongoing training
programmes.
Medical Products, Vaccines, and Technologies
All informants agreed that Global Fund projects have ensured availability of high quality drugs at
competitive prices through mechanisms like the Global Drug Facility (GDF) and the Green Light
Committee (GLC). The HIV/AIDS programme has developed a logistic management information
system and trained over 265 pharmacists and dispensing technologists on logistic management of
antiretrovirals (ARVs). ARVs are distributed to the ART sites based on their rates of consumption.
Both programmes use the national medical stores and logistic supply chain for the distribution of
drugs, and they are available through pharmacies within the health institutions. The (national)
Public Health Laboratory coordinates the supply of laboratory supplies and consumables.
Although there is clearly a need for additional funds to further strengthen warehouses and
transport, these activities have not been budgeted under the grants, indicating greater ownership
by the national health system.
Information
The demand for health information to monitor performance-based funding (input-process-outputoutcome
and impact) is known to exert great strain on project staff and fragile health systems in
most low-income countries, and Global Fund-supported projects are no exception.
Although the Global Fund project has supported M&E staff at key levels, the burden of reporting
and the emphasis on completeness and timeliness has created vertical programme reporting.
Ghana has invested appreciably in the national Health Management Information System (HMIS),
which is electronic at district level and upwards, but the programmes are reluctant to integrate
them, more out of lack of trust rather than technical obstacles. Integration is made difficult by the
demand for information on innumerable indicators, and also the frequent changes in reporting
formats.
One key informant on the staff of a development partner suggested ironically that it would be
interesting to explore whether the Global Fund or the UN agencies, including WHO, place more
70

urden on public health personnel with their frequent demands for information to fill in reports
and analytic publications.
Service Delivery
The TB and HIV/AIDS programmes have consciously used existing service delivery systems for
rolling out interventions; and, have therefore, avoided creating parallel structures. To coordinate
and monitor programme interventions, existing staff have been designated as nodal officers, but
continue to contribute to other public health interventions like leprosy control, guinea worm
control and promotion of family planning. The expansion of voluntary counseling and testing
(VCT) services to sub-district level, and ART facilities to district level and some polyclinics, has
definitely led to improved collaboration between the two programmes as well as improved client
convenience.
Global Fund projects have contributed significantly to strengthening service delivery through
refurbishment of health facilities; provision of equipment (microscopy/auto-analyzers); ensuring
uninterrupted supply of high quality TB drugs and improved access to ART; and vehicles for
monitoring and evaluation, which are used by the system at large.
Though these contributions are important, critics often complain that they only support
programmes and do not often lead to overall improvements in other areas of service delivery –
such as institutional density, health workforce and availability of essential drugs – that constitute
major challenges for many health systems. In spite of GHI support, private out-of-pocket
expenditures continue to be high and are rising as a percentage of overall health sector
expenditures. Comparative coverage indicators for non-GHI-supported initiatives like maternal
and child health – these include antenatal care (ANC) coverage, family planning acceptance, and
deliveries by skilled attendants - have failed to show comparable improvements during the same
time frame.
Community/Civil Society
Civil society, including FBOs, have been partners with the public health system for delivery of
health care to such an extent that a portion of the Christian Health Association of Ghana’s (CHAG)
staff salaries are borne by the government. The Global Fund projects have facilitated the extension
of these formal arrangements with individual private providers and private associations.
However, a perception also exists that, with large sums of money being made available there has
been an unprecedented surge in the number of NGOs, and identifying credible ones has become a
challenge. Inputs have been sub-optimal for strengthening local NGOs and community-based
organizations (CBOs), and the bulk of resources are channelled through international or large
NGOs.
71

Discussion
GHIs are critical for rapid scale-up of services, which would not otherwise be feasible with scarce
in-country resources. To achieve rapid scale-up and impressive results, the programmes may have
to work independently/vertically during the initial years. But unlike disease eradication
programmes, TB, HIV and malaria are long-term challenges. As countries work to control these
epidemics, they need to avoid creating parallel structures that are unsustainable in the absence of
external funding.
Ghana is an ideal example of a country with an existing system for improved donor coordination
and with scope for an integrated approach to address systemic challenges in the delivery of
disease-specific interventions. To maximize the benefits and limit negative spillover effects, the
Global Fund needs to be more actively engaged at the country level, moving away from a model
that involves repeated rounds of competitive proposal submission and project approval to a more
predictable and sustainable funding mechanism.
72

References
* Imperial College London/Global Fund to Fight AIDS, TB and Malaria
[1] The World Bank Group. Ghana at a glance. (http://devdata.worldbank.org/AAG/gha_aag.pdf).
[2] United Nations Development Programme. Human Development Report: Ghana.
(http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_GHA.html).
[3] Government of Ghana. Scaling-up health investments for better health, economic growth and
accelerated poverty reduction: Ghana Macroeconomics and Health Initiative: National
Development Planning Commission; 2005.
[4] National Health Insurance Authority. National Health Insurance Scheme.
(http://www.nhis.gov.gh/).
[5] Ministry of Health Ghana. Review of Ghana health sector programme of work, 2008.
[6] World Health Organization. WHO Statistical Information System (WHOSIS).
(http://www.who.int/whosis/en/).
[7] The World Bank Group. World Development Indicators (WDI) Online. (http://ddpext.worldbank.org.ezpprod1.hul.harvard.edu/ext/DDPQQ/member.do?method=getMembers).
[8] Atun, R., et al. (2004). A framework and toolkit for capturing the communicable disease
programmes within health systems: tuberculosis control as an illustrative example. Eur J Public
Health 14(3): 267-73.
[9] Atun, R., et al. (2008). When do vertical (stand-alone) programmes have a place in health
systems? Copenhagen, World Health Organization Regional Office for Europe.
73

Haiti: Maximizing Positive Synergies between
Global Health Initiatives and the Health
System
Abstract
Background
LC Ivers 11 12 13, JG Jerome 10, E Sullivan 14 15, JR Talbot 11 14, N Dhavan 11 14, M StLouis, W Lambert
10, J Rhatigan 11 12 14, JS Mukherjee 10 11 12.
Increasing attention is being paid to the impact that GHIs such as the Global Fund and PEPFAR
have on health systems as a whole. We aimed to evaluate the impact of these two GHIs on the
health system in Haiti.
Methods
We used mixed qualitative and quantitative methods: semi-structured interviews, observation,
examination of documentary material, and collection of quantitative data. Twelve key informants
from nongovernmental organizations (NGOs), international organizations and the Ministry of
Health (MOH) participated in semi-structured interviews.
Results
The Global Fund /PEPFAR were largely associated with strengthening the health system in Haiti,
particularly in geographic areas and programmes where NGO leaders and/or MOH specifically
designed and coordinated their interventions to have such an impact. The Global Fund/PEPFAR
were associated with overall improvement in human resources capacity and retention but did
cause some tensions due to salary inequities. The Global Fund/PEPFAR funds were not received
directly by the government of Haiti, and this empowered NGOs to work independently of the
MOH.
Discussion
GHIs such as the Global Fund and PEPFAR can successfully strengthen health systems if this
objective is included in planning and design and if leadership and coordination are ensured.
11 Partners In Health, Boston, USA
12 Brigham and Women’s Hospital, Boston, USA
13 Harvard Medical School, Boston, USA
14 Harvard School of Public Health, FXB Center for Health and Human Rights
15 Global Health Delivery Project, Boston, USA
74

Background
The Republic of Haiti occupies the western third of the island of Hispaniola, sharing its 360
kilometre eastern border with the Dominican Republic. Haiti is home to more than 9.6 million
people (see Table 1), up from just over three million in 1955 [1]. As of 2007, approximately 1% of
the population controlled nearly half of Haiti’s wealth [2]. In 2005, Haiti ranked 146 th out of 177
countries on the UN Human Development Index [3].
Official Development Assistance (ODA) to Haiti in 2006 was US$ 581.4 million [1], but the country
remains highly indebted due largely to loans incurred from 1964 to 1986 [4]. At the end of 2007,
Haiti’s total external debt was US$ 1.463 billion or 24% of gross domestic product (GDP) in 2006
[5].
Widespread deforestation, soil destruction, and poor agricultural output contribute to food
insecurity; the island’s vulnerability to natural disasters, including floods and hurricanes,
exacerbate the situation. A 2004 hurricane killed approximately 3000 people and caused damage
estimated at 3.5% of GDP [6], while a 2008 series of hurricanes killed over 800 people, displaced 1
000 000, and wiped out 60% of the year’s harvest [7].
Haiti has some of the worst health statistics in the Americas (see Table 1), including the highest
infant mortality and maternal mortality rates in the Western hemisphere (60.0 per 1000 in 2006
and 670 per 100 000 live births in 2005 respectively). WHO estimated that incidence of smearpositive
pulmonary TB in Haiti was 133 per 100,000 in 2006; prevalence was 402 per 100 000. First
identified in Haiti in 1981, HIV/AIDS became the leading cause of death in 1999. In 2007, 120 000
people in Haiti were living with HIV [8].
In 2003, health services reached only 60% of the population. In addition to the public sector, which
comprises about 35.7% of the health infrastructure, the health system includes for-profit service
providers catering to urban professionals; private non-profit organizations; and mixed non-profit
facilities, including Ministry of Health (MOH) personnel working in private institutions or religious
organizations [9]. About 40% of the population—mostly in rural areas—relies on traditional
medicine, lacking access to other services [9]. Physicians are concentrated in the capital city of
Port-au-Prince and in other large towns. In some provinces, there is one physician for every 67 000
people [10]. Nearly 30% of health professionals left the country between 2005-2008 for the U.S. or
Canada [11]. The government has regularly increased the health budget, but the recent funds are
lower in real terms than in the 1980s due to inflation [12]. Currently, 18.9% of the country’s total
health budget comes from external sources [11,13].
Haiti was among the first set of grant recipients from the Global Fund in 2003, and since then, the
country has received two additional grants for HIV/AIDS and one for TB, with a total approved
funding amount of approximately US$ 248 million, though only US$ 130.1 million has been
disbursed (see Table 2). The U.S. government began funding HIV/AIDS prevention, treatment, and
care interventions in Haiti through PEPFAR in 2004, working largely with existing USAID contracts
and partners, half of which were faith-based organizations (FBOs) [14].
75

Table 1 Basic Socioeconomic, Demographic and Health Indicators*
* Full data sources for all indicators are provided in Annex 1
Indicator Value Year Source
Population (thousands) 9,612 2007 World Bank
Geographic Size (sq. km) 27,560 2007 World Bank
GDP per capita, PPP
(constant 2005 international $)
1,090.37 2007 World Bank
Gini index 59.21 2001 World Bank
Government expenditure on health (% general
government expenditure)
27.7
2007 WHO NHA
Per capita government expenditure on health
(current US$)
38 2007 WHO NHA
Physician density (per 10,000) 3 1998 WHO SIS
Nursing and midwifery density (per 10,000) 1 1998 WHO SIS
Maternal mortality ratio (per 100,000 live
births)
670 2005 WHO SIS
DTP3 coverage (%) 53 2007 WHO SIS
Estimated adult HIV (15-49) prevalence 2.2 (1.9-2.5) 2007 UNAIDS
Estimated antiretroviral therapy coverage (%) 41 (33-51) 2007 WHO/UNAIDS/UNICEF
Tuberculosis prevalence (per 100,000) 366 2007 WHO GTD
Estimated malaria deaths 741 2006 WHO WMR
Table 2 Global Health Initiative Investments*
*Full data sources for all indicators are provided in Annex 1
Global Fund
Round & Disease Priority Approved Disbursed
Round 1, HIV/AIDS 159,733,983 83,684,238
Round 3, Malaria 14,431,577 12,631,744
Round 3, TB 14,034,665 13,201,730
Round 5, HIV/AIDS 18,821,754 18,821,754
Round 7, HIV/AIDS 6,199,554 2,450,766
Round 8, Malaria 33,402,457 0
TOTAL: 246,623,970 130,790,232
PEPFAR
Year Amount Disbursed
2004 28,039,418
2005 51,785,021
2006 55,606,667
2007 84,689,732
2008 100,646,286
76

Methodology
Data collection included semi-structured interviews, observation, examination of documentary
material, and collection of quantitative data. Key informant interviews addressed key elements of
the health system: governance, financing, health workforce, monitoring and evaluation (M&E),
health technologies, and communities and civil society. The sampling strategy for key informants
targeted a diverse range of implementers, policymakers and health leaders, using purposive and
snowball sampling. Most of the 12 resulting key informants had well-defined roles in the country’s
health system and included people working in the MOH at national, district and hospital level (size
= 6), a range of NGOs, which was geographically diverse and included both Haitian-based and
international NGOs (size = 5) as well as one informant from the local Haiti WHO office.
Interviews were conducted in English, French or Haitian Creole using the native language of the
interviewee where possible. They were recorded, then translated into English and transcribed.
Transcribed interviews were entered into NVivo8, a qualitative data analysis programme (QSR
International, Cambridge, MA). A thematic qualitative analysis approach and iteratively developed
set of codes were used to examine the data. To further support the authenticity of findings and
auditability of analytic processes, we engaged in inter-rater reliability activities as we created and
applied codes; wrote memos about our analytic decision-making; and conducted participant
validation exercises. For the purposes of triangulating key informant interviews, we collected and
reviewed publicly available documents (i.e. National Health Strategy, PEPFAR or Global Fund
Country Reports).
Quantitative data was collected from seven health centres in urban and rural Haiti. Convenience
sampling was used to select facilities from departments that interviewees represented and to
select facilities with different degrees of GHI funding and different models of care. Quantitative
data collection is ongoing in three facilities at the time of this report.
Results
Leadership and Governance
Many feel that GHI funding has taken away the MOH’s autonomy and control in providing health
care. The MOH first relinquished some control to other institutions during the first round of Global
Fund funding because monies were not awarded to the MOH as a Principal Recipient. Since then,
the trend for funding to go to the private sector has continued. One district director commented:
Most of the funds are not allocated to the state organizations; they are allocated
to private organizations or to non-profit – and the means that are available to us
to ensure an effective control, an effective supervision, an effective monitoring –
we don’t have those means, and as a consequence, we cannot play our role of
coordination.
It is often hard for the MOH to intervene in GHI-funded projects when they are already funded,
especially when the projects have more financing than the MOH itself. The funded NGOs often do
not want to be managed by the MOH, given that they are already reporting to the GHI. Without
getting reports from the NGOs, the MOH is unable to assess the impact of interventions or keep up
77

with need. With GHI funding, district level coordination has been easier to maintain than national
level control, but there has been a lack of experienced or visionary leadership in many districts:
When there is absence of leadership in a specific field, there is always someone
that takes the leadership. That’s what we see in Haiti: the leadership in the fight
of HIV/AIDS is out of the Ministry, is out of the public sector. It belongs to the
NGOs - and that is the funders - and this is not the ideal situation.
Other data suggest that the GHIs have, in fact, led to stronger partnerships between NGOs and the
MOH in some instances. NGOs that were previously strong and working with the MOH are even
stronger now and continue to work with and support them. Some suggest that the MOH has
achieved something of a coordination role in certain departments and is better able to help NGOs
make a greater impact, despite the need for further efforts.
The study showed that GHIs have enhanced leadership capacities in the country in various ways,
despite needing further support. Organizations dealing with salary inequalities among workers
were forced to learn new means of human resource and budget management. The culture of
programme management that has developed with GHI funding is highly regarded and valued in
the public and private sectors. In part, organizations know that they must efficiently manage
finances as well as reporting mechanisms in order to continue receiving funds.
On the negative side, data suggested that GHIs and some NGOs neglect national or local goals and
priorities in favour of their own. One informant from an NGO explained, “The fact that we have
funds that are three times greater than the MOH funds that go to HIV/AIDS—that introduces
distortions, especially if those funds are not associated with a vision to intervene on systemic
elements.” The GHIs do not necessarily take state goals or priorities into account when planning.
PEPFAR came “from the White House” with its own vision, without local field experience; it “had a
rather obtuse vision, which was the vision of the CDC that knew they know better than anyone; the
field experience was not their concern.”
Informants reported that some NGOs entered health departments in the same way, with their
own vision, reporting to the health directors what they were doing after being awarded the
financing, rather than discussing plans in advance.
GHIs may also force recipient NGOs to choose priority areas and hinder NGOs’ flexibility. The
decision to get either ABC (Abstinence, Be Faithful, Condoms) or ABY (Abstinence, be Faithful for
Youth) funding from PEPFAR, for example, can divide organizations and refocus their goals:
There are definitely different camps, obviously, or philosophies to the approach
of how to address the issues …. But even for us… our general philosophy … is
not to exclude condoms– but we took on funding from Track One, ABY, which
excluded us from handing out condoms or including them. We were able to
promote via references to where you can get condoms, but – so we were
obviously forced to come up with ways that probably via our technical advisors
were against their general operating principles, or simply what they thought was
probably the most effective approach.
On the positive side, some data show GHIs have, in fact, come to support national goals. While
PEPFAR looked very vertical at the beginning, it has changed. The Country Coordinating
78

Mechanism (CCM) also has helped unify the goals of the state and the Global Fund in the eyes of
some. In addition, some of the GHI-funded NGOs have used funds to benefit the system and
implement the visions held by local communities. The initial GHI emphasis on getting people on
treatment has been muted somewhat by the new understanding that primary care and other
community needs are important too. The “trickle-down effect” has not impacted the indicators yet,
but many informants believed that a shift is underway. The Global Fund presented an opportunity
to apply for health system strengthening in Round 8.
Financing
One informant explained, “Those two programmes together [PEPFAR and the Global Fund]
account for five times the budget of the MOH. There isn’t any rationale between the burden of
diseases, the very serious public health problems linked to HIV, and the funds invested in them.”
Organizations often feel compelled to spend the GHI-awarded money as specified rather than in
ways that align with local or national health needs.
Another informant explained that, despite the focus on target diseases, it is possible for
organizations to use the GHI funds where they are most needed:
There’s been so much of a focus on HIV that people aren’t looking at the overall
epidemiologic problems in Haiti in terms of the other diseases that are killing
people. And I think for us, nobody ever says, ‘There’s going to be a fund for acute
respiratory infection or a fund for childhood diarrhoea,’ although, I think the PEPFAR
money and the Global Fund money, if you’re aware of the fact that it doesn’t cover
those things, and you’d like to work on what’s going to be the best thing for the
public’s health, then I think then you can use that money to address those other
issues.
Others also have become aware of the GHIs’ limits and learned ways to use the funding to address
issues they see as important to public health by making a conscious effort to do so rather than by
obtaining specific GHI approval.
The funding process itself impacts recipients both positively and negatively. It leaves recipients
with some uncertainty around the rules of how the money can be spent and the timing and
continuity of funding. Among secondary beneficiaries of PEPFAR who rely on primary beneficiary
interpretations, the rules are often unclear or change:
I think the hard part for us is because we’ve been a secondary grantee, we don’t
always know what PEPFAR is deciding to do. I mean, we do our own background
research and try and figure it out. But it’s very easy for the programme managers to
hide behind some sort of a ‘The government’s making us do this this year.’ …
There’s a lot of – people are just reading the rules how they would like to.
The Global Fund funding process was also unclear, especially to those not receiving funding and
without connections to the CCM:
79

I know that within [the organization] there’re a couple of us who understand, or
somewhat understand, the process involved, but if you take other people who’ve
been working with the foundation and would like to get involved in the Global
Fund, they wouldn’t even know – it hasn’t been advertised well enough for them to
understand what channels to enter.
Respondents also suggested the Global Fund has a tendency to disburse funds late, which leaves
some organizations needing to rearrange their budgets to create “bridge funding.”
Although some appreciate the CCM for its ability to bring people together at the same table,
neither the MOH nor any other national entity must approve all GHI-funded proposals, leading to a
lack of cohesion and coordination among them.
There is also general uncertainty about the continuity of both Global Fund and PEPFAR funding. If
GHI funding ends, many programmes will be forced to end, leading to what one informant called
an “obsessive fear”.
The funding process, on the positive side, may make organizations more efficient, accountable, or
help to reinforce administrative and financial capacities:
The institutions [getting PEPFAR /Global Fund funding]… know that if they are not
performing, they will not have financing. They make lots of efforts. There you see a
change of behaviour in the management of funds….This contributes to a better
management.
Human Resources/Health Workforce
Both qualitative and quantitative data showed that GHI funding has increased the staffing levels in
areas previously understaffed as well as in areas that were previously not covered at all by health
personnel. There are significant increases in the absolute number of community health workers,
pharmacists, doctors, nurses, laboratory technicians, obstetricians and counsellors, and this
included both those specialized in HIV care, as well as generalists. More often these positions have
been financed through PEPFAR rather than Global Fund. One informant suggested that some
positions have not made as much of an impact as they might have, because stipulations required
funding only PEPFAR staff (as opposed to MOH staff).
GHI funding contributed significantly to workforce training for community health workers, agents
de santé, peer educators, departmental leaders, doctors, nurses, social workers and laboratory
technicians. While HIV/AIDS was often the focus of trainings, trainings also taught health care
workers about supervision and managerial capacity, how to use computers, management of
multidrug resistant tuberculosis (MDR-TB), sexually transmitted infections, family planning and the
treatment of other diseases Some trainings have allowed nurses to gain skills in areas previously
reserved for physicians, permitting planned “task-shifting,” a transfer of tasks from higher level
cadres to less trained cadres in healthcare. While the health workforce was described previously as
being unmotivated, given poor working conditions as well as limited tools and resources, trainings
brought new ideas, information, and knowledge that reinvigorated some and improved the
general attitude among health workers.
One informant suggested that while significant investments in training have been made, the
investments are not coordinated, so the same people may attend the same trainings multiple
80

times, and there is not always a clear goal to the trainings. At the national level, there is no
coordination of the workforce or human resource management to mitigate this type of problem.
GHI funding has also contributed to new means for recruiting and retention. One health
department director working closely with an NGO explained that GHI-funded construction of
residences in his district allowed staff to stay comfortably and have access to the Internet and
other amenities, thus improving retention. Doctors were willing to work in areas that previously
did not appeal. Other incentives made possible by GHI money included professional development
activities and funding for international congresses provided by the Global Fund. Higher salaries
associated with GHIs, specifically PEPFAR-funded programmes, and improved working conditions,
helped staff feel more valued, leading to improved retention. Community health workers who had
previously volunteered were given stipends, funded in part by GHI money. GHI monitoring and
evaluation requirements have helped with motivation:
The productivity of our staff seems to be stronger and people seem to be more
motivated with goals and indicators…. When they can see that they’re reaching
indicators or reaching goals, they feel that much more motivated about the job that
they’re doing.
The benefits that GHIs have provided for some have caused problems for others. While the
funding has helped keep healthcare workers in the country, many do not stay within the MOH
system where, when not integrated with GHI-funded activities, salary and working conditions have
not improved equivalently. Other health workers have migrated out of non-target disease
programmes into HIV/AIDS. This type of internal drain brain has “sucked the best minds out of
primary healthcare, women’s health, whatever, paediatrics – you have surgeons, for heaven’s
sakes, running PEPFAR programmes, and they’re not doing the operations; they’re running PEPFAR
programmes.” Within institutions that have multiple programmes, only some of which are funded
by GHIs, inequalities among the staff have also caused divisions or friction. As one informant
explained:
The staff members who don’t get the stipend are more inclined to leave the work
than those who get the stipend. What we have tried to make them understand - but
we cannot convince them 100% - is that, for example, AIDS is a disease like any
other disease, and that even if those funds didn’t come, the system would have to
manage it like any other disease….
Where support from PEPFAR or other GHIs is being received, workers not being paid by these
funders are often even more discouraged than they had been prior to the funding influx, because
of the inequity, although those receiving the stipend are not considered to be “overpaid.”
On the other hand, GHI funding is seen as having improved health workers’ capacities in the public
sector, including the capacity to manage:
I started as a pure clinician, nothing else, but with PEPFAR… I developed managerial
capacity.… And I am not the only one in this situation; many of my colleagues, they
all say it, doctors, nurses, psychologists, social workers.... I think the programme of
PEPFAR, and Global Fund, as well, have added to the system, in terms of capacity
building.
81

Much of the initial enthusiasm about PEPFAR funding was due to its ability to increase salaries, not
its ability to help patients, one informant explained. People working in the funded disease areas
appeared to feel that they were being paid by PEPFAR for AIDS work, no longer working for the
organization as a whole, and there were divisions among programmes within the organization.
While one informant suggested that the higher salaries paid to healthcare workers did not
contribute to improved care or result in other constructive changes, many of the human resource
strategies have led to greater access to services. GHI-funded staff hired for targeted disease
activities often reinforced institutions by providing non-targeted services, as one informant
reported: “This same staff engaged in the context of the project don’t work only on HIV/AIDS, but
they are more or less reinforcement for the institutions.” Informants displayed some concerns that
the workforce will return to its pre-GHI funding status when GHI funding dries up, and the state
once again becomes responsible for paying salaries. Informants were also concerned that doctors
will continue moving away from non-GHI funded fields unless the funding structure is changed to
equalize the fields.
Medical Vaccines, Products and Technologies
GHI funding has been used to a large extent to build, restore and renovate health system
infrastructure, especially the spaces used for antiretroviral therapy (ART) clinics. It has also
increased laboratory and diagnostic capacity, including general diagnostic capacities for TB and
MDR-TB, as well as infection control equipment. Technical capacity has increased with new
operating rooms as well as functioning Internet, made possible with generators and other
enhancements. GHI funding has improved electricity and water supply in health centres. Data
showed that PEPFAR has had a more significant response than the Global Fund in terms of
infrastructural enhancements, as well as in laboratory equipment, and funding for technologies
and material support.
While most of the changes in technology have been positive, one organization faced the challenge
of having to incorporate PEPFAR’s required equipment that they found to be inefficient.
Laboratory and diagnostic capacities created for target diseases have expanded non-target
disease diagnostic capacities. Since the basic biochemical equipment is now in place, even in some
of the most remote areas, it costs little more to offer additional tests in those settings, where
health care was once considered inaccessible. Even so, laboratory services in many places remain
prohibitively expensive for some patients, as services are provided free of charge only to those
with HIV.
Information
GHI requirements for monitoring and evaluation (M&E) have encouraged better data collection
techniques, improved information systems and the use of data for both targeted and non-targeted
diseases. These gains stem from improvements to information systems themselves and from the
additional personnel trained and hired to work with these systems. Organizations, including the
MOH, have been forced to become “more professional” about M&E. The MOH put a new
monitoring team in place to help reinforce data collection and analysis activities. The MESI
[National information system] is financed by PEPFAR. As one informant explained:
82

It is a very good system that allows having information. This is a revolution in
Haiti. One had to wait three, four months to know what happened the previous
month; now the data is available right away….Now for the [public health] system
we are thinking of a way to put these same systems in place, that allows having
information rapidly accessible to authorized personnel, in reasonable time.
GHI reporting both enhances and hinders funded programmes. Having to report and meet specific
goals enhances some workers’ productivity. But, reporting can be time-consuming and tedious, as
well; sometimes requests for reports come at the last-minute. One informant suggested that
current reporting models also allow organizations to set unrealistic goals, obtain funding, and
make fictitious claims about their achievements. GHI reporting requirements have also led to
duplicate or parallel system development in some instances.
Service Delivery
Since GHIs have begun operating in Haiti, more people have been getting tested for both TB and
HIV. They come seeking care due to the infrastructural improvements and because they know that
they will “get service.” Not only are TB and HIV patients coming, but “where you went from seeing
16 to 20 people, you are seeing 300 people per day, all categories, pregnant women coming for
care, small children, whether for vaccination … [or other services]. Now everybody comes.”
These comments were confirmed by quantitative data from health centres studied. Reduced costs
and the availability of services have also increased care-seeking behaviour. GHI-funded
programmes that offer treatment have brought hope to those with disease and reduced the level
of fear. “People with HIV don’t feel like they have a death sentence anymore,” one informant
reported. Some organizations are paying attention to stigma and working specifically to reduce it:
It is true that there is still stigmatization, but the fact that people know that they
can get free access to care, well, people who are positive come to get care,
without problems. One feels that the tendency to stigmatize and discriminate
toward people who have the virus has diminished. It is not the same thing
anymore.
There are many more sites offering HIV care because of GHI funds, including 48 sites for ART
(compared to two sites prior to GHI support). GHIs have reduced or eliminated the costs of HIV
testing and care for many patients. More people are getting tested and treated for TB, but there is
still not a national TB supply chain. GHI funding for infrastructural enhancements has increased
access to health facilities for patients with non-target diseases, as well. “We see that at those sites
the frequentation has multiplied by even ten,” one informant explained. “This is phenomenal in
terms of… accessibility, of availability of care,” another added. Vaccination coverage in one area
went from insignificant levels to 100%. The availability of medication for non-targeted diseases has
also increased; although in many programmes these services are available on a fee-for-service
basis to patients who do not have HIV, some programmes provide all medications and services
free of charge.
In terms of integration, informants felt that many GHI-funded programmes were not integrated
into the health system, despite the benefits that integration might offer. This may be due, in part,
to the fact that indicators are HIV-based, not requiring organizations to be accountable for other
83

services, or for measures of quality of care and service delivery that go beyond getting patients on
HIV treatment. As one informant put it, PEPFAR pays attention primarily to “how many people are
on treatment, how many people are on treatment, how many people are on treatment.”
There is no grant stipulation that encourages integration. The GHIs do not coordinate investments
among themselves, and there is no single overarching authority that pushes organizations to be
integrated. The state is in a weak position to exercise coordination, given its limited human and
financial resources, and state officials understandably find it challenging to regulate organizations
with more money than the government itself has. Many organizations receive funding without
informing the MOH. Others do provide information to the MOH, but without paying attention to
the department plan or using feedback to integrate the programme. This can cause lack of
coordination in service delivery. The NGOs’ relative financial strength can disempower the state.
One health department director recounted what he had to do to ensure that an NGO in his district
complied with the district health plan:
[The organization] was going to distribute bed nets. …. There was a discrepancy
between the areas where they were distributing them and the zones that had a
high prevalence of malaria and of filariasis. With the programme managers in the
district, we had decided the zones that should be covered, but they said “no,” they
cannot do that in those zones …. So I told them, “Sorry, madam, go back to Port-au-
Prince with your bed nets.” And the activities of that year didn’t happen. But, this
year is different because they took into account the zones that we had selected for
the distribution of the bed nets, and that’s where the bed nets were distributed.
Data showed that the leadership of the district could have a substantial impact on how NGOs were
integrated into service delivery in the area.
Different health departments have made various efforts to coordinate programmes, despite the
obstacles. One district has set up “District Cooperation Committees” through which the District
Health Director tries to coordinate and monitor the activities in the district. He explained the role
of the Cooperation Committees and the challenges he continues to face:
Through this District Cooperation Committee, we coordinate all the activities of the
stakeholders, but there is a missing element: there is information that the directors
should have had – and sometimes it’s after the partners got the approval from the
donors that we are asked to intervene. But if from the beginning, once the partner
would have presented its proposal, they [the donors] had required that it get the
approval of the district management, we could have started the control. There are
interventions that we would not have authorized, because we would have seen that
they don’t fit to the framework of the district. They don’t fit our objectives, our
vision in the department…. Secondly, the information about the means that are
available to the stakeholders inside of the sector or the department are sometimes
not available to us. In consequence, we cannot measure if the interventions
correspond to the amount that was available for this intervention, to compare the
impact with the investment that was made.
Some informants believe that PEPFAR and the Global Fund have been well managed in the field
and that there has not been duplication. They refer to one particular community-based NGO as an
example of how organizations can and should work with the MOH to ensure integration. Review of
facility data demonstrated that this NGO model showed greater improvements in both targeted
84

and non-targeted disease measures, when compared to other models of care. Qualitative data
demonstrated that the model was specifically designed to be comprehensive and to work within
the public sector.
Informants reported there had been recent shifts towards integration. One informant commented
that the Global Fund has helped reinforce the MOH in terms of coordination in the last few years
and has begun working more closely with other funders in the last year to create synergies with
them. “PEPFAR…came with its own indicators, its own vision that is not integrated…. It came with
its own employees…. That is how it was…. It was vertical, in a straight line …. Now, little by little, it
has changed.” Data suggested that both the Global Fund and PEPFAR are now more aware of the
need to deliver more comprehensive care, rather than vertical programmes.
GHI funding has improved access to and quality of primary care and non-target diseases in some
cases. GHI-funded programmes have made more health information available. Organizations that
have hired field agents for HIV/AIDS use them to bring information on other diseases to the
community, as well as to detect and refer people for treatment. Human resource capacity in health
centres is also improved. There are more health workers available, their attitude and motivation is
generally improved, and they are better trained.
Community/Civil Society
GHI funding reinforced NGOs in terms of their size, their power, and their roles. NGOs often have
bigger budgets than the districts in which they work. One informant suggested that GHIs have
empowered NGOs beyond what is normal or healthy for the system:
The NGOs are taking places that should not be theirs in this country. They often
make their own rules. They comply very little with the very few norms of the
national health authority. And that being said, this is not the fault of the NGOs;
this is more the fault of the public authority, the fault of the funders, the donors
who, during several years, especially during the period of political crisis, have
abandoned the financing of the state to give the funds to the NGOs.
Given the large number of NGOs working in the country and lack of state authority, there is little
coordination among NGOs. Greater coordination could presumably serve to strengthen NGOs’
impact. GHI funding has also affected the role of NGOs by encouraging them to shift their focus to
target diseases. The number of NGOs working on HIV has multiplied with GHI funding, and the
organizations previously working on target diseases have been strengthened further in those
areas.
Discussion
The majority of interviewees report that the Global Fund and PEPFAR have had a positive impact
on the Haitian health system as a whole; positive effects on infrastructure, M&E, and health
workforce are most evident. In particular, informants note that all those people seeking healthcare,
not just those with HIV, benefit from improvements in physical infrastructure and expansion of the
trained health workforce. Many respondents believe that the most pressing issue is that of creating
an integrated healthcare system. This would require fewer vertical programmes related to GHI
85

funds and more funding focused on improving the system as whole. GHI funding has introduced a
tension in the health system by empowering NGOs while at the same time constraining somewhat
the Haitian government’s ability to coordinate GHI activities within a national plan and set of
healthcare priorities. As a result, there is a desire for the funding to help strengthen the MOH and
support its role as the primary coordinator for health action in the country. To achieve complete
integration, improved coordination will be needed between the government, the Global Fund and
PEPFAR. For example, in Haiti, the Principal Recipient for Global Fund funds is not the government,
a fact that perpetuates the lack of coordination.
Although initially both the Global Fund and PEPFAR were considered vertical programmes by both
the government of Haiti and by implementing NGOs, data showed that certain implementers were
willing to challenge the vertical approach of these GHIs from the beginning and to push the
funders to allow a more diagonal approach to health system strengthening. This approach, over
time, became more accepted by local GHI officials and has also become more acceptable for GHI
objectives and programming. The Global Fund, for example, now encourages applications for
funding for health system strengthening, and PEPFAR, particularly in its second phase, is described
by interviewees as being more open and encouraging of an integrated approach.
86

References
[1] World Development Indicators 2007. Washington DC, The World Bank, 2007 (Proprietary online
database: http://ddp-ext.worldbank.org.ezpprod1.hul.harvard.edu/ext/DDPQQ/member.do?method=getMembers;
accessed 20 March 2009).
[2] Government of Haiti, Preparatory Committee for the DSNCRP, Ministry of Planning and External
Cooperation. Growth and Poverty Reduction Strategy Paper. Port-au-Prince, Preparatory
Committee for the DSNCRP, Ministry of Planning and External Cooperation, 2007.
[3] Human Development Report 2006 - Beyond Scarcity: Power, Poverty and the Global Water
Crisis. New York, United Nations Development Programme, 2006.
[4] United Kingdom Jubilee Debt Campaign. The Debt Crisis: Haiti. London, United Kingdom
Jubilee Debt Campaign, 2008 (http://www.jubileedebtcampaign.org.uk/?lid=3113; accessed 13
November 2008).
[5] Small Island Developing States: Haiti. UN-OHRLLS, 2008
(http://www.unohrlls.org/en/orphan/92/; accessed 12 November 2008).
[6] Buss T. Failure of Foreign Aid to Haiti. Paper presented at the annual meeting of the
International Studies Association 48th Annual Convention. Chicago, IL, USA, 28 Feb 2007.
[7] Humanitarian Appeal 2004: Haiti Flash Appeal. United Nations Office for the Coordination of
Humanitarian Affairs, 2004 (http://www.un.org/depts/ocha/cap/haiti.html; accessed 11 November
2008).
[8] Epidemiological Fact Sheet on HIV and AIDS: Core data on epidemiology and response, Haiti
2008 Update. Geneva, UNAIDS/WHO Working Group on Global HIV/AIDS and STI Surveillance,
2008 (http://apps.who.int/globalatlas/predefinedReports/EFS2008/full/EFS2008_HT.pdf; accessed
20 March 2009).
[9] Haiti. Epidemiological Bulletin, 2003, 24(1):13-16.
[10] Haiti Country Coordinating Mechanism (CCM). The Global Fund to Fight Aids Tuberculosis and
Malaria, Proposal Form, Fifth Round. Geneva, GFATM, 2005
(http://www.theglobalfund.org/grantdocuments/5HTIH_1032_492_full.pdf; accessed 20 March
2009).
[11] Chatterjee P. Haiti's forgotten emergency. The Lancet, 2008, 372:615-618.
[12] Haiti. In: Health in the Americas, 2007 Edition. Volume II-Countries. Pan American Health
Organization, 2007:412-429.
[13] World Health Statistics 2009 (http://www.who.int/whosis/en/; accessed 15 March 2009).
[14] Haiti Fiscal Year 2004 Country PEPFAR Operational Plan (COP). PEPFAR, 2004.
87

INDIA: Exploring how disease-specific
programmes in TB and HIV/AIDS interact
with health systems.
Abstract
Krishna Dipankar Rao 1 , Mathew Sunil George 1 , Aarushi Bhatnagar 1 , Babita Rajkumari 1 ,
Maulik Chokshi 1 , Preeti Kumar 1 and Indrajit Hazarika 1
1 Public Health Foundation of India, New Delhi.
GHIs represent an important recent trend in international health. Several GHIs operate in India
(Global Fund, PEPFAR and GAVI). Global Fund is the biggest contributor to India’s health sector.
The aim of this study is to understand how programmes supported by the Global Fund in India,
National Aids Control Programme (NACP-3) for HIV/AIDS and the Revised National TB Control
Programme (RNTCP) for TB, interact with existing health systems. Key informant interviews and
data from health facilities were collected in three states of India. Findings indicate that the RNTCP
and especially the NACP have great potential for strengthening health systems in several ways: (1)
by placing key staff at health facilities, especially at the periphery, who then function as a resource
for the entire health facility; (2) by strengthening laboratory services, sharing supplies and
resources with labs meant for general patients or providing services for non-programme specific
patients; (3) by improving patient perceptions of public health facilities through the presence of
relatively well-run programmes and creating demand for better quality services. The Global Fund
has also helped to strengthen the presence, voice and capacity of civil society organizations (CSOs)
to participate in national health programmes. Yet, these synergies did not take place at all health
facilities and the RNTCP was more conducive to generating these synergies. Here, three factors
appear to be important: (1) the explicit intention of the programme to strengthen health systems;
(2) the locus of administrative control over programme-specific contractual health workers; and (3)
the locus of administrative control over the programme as a whole.
Background
The largest country in South Asia, India is bordered by Pakistan to the west; China, Nepal, and
Bhutan to the northeast; and Bangladesh and Myanmar to the east. India is home to 1.1 billion
people with an average annual population growth rate of 1.4% [1]. Between 1997 and 2007, its
GDP grew at 6.9% per year on average, making India one of the ten fastest growing economies in
the world [1]. India ranked 132nd out of 179 countries on the UN Human Development Index in
2006 [2]. Premature death and illness due to major environmental health risks - such as a lack of
safe water, vector borne diseases, and agro-industrial chemicals - account for nearly 20% of the
total burden of disease [3]. India's public health spending is low at around 1% of GDP. The
National Rural Health Mission (NRHM), launched in 2005 to strengthen the rural public health
system, aims to increase public spending on health to 2-3% of GDP by 2012 [4]. In 2002, the
Ministry of Health and Family Welfare in India estimated that less than 20% of the population
which seek outpatient services, and less than 45% of that which seek inpatient treatment, make
use of such services in public hospitals [5]. The private health sector in India is growing at a
remarkable rate. In 2005, India ranked among the top 20 of the world's countries in its private
spending. Employers paid for 9% of spending on private care, health insurance 5-10%, and 82%
was from personal funds. More than 40% of all patients admitted to hospital had to borrow
money or sell assets to cover expenses [6].
88

Table 1 Basic Socioeconomic, Demographic and Health Indicators (*)
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 1,123,319 2007 World Bank
Geographic Size (sq. km) 2,973,190 2007 World Bank
GDP per capita, PPP (constant 2005
international $)
2,598.59 2007 World Bank
Gini index 36.8 2005 World Bank
Government expenditure on
health (% of general government
expenditure)
3.2 2007 WHO NHA
Per capita government
expenditure on health (current
US$)
9 2007 WHO NHA
Physician density (per 10,000) 6 2004 WHO SIS
Nursing and midwifery density (per
10,000)
13 2004 WHO SIS
Maternal mortality ratio (per
100,000 live births)
450 2005 WHO SIS
DTP3 coverage (%) 62 2007 WHO SIS
Estimated adult HIV (15-49)
prevalence (%)
0.3 (0.2-0.5) 2007
UNAIDS
Estimated adult antiretroviral
therapy coverage (%)
20 2007 UNGASS
Estimated pediatric antiretroviral
therapy coverage (%)
35 2007 UNGASS
Tuberculosis prevalence (per
100,000)
283 2007 WHO GTD
Estimated malaria deaths 15,008 2006 WHO WMR
Table 2 Global Health Initiative Investments
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 1, TB 8,655,033 8,655,033
Round 2, HIV/AIDS 248,367,328 92,702,000
Round 2, TB 99,755,540 23,258,723
Round 4, HIV/AIDS 140,878,118 78,663,024
Round 4, Malaria 63,544,954 47,705,431
Round 4, TB 25,823,483 19,113,943
Round 6, HIV/AIDS: 85,688,377 30,346,423
Round 6, TB 21,308,621 8,579,594
Round 7, HIV/AIDS 30,720,116 7,599,387
TOTAL: 739,561,352 329,457,252
89

PEPFAR*
Year Amount Disbursed (in US$)
2004 20,770,000
2005 26,610,000
2006 29,585,000
2007 29,935,000
2008 29,829,900
TOTAL: 136,729,900
*Not a PEPFAR focus country; above sums represent total allocations to PEPFAR
country programmes from bilateral U.S. sources including USAID, Department of
Health and Human Services, Department of Labor, and Department of Defense.
GAVI (in US$)
Disease Priority Amount Approved (in US$)
Hepatitis B vaccine 66,163,000
Vaccine introduction grant 1,200,000
Injection Safety 25,929,319
TOTAL: 93,292,319
Disease specific programmes have a long history in India, with the central government targeting
specific conditions since the 1950’s. Currently around ten such programmes are operational,
covering a variety of diseases. They operate through the government funded public health system.
In a weak health system environment and where funds for the health sector are limited, it becomes
important to understand how such disease control programmes can contribute to strengthening
health systems.
Several GHIs operate in India, including Global Fund, PEPFAR and GAVI (Tables 1-3). The Global
Fund came to India in 2003 and has rapidly become one of the largest external donors to India’s
health sector. India has received support from various Global Funds rounds for three of its national
disease control programmes: the National AIDS Control Programme (NACP - 2 and 3), the Revised
National TB Control Programme (RNTCP) and National Vector Borne Disease Control Program for
malaria. For HIV/AIDS there have been three Principal Recipients through Round Six of Global Fund
funding: the National AIDS Control Organization (NACO) [7], the Population Foundation of India
(PFI) and the India HIV/AIDS Alliance. The last two Principal Recipients are CSOs. In Round Seven
there were two other Principal Recipients, the Indian Nursing Council and the Tata Institute of
Social Sciences. For TB, the Central TB Division in the Ministry of Health and Family Welfare has
been the only Principal Recipient in all funding rounds. Contributions by Global Fund and other
donors to the NACP-3 and RNTCP are substantial. However, HIV/AIDS has received the bulk of
Global Fund support to India. Around 45% of the NACP-3’s budget is from external sources [8] of
which approximately half comes from the Global Fund. Overall, the Global Fund contributes 26%
of the NACP-3 budget, a substantial increase from 6% for the NACP-2 [9].
90

Objectives and Methodology
The aim of this study is to understand how programmes supported by the Global Fund in India
interact with existing health systems. The study focuses on the activities of the NACP-3 as well as
the RNTCP since the majority of Global Fund funds are directed at these two programmes. It
attempts to answer the following specific questions:
1. How do Global Fund-supported programmes interact with health systems?
2. What positive synergies emerge when Global Fund-supported programmes interact with
local health systems? Which factors influence the extent and nature of this interaction?
Site Selection
The states of Andhra Pradesh, Uttarakhand and Manipur were selected on the basis of significant
funding provided by the Global Fund, epidemiological profile (HIV and TB) and geographical
spread. Andhra Pradesh has a high caseload of TB and HIV whereas Uttarakhand and Manipur have
a high disease burden of TB and HIV, respectively. Moreover, the states selected offer a broadly
representative sampling due to their location in the southern, northern and eastern parts of the
country respectively.
Selection of Health Facilities
Public sector health facilities where Global Fund funded activities are functioning were selected in
each state. One district nearest to the state capital was randomly selected. From the extant pool of
facilities in the district, the district hospital was automatically included in the sample, as was the
community health centre (CHC) nearest to the district hospital and nearest primary health centre
(PHC). For assessing the RNTCP programme, a district hospital, a CHC and a PHC (preferably a
tuberculosis unit) were selected. Similarly, for the NACO programme, a medical college hospital, a
district hospital and a CHC were chosen. In Uttarakhand all the relevant health facilities were
present in the district where the district capital is located.
Primary data collection – key informant interviews
In-depth interviews with key informants in the three study states were conducted between
February and May 2009. Key Informants were identified purposively based on their roles and
experience with the NACP, the RNTCP and the general health system of India. In addition, others
were identified using the strategy of snowball sampling. Key informants included programme
managers of the NACP, RNTCP, Directorate of Health Services, central and state health ministry and
relevant staff of health facilities.
A key informant guide was developed and used to facilitate the process of data collection. The
guides were pre-tested and a few modifications were made based on this experience.
91

Primary data collection – facility service statistics
Facility level data were collected from a total of seven public health facilities providing TB and HIV
services in Andhra Pradesh and Manipur. These included a teaching hospital in each state’s capital
city, and in the selected district, the District Hospital, a CHC and a PHC. The teaching hospitals in
both states provided tertiary care and typically housed a designated microscopy centre (DMC), a
Directly Observed Therapy, Short-course (DOTS) centre, an integrated counselling and testing
centre (ICTC), parent to child transmission centres (PPTCT) and an anti-retroviral treatment (ART)
centre. The district hospital in Manipur did not have an ART centre. DMC, DOTS and ICTC centres
were also present at CHCs. Only DMC and DOTS centres were present at PHCs.
Results
Qualitative Findings
Information gathered from key informants and health facilities shed light on several aspects of
Global Fund interaction with health systems in India. These include the perceived role of the
Global Fund as a donor, the ways in which Global Fund-supported programmes (RNTCP and NACP)
interact with health systems, and Global Fund relations with civil society organizations. Both the
RNTCP and the NACP are national programmes and they adhere to a country-wide standard
design. Consequently, what is reported here regarding the interaction of these programmes and
the health system is attributable to these programmes and not specifically to Global Fund policies.
Donors have limited influence on health policy
Donor agencies were not perceived to influence health policy as much as technical agencies.
However, donor funding priorities did influence what gets attention in the national health space
and this can be different from the health needs of the country. The Global Fund’s role as a donor
agency fits this pattern to the extent that it was perceived as having a hands-off approach towards
the programmes it supports. Key informants viewed the Global Fund’s role as being confined to
providing and overseeing the use of funds directed to the RNTCP and NACP. It played no direct
role in influencing the shape, scope or implementation of the programmes it funded. However,
key informants felt that the Global Fund does guide country proposals in certain directions and to
that extent influences existing national strategies.
Global Fund- supported programmes can strengthen human resources at health
facilities
The Global Fund- supported national disease control programmes in HIV/AIDS (NACP) and TB
(RNTCP) operate within India’s vast and multi-tier government funded public health system.
Human resources for health is a critical issue in India. One of the important problems in this area is
placing health workers in rural areas. Health facilities, particularly those at the periphery, are
plagued with a large number of vacancies severely compromising their ability to provide health
care.
The RNTCP delivers services through health facilities in the public health system. These include
designated teaching and district hospitals, community and primary health centres. Where there
92

are vacancies of regularized doctors and lab technicians at health facilities designated for TB
services, the RNTCP follows a policy of filling a fixed percentage of these vacancies with
contractual doctors and lab technicians. In one of the states studied, the programme had even
created a new cadre of lab technician called the ‘sputum microscopist’ to overcome the acute
shortage of qualified general lab technicians. In this way, the RNTCP has enabled health facilities,
particularly those at the periphery of the health system, to continue providing health services.
The NACP also delivers services through a range of public health facilities and places a variety of
contractual health workers - doctors, lab technicians and counsellors – at designated health
facilities. Unlike the RNTCP, the programme does not seek to fill vacancies in the existing public
health workforce. However, some of the NACP lab technicians reported doing a variety of
laboratory tests and did not confine themselves to only HIV testing.
The presence of contractual staff at health facilities need not translate into human resource
strengthening. Several factors influence whether contractual health workers limit themselves to
specific programme obligations or not. First, the type of health facility these health workers were
placed determines the scope of their activities. Contractual health workers in both the RNTCP and
the NACP almost never went beyond their programme duties at district or teaching hospitals. In
contrast, their counterparts at peripheral health facilities usually went beyond their contractual
obligations because they typically were the only resource available and programme specific
workload was lower. Another factor was the attitude that senior programme managers and their
supervisors had towards integration. Where programme managers and supervisors felt that
integration was a good, programme specific health workers tended to be less restrictive in the
scope of their duties.
Global Fund-supported programmes can strengthen lab services by sharing resources
The presence of the RNTCP and the NACP programmes at health facilities strengthened laboratory
services at peripheral health centres in several ways and beyond the provision of human resources
or enabling a diversity of lab tests to be conducted at these facilities. These included the using of
lab supplies purchased for these programmes with non-programme specific patients. Further, at
some health facilities programme specific labs provided free services to general patients,
especially those who were poor.
Global Fund- supported programmes can increase trust in the public health facilities
The presence of disease control programmes at health facilities, according to key informants,
appears to have contributed to increasing trust in the public system. By placing qualified health
workers at health facilities where there were none earlier and sharing resources with non-disease
specific health services, both the RNTCP and, to a lesser extent, NACP have contributed to the
ability of health facilities to provide quality health services. Further, by providing good quality
services, disease control programmes improve trust in public health facilities and build public
expectations for better overall service quality.
93

The Global Fund has strengthened the role of Civil Society Organizations in disease
control programmes
Global Fund policies ensure that CSOs are directly involved in the consultative process for country
proposals. Further, Global Fund’s policy of ‘dual track’ funding ensures that CSOs are part of the
implementation process as well. Key informants in all the states studied were of the view that CSOs
had contributed towards increasing the quality and coverage of TB and HIV/AIDS services in India.
In both these programmes, CSOs actively participate in service delivery like running TB labs,
providing care to patients, and helping the programme reach vulnerable and difficult to reach
groups by engaging with them.
The engagement of CSOs in provision of health services has a long history and existed long before
the Global Fund’s arrival. However, the Global Fund has definitely institutionalized their
involvement and provided them a legitimate space for influencing the design and implementation
of Global Fund- supported programmes. Equally important, this engagement has made CSOs a
partner in the national disease control programmes of TB and HIV/AIDS, a position which they did
not previously have.
Quantitative Findings
Health Management Information Systems (HMIS) data collected from the sampled health facilities
and interviews with facility staff in Andhra Pradesh and Manipur indicated that, in general, the
presence of Global Fund funding appears to have increased availability and uptake of health
services at the sampled health facilities. Notably, Multi-drug Resistant Tuberculosis (MDRTB) drugs
became available only after Global Fund funding started. Temporal trends also suggest that
uptake of certain TB services (e.g. TB suspects evaluated) increased when Global Fund funding was
present, though trends in new patients started on DOTS does not follow this trend. HIV/AIDS
services also appear to have increased with Global Fund funding – both testing and counselling
witnessed impressive increases at most health facilities. It is important to note that these findings
are based on a small number of health facilities which are not representative of the health facilities
in either of the two states. In addition, causal inferences should not be drawn from the information
presented here since several factors not connected with Global Fund funding could be responsible
for the observed trends. For instance, the National Rural Health Mission has made considerable
resources available for strengthening service delivery in the public health system.
94

Discussion
Global Fund- supported programmes have been critical for making TB and HIV/AIDS services
available to the population and are the health system’s primary means of controlling these
diseases. In particular, Global Fund funds contributed importantly to the expansion of ART services
in India. Global Fund proposals have consciously required applicants to detail how their disease
specific strategies can strengthen health systems. Further, the latest round of Global Fund funding
to India includes the Nursing Council of India as a primary recipient. This should provide much
needed support to strengthen the position of nurses in India and the health system.
Both the RNTCP and especially the NACP have a great potential for strengthening health systems
in several ways. First, they place health workers at health facilities, especially at the periphery, who
function as a resource for the entire health facility and not just for their specific programme. Key
functions of health facilities like lab services can also be strengthened by sharing supplies and
resources with labs meant for general patients or providing services for non-programme specific
patients. Furthermore, the presence of these well run programmes at health facilities improves
patient perceptions of public health facilities and also creates demand for better quality services.
The Global Fund has also contributed in important ways to strengthen the presence, voice and
capacity of civil society organizations to participate in national health programmes.
While there is tremendous potential for Global Fund-supported disease control programmes to
strengthen health systems, these synergies did not take place at all health facilities. Programmes
like the RNTCP seem more conducive to generating these synergies. It is important to understand
why disease control programmes have not been able to contribute more to strengthening health
systems. Three factors appear to be important: (1) the explicit intention of the programme to
strengthen health systems; (2) the locus of administrative control over programme-specific
contractual health workers; and (3) the locus of administrative control over the programme.
The RNTCP was designed to work through the existing health system infrastructure and with
human resources available in the public health system. Where these are lacking, the programme
has an explicit policy of supplementing it, thereby strengthening the capacity of health facilities.
The NACP also works through the public health system and there were several instances where
resources for the NACP contributed to health services beyond services specific to HIV/AIDS care.
Yet, there did not seem to be an emphasis within the programme and by its managers on building
connections and contributing to non-HIV/AIDS specific services.
While both NACP and RNTCP staff receive salaries directly from the relevant programme, RNTCP
employees are supervised by the health facility and district managers since they are typically filling
an existing vacancy. NACP staff report to the AIDS control society of their state. This chain of
command gives health facility and health system managers little control over NACP staff;
consequently, they found it difficult to engage them in work beyond HIV/AIDS.
In each state, the Directorate of Health Services is responsible for organizing and managing the
delivery of curative and preventive health services through the public health system. The
management unit of the RNTCP is housed within the Health Directorate, while the NACP is
administered by the State AIDS Control Society (SACS) which is housed outside the state Health
Directorate (though in some instances officers from the directorate are also SACS officers). The
programme management unit’s location influences the degree to which programme activities and
resources contribute to strengthening the health system. Since the Directorate is responsible for
the entire health system in a state, close coordination with it enables better alignment of disease
control programme activities and resources with those of the health system.
95

References
[1] The World Bank Group. World Development Indicators (WDI) Online. (http://ddpext.worldbank.org.ezp-prod1.hul.harvard.edu/ext/DDPQQ/member.do?method=getMembers;
Proprietary website, accessed 19 February 2009).
[2] United Nations Development Programme. Human Development Reports: India.
(http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_IND.html; accessed 19
February 2009).
[3] World Health Organization. India Health System Profile.
(http://www.searo.who.int/EN/Section313/Section1519_10852.htm; accessed 19 February 2009)
96

Kenya: The Impact of Global Health Initiatives on
the Health System
Abstract
Joseph Rhatigan *† , Erin Sullivan ‡† , Kileken ole-MoiYoi ‡† , George Kimathi § , Nayana
Dhavan *† , Ephantus Kabiru §
There has been an increasing interest in understanding the effects that GHIs, such as the Global
Fund and PEPFAR, have had on nations’ overall health systems. We sought to evaluate the impact
of these two GHIs on Kenya’s health system.
We conducted semi-structured interviews with 37 key informants from across the health sector
and examined documentary material. Interviews were recorded and transcribed. The data was
then coded and analyzed using qualitative data analysis software.
Informants reported that PEPFAR and the Global Fund had an overall positive effect on Kenya’s
health system. Informants identified PEPFAR’s governance, procurement systems, and monitoring
and evaluation (M&E) systems as key drivers of its achievements. Informants identified the Global
Fund’s ability to coordinate funding with government priorities as a key driver of its positive
effects.
Informants also identified opportunities for maximizing GHIs’ positive effects on the health system
and minimizing negative effects. PEPFAR’s lack of coordination with government decision-making
processes and the constraints it imposes on the use of funding were described in unfavourable
terms. Similarly, the Global Fund’s complex grant funding process and its perceived lack of
oversight and accountability were described unfavourably.
Although GHIs such as the Global Fund and PEPFAR have successfully strengthened the Kenyan
health system, there are opportunities for them to have a greater positive effect.
* Brigham and Women's Hospitals, Department of Global Health Equity
† Global Health Delivery Project
‡ Harvard School of Public Health, François-Xavier Bagnoud Center for Health and Human Rights
§ Kenyatta University, Nairobi
97

Background
The Republic of Kenya lies on the equator in East Africa and is home to 37.5 million people (see
Table 1) [1]. In 2005, Kenya ranked 148 th out of 177 on the UN Human Development Index [2].
Seventy-nine percent of all Kenyans live in rural areas. Most of the rural population relies on smallscale
subsistence agriculture. Fifty-eight percent of all Kenyans live below the poverty line, and
unemployment is 40%. Official development Assistance (ODA) to Kenya was US$ 934.4 in 2006 [3],
while total external debt was US$ 6.5 billion or 28.7% of GDP.
The Kenyan Ministry of Public Health and Sanitation and the Ministry of Medical Services (which
until April 2008 were united as the Ministry of Health) direct the public health activities in the
country. In 2006, the Ministry of Health (MOH) managed approximately 52% of all health facilities,
with the remainder managed by private and non-profit entities. Out-of-pocket expenditure
compr 16 ised 80% of all private expenditure on health, while private prepaid plans comprised 6.9%
in 2005. In 2006, external resources made up 14.8% of the total expenditure on health [3]. An
estimated 40% of rural households had no access to health services in 2000. Approximately 12% of
registered physicians living in Kenya worked in the public sector, and more than 80% of all
physicians practiced in urban areas [4, 5]. Only 25% of Kenyan health facilities have year-round
access to water, and 50% have a regular supply of energy or a back-up generator [6].
Table 1 Basic Socioeconomic, Demographic and Health Indicators (*)
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 37,531 2007 World Bank
Geographic Size (sq. km) 569,140 2007 World Bank
GDP per capita, PPP (constant 2005
international $)
1,456 2007 World Bank
Gini index 42.5(†) 2007 UNDP
Government expenditure on health
(% of general government
expenditure)
9.7 2007 WHO NHA
Per capita government expenditure
on health (current US$)
18 2007 WHO NHA
Physician density (per 10,000) 1 2002 WHO SIS
Nursing and midwifery density (per
10,000)
12 2002 WHO SIS
Maternal mortality ratio (per 100,000
live births)
560 2005 WHO SIS
DTP3 coverage (%) 81 2007 WHO SIS
Estimated adult HIV (15-49)
prevalence (%)
7.1-8.5
2007
UNAIDS
Estimated antiretroviral therapy
coverage (%)
42 2007 UNGASS
Tuberculosis prevalence (per
100,000)
319 2007 WHO GTD
Estimated malaria deaths 27,049 2006 WHO WMR
16
98

AIDS was first documented in Kenya in 1984, and by 2007, between 1.5 and 2.0 million people
were living with HIV/AIDS [7]. Tuberculosis (TB) incidence in Kenya has increased steadily over the
past two decades (see Table 2) [8]. Of all new cases, 52% occurred in HIV-positive individuals [9]. In
2009, Kenya ranked 13th among the 22 highest TB burden countries, as designated by the Stop TB
Partnership, but had become the first country in sub-Saharan Africa to achieve global targets for
both case detection and treatment success [10].
All tuberculosis treatment, monitoring, and drug procurement activities are coordinated by the
Kenyan Ministry of Health’s Department of Leprosy, Tuberculosis, and Lung Disease (DLTLD) with
support from international agencies such as the Global Fund. Since 2003, the Global Fund has
approved US$ 326 million in funding for Kenya, although only US$ 160.1 million of those funds
have been disbursed to date (see Table 3). In addition to the Global Fund, PEPFAR, the US
government programme for international HIV control, is a major donor to HIV care providers in
Kenya. Kenya was designated one of PEPFAR’s 15 focus countries at the programme’s inception in
2003, and has received US$ 1.3 billion for HIV treatment through the end of 2008 [11].
Methodology
We collected data through semi-structured interviews, observation, and examination of
documentary material. Key informant interviews followed a standardized semi-structured
interview template designed to evaluate how GHIs—primarily PEPFAR and the Global Fund—have
interacted with Kenya’s health system. Embedded units of analysis were limited to PEPFAR and the
Global Fund to focus the scope of the initial nine-month research programme. Interviews
addressed the following health system building blocks as defined by the WHO: governance,
financing, health workforce, M&E, and health technologies.
The key informant sampling strategy targeted a diverse range of implementers, policymakers and
health leaders. The study used two sampling methods: purposive sampling and snowball
sampling. Key informants had extensive experience with PEPFAR, the Global Fund, or both. The
study had a sample size of 37 key informants: 12 Government of Kenya (GoK) officials working
within key ministries at the national and provincial levels; 10 members of various Kenyan-based
and international nongovernmental organizations (NGOs); three members of international
multilateral organizations; eight US government employees; and four informants from the private
sector.
Interviews were recorded and transcribed to both ensure accurate data collection and facilitate
qualitative analysis. Interviewers also took field notes based on their observations and postinterview
debriefings. These notes were included in the qualitative data set. Transcribed interviews
were entered into NVivo8, a qualitative data analysis tool (QSR International, Cambridge, MA). A
thematic qualitative analysis approach with an iteratively developed set of codes was used to
examine the data. To further support the authenticity of findings and auditability of analytic
processes, we engaged in inter-rater reliability activities as we created and applied codes,
developed memos about our analytic decision-making and conducted participant validation
exercises.
For the purposes of triangulating with the data collected in key informant interviews, we collected
and reviewed publicly available documents (i.e. National Health Strategy, Demographic Health
Surveys, and PEPFAR and Global Fund Country Reports).
99

Results
Leadership and Governance
Data showed that GHI governance was an important determinant of impact. PEPFAR’s leadership
and management structures were cited as integral to the rapid increase in the number of people
on antiretrovirals (ARVs) in Kenya. The Global Fund leadership and management structures were
perceived as presenting persistent challenges to achieving the full potential of its funding.
Informants indicated that it was relatively easy to communicate and problem-solve with PEPFAR
representatives, due to the established relationships with US government (USG) agencies in Kenya
and the resultant familiarity with their organizational structures. Access to PEPFAR leadership,
however, was somewhat hindered by its location in the US Embassy compound and the associated
security protocols. Respondents indicated that unclear mandates and poorly delineated roles and
responsibilities of the USG agencies—US Agency for International Development (USAID), the
Department of Health and Human Services (HHS), Centers for Disease Control and Prevention
(CDC), Walter Reed Medical Research Institute, Department of Defense (DOD), Department of State
and Peace Corps—created interagency tension and competition that led to inefficiencies.
However, informants also reported that these agencies’ scope of work was better defined and
clarified over the past year, and that this may help reduce tensions.
Unlike PEPFAR, the Global Fund does not have in-country representatives. All concerns must be
raised with the main in-country governance mechanism, the Country Coordinating Mechanism
(CCM), or by contacting the country portfolio manager in Geneva. Additionally, any programmatic
changes after the initial grant approval must be directed to Geneva for review, which causes
delays in implementation and reporting. Although informants believed that the Global Fund
country portfolio manager in Geneva was accessible and receptive, they indicated that Global
Fund-funded programmes would significantly benefit if portfolio managers were located in Kenya.
The Global Fund CCM elicited both positive and negative assessments from respondents. Some
informants lauded the creation of the CCM as a country-led team that brought together different
stakeholders in the health system and enabled donor resources to be better aligned with national
health priorities. Others felt CCM members did not adequately represent the different sectors
within the health system and that the government representation was disproportionately large.
They suggested that the CCM was a parallel system that was imposed by the Global Fund and that
it did not fit with the pre-existing national structures, since it only coordinated activities around
three diseases rather than across the entire health system.
Civil society organization (CSO) representatives to the CCM were elected democratically, however
the CCM chair and vice chair were not elected. Informants indicated that there were tensions
within the CCM, including a lack of trust between CSO and government representatives. Many
CCM members were members of the Global Fund sub-recipient organizations. Several informants
criticized CCM members who worked to benefit their own sectors exclusively or who attempted to
obtain Global Fund funds for their own programmes. Informants believed that this conflict of
interest hindered the transparent functioning of the CCM and the overall effectiveness of the
Global Fund in Kenya. Respondents criticized the CCM leadership for heavy infighting and
irregularities in reporting. Several respondents indicated that the CCM was not a legal entity and,
as a result, had unclear accountability.
100

Table 2 Global Health Initiative Investments (*)
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 1, HIV/AIDS 220,875 220,875
Round 1, HIV/AIDS 2,650,813 2,650,813
Round 2, HIV/AIDS 106,786,807 68,006,881
Round 2, Malaria 27,700,377 4,640,447
Round 2, TB 8,761,405 3,299,522
Round 4, Malaria 162,173,085 76,103,617
Round 5, TB 13,499,895 3,511,242
Round 6, TB 4,206,357 1,710,684
Round 7, HIV/AIDS 46,663,557 16,538,950
TOTAL: 372,663,171 176,683,031
PEPFAR
Year Amount Disbursed (in US$)
2004 92,474,390
2005 142,937,153
2006 208,269,879
2007 368,129,182
2008 534,794,604
TOTAL: 1,346,605,208
GAVI
Disease Priority Amount Approved (in US$)
Pentavalent vaccine 171,060,000
Yellow fever vaccine 325,000
Vaccine introduction grant 100,000
Injection Safety 1,245,000
Immunization services support 8,460,680
Health systems strengthening 9,903,000
TOTAL: 190,093,357
World Bank MAP
Title FY Approved/Closing Date Commitment (in US$)
HIV/AIDS Disaster Response Project 2001/2005 50,000,000
Total War Against HIV and AIDS
(TOWA) Project
2007/2011 80,000,000
TOTAL: 130,000,000
101

Informants noted that a general lack of consultation with local governments and communities is a
common criticism of GHIs and international NGOs involved in global health. Many informants,
particularly those in government, described PEPFAR’s limited interaction with the government as a
shortcoming, and identified this as a significant barrier to achieving broader health system
improvement. While respondents appreciated the connection that PEPFAR had at the community
level through the implementing partners, many informants supported increased engagement by
PEPFAR at community level as well.
Informants believed that the Global Fund’s interaction with the government indicated its desire to
strengthen national capacity by creating a country-led approach. Although constituency
development funds (CDF) have benefited local communities, most beneficiaries were not aware of
the source of funding and the actual magnitude of the impact was often questioned by
informants.
There was general agreement that PEPFAR succeeded in attaining its initial goals, although many
informants questioned the alignment of these goals with national priorities and the sustainability
of PEPFAR’s progress. Many informants criticized disease-focused approaches and explained that
the allocation of both PEPFAR and Global Fund funds was disproportionate to the health needs of
the population. Many respondents voiced concern about the health issues GHIs do not address,
such as maternal and child health, improving access to clean water, better sanitation and food
safety. However, given the programme’s objectives, respondents believed that it was necessary for
PEPFAR to assume a vertical structure to achieve its treatment goals in a timely manner.
Respondents believed the Global Fund had a more comprehensive approach to health system
improvement because of its high level of interaction with the government.
Both PEPFAR and the Global Fund could improve their coordination with existing government
priorities, systems and structures. Managerial capacity was identified as posing a significant
challenge to Kenya’s ability to maximize the benefits of GHI funding, particularly from the Global
Fund. Limited human resources and poor procurement systems hindered the country’s ability to
absorb and implement funding effectively. Informants noted that the current training system is
focused on increasing the numbers of doctors and nurses, but not on management, motivation, or
productivity. The health system would benefit from providing management training to health
workers to improve their skills in grant application processes, reporting procedures, fund
management, and controlling the high rate of turnover, particularly in the public sector.
Financing
Data showed that the influx of funding from PEPFAR and the Global Fund had a positive impact on
Kenya’s health system. Key informants identified PEPFAR’s timely disbursement of funding as an
important factor contributing to its impact. Some respondents, however, expressed concern that
the increased disease-specific funding shifted health priorities within Kenya.
Informants reported that initially understanding the guidelines for the use of PEPFAR money was
challenging. Informants believed that this has generally improved over time, and that now these
processes are better integrated and easier to manage. Respondents expressed an understanding
that PEPFAR is unable to disburse money directly to Kenya’s Treasury, and generally viewed this
neutrally or unfavourably. In comparison, Global Fund funds were disbursed directly to the GoK,
which was viewed both favourably and unfavourably by informants. Respondents felt this allowed
improved alignment with national priorities but raised concerns about accountability and
transparency.
102

PEPFAR was perceived as having a more efficient chain of fund disbursement than the Global
Fund. The transfer of Global Fund funds through multiple accounts was reported to take up to a
year before the monies reached the intended programme. Furthermore, due to the performancebased
financing system, subsequent disbursement of approved Global Fund funds can be delayed
while evaluations are reviewed by the relevant ministries and by external auditors before being
sent to Geneva.
The complexity of the Global Fund grant application system was highlighted as an impediment to
efficient programme implementation. Many informants indicated that an excessive amount of
time and effort was spent applying for funds. Several informants criticized the Global Fund grant
application process as complex, stressful, and frustrating for staff, particularly because it occurs
simultaneously with reporting on previous Global Fund grants. Informants also reported that
initially, the Global Fund grant application and general funding procedures were not clear. Due to
poor performance and poor reporting, Kenya did not receive funding from Global Fund Rounds 7
and 8; only 40% of the approved funding from Round 2 was disbursed by the Global Fund for
similar reasons. Some respondents attributed underperformance to Global Fund policies and
procedures being difficult to understand and follow.
Many believed both PEPFAR and the Global Fund could better communicate and disseminate
information on their funding and implementation activities.
Health Workforce
Since the advent of GHIs, health workers have typically moved out of the MOH into GHI-funded
projects due to better compensation packages. Informants suggested that although GHIs have
significantly increased the number of health professionals in Kenya’s health system—through
improved recruitment, training and compensation—internal migration of highly skilled health
workers remains a significant problem.
Several informants criticized PEPFAR for creating competition for staff among NGOs and between
NGO’s and the GoK. Some informants did suggest that PEPFAR has attempted to improve
harmonization of recruitment, training, and compensation across the health system. The Capacity
Project—a PEPFAR funded programme of IntraHealth, a US-based NGO—coordinates its activities
with the GoK to plan human resources, standardize compensation, provide health worker
development, and support health workers once they are in place.
Despite collaborative efforts such as the Capacity Project, many NGOs offered higher salaries to
highly-skilled workers than the government did. Some informants suggested that Kenya has an
excess of skilled health workers who were not motivated to work in government programmes due
to low compensation and programmatic frustrations. Instead these professionals looked to work
for NGOs in Kenya or elsewhere in sub-Saharan Africa.
Like PEPFAR, the Global Fund also increased the number and capacity of staff in Kenya. Some
Global Fund human resource funds were used to pay personnel who were expected to deliver a
comprehensive care package and not just HIV, TB and malaria services. GoK also used Global Fund
funds to train district-level health care workers regardless of their focus. These actions were seen
as strengthening Kenya’s health system.
Although there is a significant need for trained health workers in Kenya, particularly in rural areas,
the government’s ability to recruit and compensate more health workers is limited. Informants
reported that both the Global Fund and PEPFAR had asked the GoK to absorb staff after certain
103

time periods. Informants believed that staff absorption by the government was a challenge
because the MOH had neither a system in place to manage more health workers nor the funds to
compensate those workers. Some respondents reported that a nationwide human resource
management system that will track all the health workers in Kenya is being developed and may
substantially aid this process.
Several respondents indicated that the health workforce represents an impediment to long-term,
sustainable progress in Kenya’s health system. Although the government has exceeded its
capacity to accommodate and compensate health care workers, the need for trained health
workers remains unmet in many rural areas.
Medical Products, Vaccines and Technologies
Many respondents agreed that both the Global Fund and PEPFAR have strengthened the health
system through the provision of medical products and technologies including vehicles, medicines,
food supplies and computers. GHIs have helped lower the cost of antimalarials, bed nets,
antiretrovirals (ARVs), HIV kits and other related commodities. In addition, they have provided
infrastructure, including facility renovations, new laboratories, information systems, and
procurement support. There were a variety of perceptions on the significance of these
contributions to infrastructure. Some informants felt that PEPFAR’s contributions to physical
infrastructure were minimal given the significant need. Others felt that PEPFAR has recently
increased its contribution to physical infrastructure, especially laboratory development. Regarding
the Global Fund, some respondents suggested that the Global Fund contributed to infrastructure
by providing funding for mobile voluntary counselling and testing (VCT) centres, vehicles, and lab
equipment. Others felt these contributions were insignificant.
PEPFAR’s objective in Kenya focused on rapidly increasing access to ARVs, which required effective
operations, management structures, and efficient supply chains. Although PEPFAR supported
strengthening logistics management within the national Kenya Medical Supplies Agency (KEMSA),
PEPFAR procurement of ARVs was primarily managed by the Mission for Essential Drugs and
Supplies (MEDS), a non-profit organization based in Nairobi. Informants reported, however, that
PEPFAR often supplied the government with ARVs when stock-outs in public facilities seemed
imminent, increasing the number of patients with reliable access to ARVs.
Informants reported that the Global Fund enabled greater access to drugs such as ARVs and
antimalarials as well as to health commodities, including long-lasting insecticidal bed-nets (LLINs)
and coverage with indoor residual spraying (IRS). However, informants’ opinions differed on
whether Global Fund processes and requirements helped stabilize and form better systems for
procurement, service delivery, and management. For example, the Global Fund imposed new drug
procurement and distribution processes for KEMSA, only some of which were viewed as
improvements. Although respondents believed that the Global Fund, along with other donors,
had helped support the planning process at the district level and introduced standard operating
procedures for KEMSA, it was noted that inefficiencies remain which need to be remedied.
Most respondents did believe that the increased availability of medications and improved stock
management systems improved health outcomes. Respondents also reported that support for
infrastructure and health technologies improved the health system beyond the target diseases, as
patients are able to benefit from the expanded diagnostic capacities provided by GHIs.
104

Information
Many respondents agreed that both the Global Fund and PEPFAR improved M&E systems. However,
informants indicated that reporting and performance-based financing structures in Kenya remain
weak. Many informants reported that both PEPFAR’s and the Global Fund’s reporting structures were
too extensive and time consuming. PEPFAR, the Global Fund, and the government each required
different information and indicators in their reports, and respondents believed that a better
harmonization of systems was needed.
PEPFAR requires specific data to be routinely collected. Respondents felt that this necessitated a high
level of staff training to adequately evaluate and report on multiple indicators. Respondents reported
that the M&E requirements were time-consuming. Some respondents commended PEPFAR’s effort to
help organizations understand the importance of monitoring their programme’s progress using
appropriate data. A few informants hoped that training health workers to conduct quality M&E would
benefit the health system overall.
A number of interviewees cited challenges with Global Fund indicators. In particular, they noted that
indicators needed to be better aligned with country constraints because limitations in infrastructure
hindered the effective tracking of some indicators. Additionally, the Global Fund has, at times, added
indicators that were not in the original grant application forcing the GoK to set up separate monitoring
systems for such indicators. A few respondents blamed reporting issues on a lack of communication
between the Global Fund and the government. Informants reported that poor M&E systems
contributed to the Global Fund’s decision to withhold approved disbursements. Some respondents
suggested strengthening accountability throughout the reporting process as a way to improve the
reporting system. It was clear that performance-based funding makes effective and timely reporting
imperative.
Discussion
Informants reported that PEPFAR and the Global Fund had an overall positive effect on Kenya’s health
system. PEPFAR significantly increased the number of people living with HIV/AIDS on ARVs and created
or strengthened the systems providing this treatment. Specifically, informants identified PEPFAR’s
governance, procurement systems, and M&E systems as key drivers of its achievements. Global Fund
funding had perceived positive effects, particularly in malaria and TB control, as well as on HIV
treatment and prevention. Specifically, informants identified Global Fund’s funding coordination with
government priorities as a key driver of its positive effects.
While acknowledging PEPFAR and Global Fund’s overall positive effect on Kenya’s health system, many
informants identified additional opportunities for maximizing these effects and minimizing negative
effects. PEPFAR’s lack of coordination with government decision-making processes and the constraints
it places on the use of funding were described in unfavourable terms. Similarly the Global Fund’s
complex grant funding process and its perceived lack of oversight and accountability were described
unfavourably.
Informants thought that PEPFAR’s and the Global Fund’s impacts on Kenya’s health system were closely
related to the differences in how they disbursed their funding. PEPFAR was viewed unfavourably for
not working closely with the government, but it was viewed favourably for being able to achieve
significant results quickly by disbursing funds efficiently to implementing organizations. The Global
Fund, which was viewed favourably for disbursing funds through government bodies, was felt to have
inefficient procedures for grant approval and reporting that diminished its potential positive impact.
Most informants had concerns about the long-term viability of GHI-sponsored programmes without
lasting commitments from the GHIs. Strengthening Kenya’s health system was deemed essential to
achieving sustainable positive effects from GHI funding.
105

References
[1] The World Bank Group. Kenya Data-at-a-Glance. Washington, DC, World Bank, 2008
(http://devdata.worldbank.org/AAG/ken_aag.pdf; accessed 20 May 2009).
[2] Report on the Global AIDS Epidemic 2008. Geneva, UNAIDS, 2008.
[3] World Development Indicators 2008. Washington, DC, World Bank (proprietary online database,
accessed 12 November 2008).
[4] Siringi S. Kenya government promises to increase doctors' salaries to curb brain drain. Lancet,
2001, 358: 9278:307.
[5] Kimalu PK, Nafula NN, Manda DK, Bedi A, Mwabu G, Kimenyi MS. A Review of the Health Sector
in Kenya. Kenya Institute for Public Policy Research and Analysis Working Paper. Nairobi, 2004.
[6] Government of Kenya National Coordinating Agency for Population and Development, Ministry
of Health, Central Bureau of Statistics, and ORC Macro. Kenya Service Provision Assessment Survey
2004. Nairobi, National Coordinating Agency for Population and Development, Ministry of Health,
Central Bureau of Statistics, and ORC Macro, 2005.
[7] World Health Organization Statistical Information System (WHOSIS). Geneva, WHO, 2008
(http://www.who.int/whosis/en/; accessed 27 February 2009).
[8] Zumla A, Malon P, Henderson J, et al. The impact of the human immunodeficiency virus (HIV)
infection epidemic on tuberculosis. Postgraduate Medical Journal, 2000;76:259-268.
[9] World Health Organization. Global Tuberculosis Control: Surveillance, Planning, Financing.
Geneva, WHO, 2008.
[10] Stop TB Partnership. Tuberculosis in countries online database. Country profile: Kenya.
Geneva, Stop TB Partnership, 2009. (http://www.stoptb.org/countries/GlobalReport2009/ken.pdf;
accessed 21 May 2009).
[11] PEPFAR Country Profile: Kenya. Washington, DC, United States Department of State, Office of
the Global AIDS Coordinator (http://www.state.gov/s/gac/countries/fc/kenya/; accessed May 30,
2008).
106

Kyrgyzstan: Tracking global HIV/AIDS initiatives and
their impact on the health system
Abstract
Gulgun Murzalieva 17 , Julia Aleshkina 15 , Arnol Samiev 15 , Neil Spicer 18 , Gill Walt 16
This study examines the effects of two HIV/AIDS-focused GHIs – the Global Fund and the World
Bank Central Asian AIDS Project (CAAP) – in three regions of Kyrgyzstan: Bishkek/Chui, Issyk-Kul
and Osh/Jalalabad. The Global Fund is the biggest GHI for HIV/AIDS in the country, and
contributed more than US$ 17 million between 2004 and 2008.
The research finds that these GHIs have contributed to substantial scale-up of HIV/AIDS services,
including information/education programmes, voluntary counselling and testing (VCT), harm
reduction programmes and antiretroviral therapy (ART). However, stigma and discrimination are
major barriers to people seeking to access these services. Most HIV/AIDS services are provided by
non-governmental organizations (NGOs), many of which depend on GHIs for a high proportion of
overall funding. GHI support has led to substantial increases in staff working for NGOs, especially
social workers and peer-to-peer outreach workers, and has contributed towards developing staff
and organizational capacity.
GHIs have also strengthened national coordination mechanisms, which now play a significant role
in HIV/AIDS-related decision-making. While coordination and cooperation between government
and NGO HIV/AIDS services has improved, especially in terms of referrals between services,
Kyrgyzstan continues to face several problems concerning the engagement of key actors;
information sharing and the devolution of decision-making powers to sub-national coordination
structures remain problematic.
Background
Kyrgyzstan is a landlocked, mountainous country bordering Kazakhstan, Uzbekistan, Tajikistan,
and China. It is home to 5.2 million people and averaged a 0.9% annual population growth rate
between 2001 and 2007 [1]. Kyrgyzstan ranked 122 nd out of 179 countries on the UN Human
Development Index in 2006 [2]. The country has faced economic problems in its transition from a
command economy to a market economy. Whilst Kyrgyzstan possesses oil and gas resources, it
imports the bulk of what it needs. Economic challenges are particularly acute in the southern
region of the country, where agriculture remains the main industry, unemployment is very high,
and poverty is widespread [3].
17 Centre for Health System Development, Kyrgyzstan
18 London School of Hygiene and Tropical Medicine, Department of Public Health and Policy
Acknowledgement: The research for this study summary was funded by The Open Society Institute, New York. The study
summary was produced with the support of the GHIN Network (www.ghinet.org).
107

A comprehensive, long-term reform of Kyrgyzstan’s health system is in progress. Beginning in
1996, health financing reforms have led to decreased costs, improved quality of care, and
increased spending [4,5]. Primary care services were reorganized in Family Group Practices (FGPs),
which by 2004 had enrolled approximately 98.5% of the population [5]. This has made Kyrgyzstan
the leader in family medicine reform in post-Soviet Central Asia [4]. There is still progress to be
made, as out-of-pocket expenditures still account for more than half of total health expenditure
[6]. In 2006, external resources made up 6.1% of total expenditure on health [6].
Official development assistance (ODA) to Kyrgyzstan was US$ 311.2 million in 2006 [7]. The country
remains highly indebted; it’s total external debt was US$ 2.38 billion, or 84.5% of GDP, in 2006 [1,7].
Table 1 Basic Socioeconomic, Demographic, and Health Indicators (*)
(*) Full data sources for all indicators are provided in Annex 1
Indicator Value Year Source
Population (thousands) 5,235 2007 World Bank
Geographic Size (sq. km) 191,800 2007 World Bank
GDP per capita, PPP (constant 2005
international $)
1,894 2007 World Bank
Gini index 33 2004 World Bank
Government expenditure on health (%
general government expenditure)
8.7 2007 WHO NHA
Per capita government expenditure on
health (current US$)
19
2007 WHO NHA
Physician density (per 10,000) 24 2007 WHO SIS
Nursing and midwifery density (per 10,000) 58 2006 WHO SIS
Maternal mortality ratio (per 100,000 live
births)
150 2005 WHO SIS
DTP3 coverage (%) 94 2007 WHO SIS
Estimated adult HIV (15-49) prevalence (%) 0.1 (

PEPFAR*
Year Amount Disbursed (in US$)
2006 1,265,500
2007 1,020,000
2008 721,000
TOTAL: 3,006,500
*Not a PEPFAR focus country; above sums represent total allocations to PEPFAR
country programs from bilateral U.S. sources including USAID, Department of
Health and Human Services, Department of Labor, and Department of Defense.
GAVI
Disease Priority Amount Approved (in US$)
Hepatitis B vaccine 1,608,000
Vaccine introduction grant 100,000
Injection Safety 178,000
Immunisation services support 256,000
Health systems strengthening 1,155,000
TOTAL: 3,297,199
Methodology
The Centre for Health System Development in Kyrgyzstan, with the London School of Hygiene and
Tropical Medicine and the Royal College of Surgeons in Ireland, conducted a three-year study
between 2006 and 2008 to track GHIs and their impact on the health system in Kyrgyzstan. The
data used for this case study draws primarily on the findings reported in 2008.
Multiple qualitative and quantitative data collection methods were used. These include: an
analysis of policy and programmatic documents and secondary data; in-depth interviews with
national and sub-national level key informants; structured surveys with HIV/AIDS service providers
and clients; in-depth interviews with clients; and focus groups. The research focused on the effects
of two GHIs in Kyrgyzstan – the Global Fund and CAAP—in three regions: Bishkek/Chui (capital
and surrounding administrative region), Issyk-Kul (north of the country) and Osh/Jalalabad (south
of the country). The following research questions were addressed:
The effects of GHIs on the scale-up of HIV/AIDS services:
• The levels and types of HIV/AIDS services delivered, including prevention, treatment
and support services, with a concern to demonstrate trends over time;
• The perceived quality of HIV/AIDS services;
• Coordinated service delivery.
The effects of GHIs on equitable access to HIV/AIDS services:
• Accessibility and patterns of utilisation of HIV/AIDS services;
• Institutional, household and community factors that determine the accessibility of
HIV/AIDS services.
109

The effects of GHIs on the capacity of health systems:
• National and sub-national HIV/AIDS coordination mechanisms;
• Health workforce.
Data were collected from more than 40 HIV/AIDS service delivery outlets, and approximately 230
interviews were conducted at national level and across the three selected regions.
Results
Leadership and Governance
The study suggests that coordination and cooperation between government and NGO HIV/AIDS
services has improved since the Global Fund and CAAP programmes were introduced. Most
HIV/AIDS service providers participating in the survey indicated that they coordinated their
activities, including client referrals, with other organizations. By 2008, all organizations practiced
client referral, which was practiced extensively between NGOs and government AIDS centres; and
most organizations referred patients to narcology (drug addiction treatment) centres or legal and
support services for people living with HIV or AIDS (PLWHA). Some HIV/AIDS organizations have
signed Memorandums of Understanding formalizing these arrangements. Other forms of
coordination are practiced between HIV/AIDS organizations including: coordinated strategic
planning; information sharing; integrated resources; common protocols; and using a common
monitoring and evaluation system. However, the study also suggests that the practice of these
forms of inter-service coordination is only starting to emerge, rather than already being widely
implemented; indeed, only 23% of clients indicated that they were referred to an HIV/AIDS service
provider by other organizations, and 55% received information about HIV/AIDS through personal
contacts rather than through HIV/AIDS organizations.
National HIV/AIDS coordination council
The coordination of HIV/AIDS activities in Kyrgyzstan has in the past been the responsibility of the
Country Multi-sectoral Coordination Committee (CMCC) for HIV/AIDS, tuberculosis and malaria. In
2005-2007, GHIs played an important role in developing this coordination mechanism; for
instance, the membership and functions of the CMCC were adapted to meet Global Fund
requirements, and the Global Fund CCM formed a sub-committee of the CMCC. In 2006-2007,
interviewees considered the CMCC to have a significant role in HIV/AIDS-related decision-making
at the national level and to be engaged with a wide group of stakeholders.
In August 2007, the CMCC merged with the Republican Special Anti-epidemiological Commission
on Socially Significant and Especially Dangerous Diseases, which focuses on more than 40 different
animal and human diseases. The World Bank CAAP implementers are a member of this body and
participate in meetings. After merging the CMCC, Kyrgyzstan has faced several coordination
problems. These are mainly related to: (1) difficulties in involving all relevant government
departments in implementing HIV/AIDS activities; (2) monitoring activities and sharing
information among stakeholders; (3) ensuring continuity of activities. The organizational structure
of this new CMCC poses a serious obstacle to effective coordination. In particular, the limited
capacity and resources available to the secretariat undermines its functioning, and there are
limited resources available from international donors and initiatives to strengthen the
coordination mechanism.
110

Sub-national HIV/AIDS coordination councils
Oblast (sub-national) Multi-sectoral Coordination Committees (OMCC) were reported by many key
informants participating in the study as being imposed from the national level and as having a
limited role in regional coordination. Sub-national experiences do, however, vary: interviewees
evaluated Osh OMCC as having a relatively substantial role in determining regional HIV/AIDS
policy, whilst coordination in Issyk-Kul was relatively underdeveloped. Barriers to effective
coordination include: high turnover of committee membership; lack of clear working procedures
and lines of accountability; lack of clarity among members about how to implement coordination
efforts; limited civil society representation; and lack of funding for coordination structures.
The Global Fund had an indirect effect on the work of sub-national coordination: roles and
responsibilities were clarified among organizations receiving Global Fund grants, making interagency
coordination more feasible; and the Global Fund grant has been used to finance
strengthening institutional relationships among NGOs, for instance through financing the
establishment of NGO coordination forums.
Financing
HIV/AIDS-related programmes in Kyrgyzstan receive funding from the state budget and
international donor organizations and initiatives. Fifty-one percent of international funds come
from the Global Fund. Since 2004, Kyrgyzstan has implemented a Second Round Global Fund
grant for HIV/AIDS. The amount of funding approved was more than US$ 17 million, and the main
grant recipient was the Kyrgyz Republican AIDS Centre, located in the Ministry of Health (MOH). In
total, 58% of the grant was spent on prevention services, 16% on treatment and 8% on support
services [8].
Kyrgyzstan has been awarded a Round Seven HIV/AIDS grant worth US$ 28.2 million for 2009-
2013. The main objective of this grant is to provide universal access to HIV/AIDS services for
PLWHA and other vulnerable populations. There is also an increased emphasis on improving intersectoral
cooperation to fight HIV/AIDS and on strengthening the capacity of the national health
system, including government and nongovernmental HIV/AIDS services. The development of
HIV/AIDS services is taking place against a background of a shrinking supply of health
professionals in Kyrgyzstan caused by high levels of international migration.
Kyrgyzstan has also received a grant from the World Bank through CAAP to fund a programme of
activities between 2005 and 2010. The total amount of funds distributed among four Central Asian
countries is US$ 25million. The objectives of the programme are to: (1) control the spread of HIV in
Central Asia; (2) establish a regional AIDS Fund as a sustainable financing mechanism during and
after the project; and (3) strengthen cooperation between the state, nongovernmental and private
sectors at both the regional and national levels.
Between 2004 and 2006, Global Fund grant disbursements to sub-recipients increased steadily
from US$ 334 000 to US$ 2 777 000. NGOs are now providing the majority of HIV/AIDS services
funded by the grant; in particular, they focus on HIV prevention. Between 2004 and September
2007, the Global Fund grant supported 102 organizations implementing HIV/AIDS activities,
including 80 nongovernmental, 18 governmental/public and four private organizations. Many
NGOs are, however, becoming increasingly reliant on this grant, which provides a high proportion
of their overall funding. For example, in 2006, 11 out of 16 NGOs surveyed reported that most of
111

their budget was comprised of Global Fund funds. CAAP grants have provided funds to 64
governmental and nongovernmental organizations in four Central Asian countries. HIV/AIDS
organizations in Kyrgyzstan received a total of US$ 138 000.
Many interviewees reported that the distribution of the Global Fund grant reflects the relatively
high organizational capacity of service providers based in the capital Bishkek, rather than HIV
prevalence rates in different parts of the country. In 2007, 53% of NGOs providing HIV/AIDS
services were located in Bishkek/Chui, while the vast majority of PLWHA (62%) lived in southern
Kyrgyzstan in and around the city of Osh. Similarly, CAAP-funded services and activities are also
concentrated in Bishkek.
Health Workforce
The Global Fund grant has led to substantial increases in staff working for NGOs (focusing on
prevention and care/support services), while the number of staff working for government medical
services has remained stable. The greatest growth is among NGO volunteer workers and social
workers. Specifically, peer-to-peer outreach workers are playing an increasingly important role:
they bring knowledge of vulnerable groups and are able to build rapport with clients. The majority
of social workers has not, however, received professional education and requires regular training.
Shortages of psychologists, psychotherapists and lawyers working in the field of HIV/AIDS persist.
The largest increase in staff numbers is observed in the north of Kyrgyzstan: Bishkek city/Chui
province. While increases in staff in Osh have also been substantial, the overall number of HIV/AIDS
workers lags behind Bishkek/Chui. Table 4 summarises the scale-up of HIV/AIDS-related personnel
based on a sample of 24 organizations.
Table 3 Personnel categories in 24 surveyed organizations
Personnel category
Bishkek, Chui
Province Organization type
Osh,
Jalalabat
Issyk-Kul Government NGOs
2006 2007 2006 2007 2006 2007 2006 2007 2006 2007
Doctors 85 97 27 33 19 17 101 111 30 36
Nurses/paramedics 159 139 33 38 7 9 184 167 15 19
Social workers 25 36 19 42 1 3 2 1 43 80
Outreach workers/
“peer to peer”
72 79 18 6 16 16 19 2 87 99
Volunteers 417 557 53 127 26 26 17 40 479 670
Administrative staff 26 35 11 12 5 7 19 21 23 33
Other 12 19 0 4 4 3 3 7 13 19
Total 796 962 161 262 78 81 345 349 690 956
112

The Global Fund has contributed substantially towards capacity building among HIV/AIDS services
in terms of providing training in HIV/AIDS issues and organizational development. Medical workers
are trained in donor blood safety, HIV/AIDS diagnostics, ARV treatment (ART) and the prevention
of mother-to-child transmission (PMTCT). In 2008, 85% of the survey respondents indicated they
had received training, and 62% felt that Global Fund-supported workshops had a positive impact
on their skills.
Service providers in government and nongovernmental HIV/AIDS organizations do not receive
financial incentives for delivering HIV/AIDS services from Global Fund or CAAP grants. Low salaries
and uncertainty about wages among government service providers, as well as interruptions in
Global Fund funding to sub-recipients, have led to low motivation, a deterioration in service
quality (in some cases) and high staff turnover. Staff in NGOs are more motivated to deliver
HIV/AIDS services, compared to staff in governmental organizations: 36% of NGO staff and 8% of
staff working at governmental medical facilities participating in the survey described themselves
as highly motivated to provide services.
Monitoring and Evaluation
Currently there is no common national monitoring and evaluation (M&E) system for HIV/AIDSrelated
programmes and services, although there are some efforts to develop one. The Global
Fund, the CAAP, and other donors do not employ a common system for monitoring the activities
they finance.
Community/Civil Society
Reflecting the increasing GHI funding, the research confirms that there has been an increase in
client numbers for a range of HIV/AIDS-related interventions, including: information/education,
VCT, harm reduction (needle/syringe exchange and substitution therapy) and ART. Coverage of
vulnerable groups, including PLWHA, young people, injecting drug users (IDUs), commercial sex
workers (CSWs), men who have sex with men (MSM) and prisoners/ex-prisoners, has increased
since the inception of the Global Fund (Table 5). New groups, including young people from rural
areas and street children, are now receiving interventions.
Table 4 Number of clients of surveyed organizations: 2004-2006 (N=24)
Previous target groups 2004 2005 2006
Youth 2,527 7,159 21,941
IDUs* 2,982 4,969 8,225
CSWs 2,491 2,549 2,620
MSM** 6,500 7,200 7,500
PLWHA*** **** **** 324
Prisoners No data 3,325 6,500
113

Quality of services
The majority of stakeholders and service providers indicated that the Global Fund and CAAP
grants supported improvements in the quality of HIV/AIDS services in Kyrgyzstan; moreover, most
clients said that they were satisfied with the quality of services (85% of clients participating in the
survey in 2007). These initiatives enabled service providers to enhance service quality due to:
increased staff numbers; better training; and improved provision of key commodities. However,
the study shows that there is room to improve quality in a number of areas including: effective and
confidential VCT services; appropriate information materials for population groups in rural areas,
particularly in the relatively high-prevalence south of Kyrgyzstan; and staff training for substitution
therapy.
Access to HIV/AIDS services
The client survey suggests that clients experience multiple problems accessing Global Fund and
CAAP-financed HIV/AIDS services. The stigmatization of HIV/AIDS was the most important barrier
to using services; using an HIV/AIDS-related service carried the risk of clients becoming known as
HIV-positive, a drug user or sex worker. Indeed, in mid-2007, a hospital outbreak of HIV among
children in the south of the country revealed a range of needs and problems in the field of
HIV/AIDS services delivery and showed that stigma remains high among the population. An
important related problem is the criminalization of injecting drug use, since police frequently
intercept drug users or sex workers, sometimes when they attempt to use an HIV/AIDS-related
service. Other significant access barriers from clients’ perspectives include shortages of medicines
and other commodities; costs of transport and out-of-pocket expenses; and limited knowledge
about HIV/AIDS services and eligibility to use them.
Discussion
The Global Fund grant in Kyrgyzstan has financed substantial scale-up of HIV/AIDS services,
including prevention, testing, treatment, care and support. However, barriers to access remain. In
particular, the stigmatization of PLWHA, the criminalization of drug use, and limited provision of
information about HIV/AIDS services to target groups have undermined efforts to scale-up
HIV/AIDS services. Though organizations providing HIV/AIDS services appear to be collaborating
more effectively following the introduction of the Global Fund and CAAP grants, cooperation is
frequently based on informal personal relationships and agreements rather than formalized
procedures.
GHIs have led to an increase in the number of organizations from different sectors involved
in HIV/AIDS-related activities. Global Fund requirements concerning national coordination
led to improvements in the national coordination structure and participation (2005-2006).
Experiences at the national and sub-national level in Kyrgyzstan suggest that the effective
functioning of multisectoral coordination councils depends on several factors:
1. An effective secretariat with sufficient resources;
2. A national coordination structure that is sufficiently focused to engender high levels of
engagement from all parties;
114

3. Political commitment to HIV/AIDS among decision-makers at the highest government
level;
4. Regular technical assistance for sub-national councils on the coordination of HIV/AIDS
activities;
5. Regular communication between secretariats of sub-national coordination councils
and the country coordination council.
The Global Fund grant has supported the strengthening of the health workforce among NGO subrecipients
delivering HIV/AIDS programmes; there has been a substantial scale-up in the number
of NGO workers. The Global Fund and СААР have also made an important contribution towards
building the capacity of HIV/AIDS service organizations in Kyrgyzstan in terms of financing staff
training. However, the GHIs have not funded staff incentives, which have a strong influence on
motivation, staff turnover and the quality of care.
115

References
[1] Kyrgyzstan at a glance. Washington, DC, World Bank, 2008
(http://devdata.worldbank.org/AAG/kgz_aag.pdf; accessed 20 March 2009).
[2] Human Development Reports: Kyrgyzstan. New York, United Nations Development
Programme, 2008
(http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_KGZ.html; accessed 20
March 2009).
[3] BBC News. Country profile: Kyrgyzstan (http://news.bbc.co.uk/2/hi/asiapacific/country_profiles/1296485.stm;
accessed 5 March 2009).
[4] Hardison C, Fonken P, Chew T, Smith B. The Emergence of Family Medicine in
Kyrgyzstan. Family Medicine. October 2007, 39(9):627-633.
[5] Kyrgyzstan Cooperation Strategy at a glance. Geneva, World Health Organization
(http://www.who.int/countryfocus/cooperation_strategy/ccsbrief_kgz_en.pdf; accessed 20
March 2009).
[6] WHO Statistical Information System (WHOSIS) [online database]. Geneva, World Health
Organization (http://www.who.int/whosis/en/; accessed 20 March 2009).
[7] World Development Indicators (WDI) Online. Washington, DC, World Bank (Proprietary online
database: http://ddp-ext.worldbank.org.ezpprod1.hul.harvard.edu/ext/DDPQQ/member.do?method=getMembers;
accessed 20 March 2009).
[8] Central Asian Regional HIV/AIDS Programme (CARHAP)/UK Department for International
Development (DFID). Analysis of financial deficit in the frame of the HIV&AIDS National Strategy
implementation, 2007 – 2012. Bishkek, CARHAP/DFID, 2008.
116

Malawi: Global Health Initiatives and Delivery of
Health Care: the case of the Global Fund
to Fight AIDS, TB and Malaria
Abstract
Victor Mwapasa 19 , John Kadzandira 17
This study was aimed at assessing the impact of Global Fund-supported activities on the delivery
of non-HIV services; health worker availability, workload, incentives and motivation; and drug
management. Two rounds of data collection took place - in December 2006/January 2007 and
June/July 2008 - at 52 randomly sampled health facilities in nine districts, interviewing 524
respondents, including health service managers and service providers. Records of clients accessing
the services were also collected for the period spanning the last quarter of 2005 to the first quarter
of 2008.
Findings from the study showed a 10% increase in clinical staff in urban and district level hospitals
and a three- to six-fold increase in the numbers of health surveillance assistants in the district and
sub-district facilities. Workload had risen three to five times, resulting in most staff (68%) working
beyond normal hours and facilities resorting to task shifting. No tangible incentives were identified
that could be associated with Global Fund-supported activities. Drug management and processes
of requisition and replenishment had improved, resulting in a reduction of drug stock-outs by 35-
60% between 2006 and 2008. Client volumes for antiretroviral therapy (ART), HIV testing and
counselling (HTC) and prevention of mother-to-child transmission (PMTCT) services rose three to
seven times in the period, but there were no concomitant declines that could be attributed to
Global Fund in client numbers for the non-HIV services.
Background
Malawi is a small landlocked country in Southeast Africa, bordered by Zambia, the United Republic
of Tanzania, and Mozambique. In 2008, Malawi was home to 13.1 million people, with an
estimated 2.8% annual population growth rate [1]. In 2006, Malawi ranked 162 nd out of 179
countries on the UN Human Development Index [2]. Official Development Assistance (ODA) to
Malawi in 2006 was US$ 501 million, while the country’s external debt was US$ 3.4 billion or 26.9%
of gross domestic product (GDP) [3, 4].
The first case of AIDS in Malawi was diagnosed in 1985 [5]. Since then, it has become the leading
cause of death for the country’s most productive age group (15-49 year-olds). In 2007, 930 000
people in Malawi were living with HIV/AIDS [6]. HIV prevalence is estimated at 12%, with higher
prevalence in urban areas (17%) compared to rural areas (11%). The national response to AIDS
dates back to the second half of the 1980s, culminating in the establishment of the National AIDS
19 University of Malawi (College of Medicine and Centre for Social Research). The research for this study was funded by
the Alliance for Health Policy and Systems Research (AHPSR), Geneva. The study summary was produced with the
support of the GHIN Network (www.ghinet.org).
117

Commission (NAC) in 2001 as a multi-sector coordinator of the response. The NAC is responsible
for mobilizing resources, both locally and externally, as well as providing overall coordination and
leadership on behalf of the government and partners.
Table 1 Basic Socioeconomic, Demographic, and Health Indicators*
*Full data sources for all indicators are provided in Annex 1
Indicator Value Year Source
Population (thousands) 13,920 2007 World Bank
Geographic Size (sq. km) 94,080 2007 World Bank
GDP per capita, PPP (constant
2005 international $)
719 2007 World Bank
Gini index 39 2004 World Bank
Government expenditure on
health (% of general
government expenditure)
12.1 2007 WHO NHA
Per capita government
expenditure on health (current
US$)
10
2007
WHO NHA
Physician density (per 10,000)

Table 2 Global Health Initiative Investments*
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 1, HIV/AIDS 342,557,595 193,794,673
Round 2, Malaria 36,773,714 17,957,714
Round 5, HIV/AIDS 17,920,636 7,708,331
Round 5, HSS 22,643,238 17,207,360
Round 7, HIV/AIDS 15,078,417 5,076,095
Round 7, Malaria 36,545,312 14,961,664
Round 7, TB 7,802,037 2,825,106
TOTAL: 479,320,948 259,530,943
PEPFAR§
Year Amount Disbursed (in US$)
2004 14,540,168
2005 15,155,307
2006 16,369,500
2007 18,887,000
2008 23,862,300
TOTAL: 88,814,275
§ Not a PEPFAR focus country; above sums represent total allocations to PEPFAR
country programmes from bilateral U.S. sources including USAID, Department of
Health and Human Services, Department of Labor, and Department of Defense.
GAVI (in US$)
Disease Priority Amount Approved (in US$)
Pentavalent vaccine 100,808,000
Vaccine introduction grant 100,000
Injection Safety 792,175
Immunization services support 3,588,500
Health system strengthening 11,343,000
TOTAL: 116,631,272
World Bank MAP
Project Title FY Approved/Closing Date Commitment (in US$)
Multisectoral AIDS Project 2004/2009 35,000,000
Malawi epitomizes the problem of “brain drain” among health professionals that has affected
many African health care systems. About half of the 248 medical doctors working in Malawi in
2007 were in central hospitals and training/research institutions in urban areas, leaving severe
shortages in rural areas [7]. The government launched a six-year Emergency Human Resources
Plan (EHRP) in 2004 to address its health professional “brain drain” and has used US$ 17.2 million
from the Global Fund for this national effort.
119

Objectives and methodology
The main aim of this study was to assess the impact of Global Fund -supported activities on the
delivery of general health care in Malawi. Specifically, the study sought to correlate trends in the
scale-up of ART, PMTCT and HTC services with trends in other areas of health systems
performance, including: coverage levels of non-HIV programmes (such as antenatal care,
immunization, malaria, TB and family planning); availability of various cadres of health workers;
workload changes; staff incentives and motivation; service integration; and the management of
drugs and medical supplies in health facilities. The results presented below are preliminary and
cover a selected initial subset of these topics.
Two rounds of data collection took place in December 2006/January 2007 and June/July 2008 at
52 health facilities in nine districts. Three of the 52 facilities were central hospitals (one from each
of the three administrative regions of the country), seven were district hospitals and the rest
(n=42) were sub-district facilities from urban and rural areas. Districts, and the sub-district facilities
within them, were sampled at random.
Data collection involved interviews with 524 respondents: nine district managers; 12 nurses in
charge of district hospitals and health centres; 50 coordinators of ART, PMTCT and HTC services;
130 staff working in human resource departments, laboratories and pharmacies managers; and
332 staff delivering the HTC, ART and PMTCT services. Interview data were captured using semistructured
questionnaires, which allowed for both pre-coded responses as well as verbatim
documentation of open-ended responses. Two Research Assistants conducted each interview and
recorded the responses. Records of clients that accessed the services at these facilities between
the last quarter of 2005 and the first quarter of 2008 were also collected.
The Malawi study was conducted jointly by the College of Medicine and the Centre for Social
Research, both of the University of Malawi. Ethical approval was sought from the College of
Medicine’s Research and Ethics Committee and the office of the University Coordinator at the
University Offices, while a formal approval to collect data from the health facilities was provided by
the MOH. All approvals were granted prior to commencement of data collection.
Results
Health Workforce
The study showed modest increases in numbers of clinical staff. An increase in nurses was found
mainly in urban areas and at district hospitals, where a 10% increase was observed. A three- to sixfold
increase in the numbers of health surveillance assistants (HSAs) was observed, mainly in the
district and sub-district facilities. Workload (determined by client volumes per health worker) had
risen three to five times in the period under observation, mainly due to HIV-related services. On
average, nurses in sub-district facilities were each attending 80 patients in the general outpatient
clinic (range 12 to 162), 87 patients in the under-five clinic (range 49 to 268) and six new antenatal
mothers (range 2 to 23).
Qualitative data from interviews with the service managers and providers also showed a general
feeling that workload was very high, resulting in most staff (68%) working beyond normal hours,
especially in rural areas and during the rainy season. Task shifting (across staff cadres and days of
the week) was the most common strategy that was being used to address high workload in the
sampled facilities. The study found that HSAs and dedicated counsellors constituted 79% of all
120

HTC service providers in the sampled facilities, 11% of ART and 13% of PMTCT providers, which
was a shift from the situation observed in 2006 (21% for HTC,

Discussion
The findings of the Malawi study show that increased funding for HIV has been associated with a
mix of both positive and negative effects on the wider health system. Overall, GHIs (mainly the
Global Fund in Malawi) are seen as having had a positive effect on health systems in the country.
Key targeted services, especially ART, are being delivered at three to seven times the coverage
levels observed in the baseline survey in 2006/07. An important early observation is a marginally
greater increase in clients attending sub-district facilities, relative to district and central hospitals.
This is expected to reduce the congestion in the district and central hospitals in the long run,
thereby freeing these facilities to concentrate on the provision of secondary and tertiary care.
Considering that the Global Fund provides more than 70% of the funds for the national response
to HIV and AIDS, the scale-up being observed in the sampled facilities (and at national level) can be
substantially attributed to the Global Fund.
With the support of the Global Fund, DFID and other donors, Malawi is implementing an
emergency human resource programme, which supports capacity building of training institutions
to produce more workers. Training of clinicians and nurses takes time (three to six years,
depending on the cadres). Therefore, the effects of such an investment have not yet been seen.
This may partly explain why the Malawi study did not find significant increases in the numbers of
clinicians and nurses in the study period. The three- to six-fold increases in the numbers of HSAs
observed in the period is likely to have resulted from the recruitment exercise that took place in
2007 with the support of the Global Fund, as one way to close the gap in human resources
required for service scale-up.
Despite shifting tasks from clinicians/nurses to HSAs/counsellors in the facilities surveyed,
workload is still considered high. Although not entirely attributable to HIV-related services, scaleup
of the services has not been properly matched with staffing additions. As a result, the process is
contributing to further overstretching of the existing staff. This appears to be happening in the
absence of incentives that could be associated with Global Fund -supported activities.
Recommendations for maximizing GHI-health systems synergies in Malawi include: (1) continue to
increase and strengthen the HTC, ART and PMTCT sites that are opening up in rural and urban
areas so that they absorb the majority of the clients. This would reduce client volumes in district
and central hospitals, enabling them to concentrate on secondary and tertiary care; (2) monitor
the task-shifting processes that are taking place in order to ensure quality of care for HIV services
and to observe trends in primary health care services for which HSAs are primarily responsible; (3)
develop mechanisms to remunerate staff partaking in task-shifting and to ensure that both newly
trained and veteran clinical staff are sufficiently motivated and are fairly distributed amongst rural
and urban areas.
122

References
[1] Government of Malawi, National Statistical Office. Population and Housing Census (2008):
Preliminary Report. Zomba, National Statistical Office, 2008 (www.nso.malawi.net; accessed 20
March 2009)
[2] Human Development Report: Malawi. New York, United Nations Development Programme,
2008 (http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_MWI.html; accessed
20 March, 2009).
[3] Malawi at a glance. Washington, DC, The World Bank, 2008
(http://devdata.worldbank.org/AAG/mwi_aag.pdf; accessed 20 March 2009).
[4] World Development Indicators (WDI) Online. Washington, DC, The World Bank, 2007
(Proprietary online database: http://ddp-ext.worldbank.org.ezpprod1.hul.harvard.edu/ext/DDPQQ/member.do?method=getMembers;
accessed 20 March 2009)
[5] Government of Malawi, Office of the President and Cabinet. Malawi HIV and AIDS Monitoring
and Evaluation Report. Lilongwe, Office of the President and Cabinet, 2007.
[6] 2008 Report on the global AIDS epidemic. Geneva, UNAIDS, 2008
(http://www.unaids.org/en/KnowledgeCentre/HIVData/GlobalReport/2008/default.asp; accessed
20 March 2009).
[7] Kadzandira JM et al. Malawi Health Sector Employee Census 2007. Lilongwe, Centre for Social
Research, University of Malawi & Ministry of Health and Population, 2007.
123

Pakistan: The impact of the Global Polio
Eradication Initiative on health systems
Abstract
Sania Nishtar 20
Pakistan is one of the four countries in the world where polio eradication has not been achieved,
despite an unprecedented inflow of resources dedicated to the programme over the last 15 years.
The objective of this case study is to analyse the impact of Pakistan’s Polio Eradication Initiative
(PEI) on health systems in Pakistan and to document the impact of the health system on polio
eradication. Qualitative methods were used for the study.
By and large, the impact of the PEI on health systems in Pakistan was found to be positive, as
evidenced by a decline in poliovirus transmission. The programme can also be credited with
establishing a sensitive nationwide disease surveillance system, building capacity for door-to-door
outreach, and achieving the largest-ever funding allocations for a disease-specific programme in
the country. The PEI’s negative effects include the time implications of organizing and
implementing Supplementary Immunization Activities (SIAs) and reported abuse of the mode of
payment used to conduct SIAs. The failure to achieve the polio eradication goal stems from
weaknesses in Pakistan’s health systems and the country’s geo-political, security and broader
governance challenges. Salient considerations include severe conflict-related limitations in
reaching children in the tribal areas of the country, along with organizational, managerial and
operational problems in achieving sufficient coverage, even in accessible areas. In many ways, the
PEI has exposed critical weaknesses in Pakistan’s health system and its inability to deliver on
targets, even when resource availability is not an issue.
Background: Pakistan and its health systems
Pakistan is the sixth most populous country in the world, with its current population estimated at
160 million. Sixty-five percent of the population lives in rural areas [1], 35% is below the age of 15
years, and more than 46% is illiterate [2,3,4].
Pakistan has been under military rule for 31 out of the 62 years since the country came into
existence. Macroeconomic growth has been significantly dependent on the level of international
support, as opposed to the mode of governance (democracy vs. military rule), with periods of
growth in the 1960s, 1980s and more recently during the period 2001-07. Pakistan’s
macroeconomic downturn from 2007 onwards can largely be attributed to geo-political and
security factors, weaknesses in governance and the impact of the global commodity crisis. The
financial crisis of 2008 did not permeate into Pakistan as it did in most East Asian countries
because of the lack of integration of Pakistan’s financial markets with the global financial system.
However, effects of the crisis are expected in Pakistan over the coming months and years. Despite
20 Heartfile
124

variations in the economic growth rate, there has been very little change in government
expenditure on health as a percentage of gross domestic product (GDP), which has ranged from
0.51% to 0.8% throughout the past 61 years [5,6].
Health is regarded as a fundamental human right under Pakistan’s Constitution [7]. Institutionally,
the Ministry of Health is responsible for policy-making, coordination and technical support. The
provincial mandate is to oversee the provision of health services. In Pakistan’s decentralized
system of government, districts are responsible for the delivery of health services [8]. A number of
vertical health systems in the country finance and provide/produce services for defined
populations. These systems have mutually exclusive governance and financing arrangements and
means of harnessing inputs. They include the healthcare system of the armed forces, the
Employees Social Security Institute, and the Fauji Foundation System. Together, these vertical
systems cover 10.29% of the population. A much wider public system and the market system run
in parallel and provide services to 89.71% of Pakistan’s population [9]. Notwithstanding frequent
blurring of lines between the two—healthcare providers in the public system routinely also
engage in private practice—the market system is predominant; more than 70% of the population
accesses private providers for healthcare [10], and out-of-pocket payments account for 57% of
healthcare financing [9].
Revenues and development allocations finance the public system, which comprises the National
Public Health Programmes and three tiers of service delivery. The former, a set of federally-led,
vertical public health programmes, are characterized by Federal government leadership as the
common denominator. These programmes have varying levels of implementation autonomy
within provinces and districts. Some of the programmes are disease-specific, such as those on
HIV/AIDS, malaria, TB and hepatitis, whereas others are cross-cutting, including the National
Expanded Programme for Immunization and the Maternal and Child Health Programme [11].
Physical infrastructure for primary health care is comprised of many categories of over 12,000 First
Level Care Facilities (FLCFs). These deliver basic clinical services. Basic Health Units (BHUs) are also
meant to deliver outreach services and serve as the implementation arms of the National Public
Health Programmes. More than 30% of FLCFs are currently non-functional, despite efforts to
restructure management of BHUs. More than 90,000 female Lady Health Workers provide
preventive, maternal and child health and related family planning services to 55% of Pakistan’s
population at the grassroots level in rural areas. Family planning services are also provided
through Family Welfare Clinics run by the Ministry of Population Welfare [12]. In addition, the PEI
leverages thousands of volunteers to deliver services during national immunization days. The
extensive private health sector is heterogeneous in terms of the qualifications of healthcare
providers, the system of medicine followed, the registration status of providers and the length of
time for which providers practice.
Health status
A recent compendium of health statistics in Pakistan concludes: “Although there have been some
improvements in the health status of the Pakistani population over the last 60 years, key health
indicators lag behind in relation to international targets articulated in the Millennium Declaration
and in comparison to averages for low-income countries....” [13] Findings from a recently
concluded household survey, the largest ever conducted in Pakistan, support this notion—the
reported Maternal Mortality Ratio of 276 maternal deaths per 100,000 live births is high by
developed country standards [14]. Although there has been a decline in Total Fertility Rate (TFR)
from 5.4 children per woman in 1990-91 to 4.1 children in 2006-07, TFR remains high by
125

international standards. The Under Five Mortality Rate has declined from 117 per 1,000 live births
in 1986-90 to 94 in 2002-06, and an upward trend in child immunization has been reported—
increasing from 35% in 1990-91 to 47% in 2006-07. Nevertheless, rates of improvements in health
outcomes have been slow, particularly in comparison with other countries in the region [15].
Table 1. Pakistan’s Key Health indicators (2005-2006)
Health Indicators
Life expectancy (years)* 63.8 a
Dependency ratio (percent)* 68.7 a
Crude Birth Rate (per 1,000)* 26.1 b
Crude Death Rate (per 1,000)* 8.2 b
Total Fertility Rate (children per woman) 4.1 c
Contraceptive Prevalence Rate (percent) 39.0 c
Pregnant women receiving at least one ante-natal consultation (percent) 61.0 c
Women who receive care from Skilled Birth Attendants (percent) 39.0 c
Neonatal Mortality Rate (per 1,000) 54 c
Infant Mortality Rate (per 1,000) 78 c
Under-Five Mortality Rate (per 1,000) 94 c
Fully immunized children (percent) 47 c
Tuberculosis Case Detection Rate 62 d
Tuberculosis Case Detection Rate for new Sputum Smear Positive Cases 49 d
Tuberculosis Treatment Success Rate 84 d
Confirmed number of polio cases (as of May 13, 2009) 12 e
Prevalence of viral hepatitis in the general population (percent) 7.4 f
Prevalence of smoking (over 18 years of age, percent) 15.75 g
Prevalence of leisure time physical inactivity (over 18 years of age, percent) 91.5 g
Prevalence of overweight and obesity (over 18 years of age, percent) 38.5 g
Prevalence of central obesity (over 18 years of age, percent) 48.35 g
Prevalence of high blood pressure (over 18 years of age, percent) 13.7 h
Prevalence of diabetes (percent) 7.65 i
Doctor-population ratio 1326 g
Nurse-population ratio 22,662 g
Dentist-population ratio 3,039 g
Households with toilets systems (percent) 74 j
Households with government garbage disposal services (percent) 37 j
*estimations
a. Federal Bureau of Statistics, Pakistan. Pakistan Demographic Survey, 1998-2003
b. Federal Bureau of Statistics, Pakistan. Pakistan Population Census, 1998
c. National Institute of Population studies and Macro International Inc. Pakistan Demographic and Health Survey,
2006-07
d. National TB Control Programme, Ministry of Health, Pakistan
e. National Expanded Programme on Immunization, Ministry of Health, Pakistan
f. Pakistan Medical Research Council, Ministry of Health, Pakistan; Unpublished data
g. Heartfile, Ministry of Health and Federal Bureau of Statistics. Gateway Paper II: Health Indicators of Pakistan, 2007
h. Pakistan Medical Research Council. National Health Survey of Pakistan, 1994
i. Diabetic Association of Karachi and WHO surveys, 1994-1998
j. Social Audit of Governance and Delivery of Public services Pakistan, 2005
126

Pakistan also suffers from a number of other health challenges, particularly in relation to polio
eradication and the prevention and control of non-communicable diseases, viral hepatitis and
HIV/AIDS. Pakistan’s key health indicators have been summarized in Table 1.
Health is an inter-sectoral responsibility. A number of factors are therefore responsible for poor
health status of the country’s population; these include broader issues implicit in the social
determinants, low public investments in the health sector and poor performance of the health
system. The latter is the result of interplay between insufficient state funding for health, a
regulatory environment which enables the private sector to operate unchecked and unregulated
in the delivery of social services, and the overall lack of transparency in governance [16]. Political
factors, conflict and disaster further worsen the situation, particularly with reference to the case
study that is described below.
Objective and methodology
The objective of the case study was to analyse the impact of the Polio Eradication Initiative (PEI)—
a part of the Global Polio Eradication Initiative—on health systems in Pakistan. However, the study
also ended up documenting the impact of the health system on polio eradication.
Qualitative methods were used for the study. These included a review of academic and grey
literature and semi-structured interviews and focus group discussions with key informants. A
Medline search was conducted with the search terms “polio eradication” and “PEI Pakistan” in
combination with search terms “health systems”, “information systems”, “workforce”, “service
delivery”, “financing”, “supply chains” and “governance”. In order to review grey literature, the
same search terms were entered into Google.
Semi-structured in-depth interviews varied in length. These were conducted with managers at
various levels within the public and private sectors. A total of 27 interviews were conducted; the
professional profile of the interviewees ranged from government clerical staff and equivalent
private sector employees to Chief Executive Officers and former cabinet members. Gender, culture,
language and social considerations were taken into account when planning interviews. The
Principal Investigator also conducted focus group discussions with 5 to 10 people on particular
issues. A total of five focus group discussions were held.
Results and analysis
The PEI in Pakistan: background
The PEI was launched in Pakistan in 1994, fifteen years after the launch of the Expanded
Programme for Immunization (EPI); the latter is mandated to undertake routine immunization
activities. Since 2000, the PEI has followed the successful approach of the western hemisphere—
supplementing routine polio immunization with huge country-wide campaigns several times a
year to deliver drops of Oral Polio Vaccine (OPV) to every child under the age of five years. Over the
last nine years, 88 rounds of SIAs have been conducted with nationwide outreach, with the help of
contributions from the Global Polio Eradication Initiative—a global partnership of WHO, UNICEF,
127

the US Centers for Disease Control and Prevention (CDC), Rotary International and other major
bilateral and private donors.
Initial success of the PEI in Pakistan was remarkable. The number of laboratory confirmed cases of
poliomyelitis countrywide declined from 1155 in 1997 to 28 in 2005—the lowest ever recorded for
a year. A very sensitive nationwide reporting system was built up to assure the detection of all
remaining polio cases. The system captures all children under 15 with acute onset flaccid paralysis
(AFP) and includes subsequent laboratory testing of stool specimens [17]. However, since 2008,
there has been a marked resurgence of polio cases, both in aggregate terms and in relation to
geographic spread (Figures 1 and 2). In 2008, eight cases were reported in Punjab, the largest
Pakistani province in the country, home to more than 60% of the country’s population. No cases
had been reported there in 2007. The determinants of this failure are discussed in a subsequent
section.
Figure 1: Number of confirmed cases of poliomyelitis in Pakistan (1997-2006)
Figure 2. Number of districts with confirmed poliomyelitis cases in Pakistan (1997-2006)
Source for Figures 1 and 2: National Surveillance Cell, Expanded Programme on Immunization, Ministry of Health,
Government of Pakistan, Islamabad
128

Impact of the PEI on the health system
Notwithstanding the failure to achieve the polio eradication goal, the Polio Eradication Initiative,
as a GHI, has had a positive impact on Pakistan’s health system in several areas.
Funding for the PEI represents the largest-ever allocation for a disease-specific programme in
Pakistan. Official Development Assistance (ODA) accounts for a major proportion of the funds for
Pakistan’s PEI. Donors contribute an average of US$ 40 million annually towards the initiative. A
significant proportion of polio resources are not delivered through conventional channels and are
therefore not accounted for when calculating health expenditure.
PEI financing is the result of effective coordination between multilateral and bilateral agencies and
the government. Key stakeholders in financing the PEI include WHO, UNICEF, the World Bank, the
UK Department for International Development (DFID) and USAID. WHO supports AFP surveillance
activities and the operational costs of the SIAs. UNICEF supports social mobilization and vaccine
procurement, whereas the World Bank, Japan’s International Cooperation Agency (JICA) and
USAID provide funds for vaccine procurement.
The most significant contribution of the PEI at the outcome level is a major reduction of wild
poliovirus transmission, moving towards the ultimate goal of eradication. An important spin-off
effect is the establishment of the Polio Surveillance system, which is sensitive enough to detect
every case of polio. This is an active surveillance system where each case is actively searched out in
the community; as such it can only be used for diseases marked for elimination or in outbreak and
emergency situations. However, many of the components and attributes of the system are also
features of other surveillance systems. These include: active surveillance visits at health facilities,
laboratory specimen collection, transport and laboratory testing, channels of data reporting,
analysis and dissemination, use of indicators to measure surveillance quality, and training and
capacity-building of staff, among others. These attributes are of relevance to broader capacity
building in the wake of the threat posed by emerging infections in Pakistan, for example with the
entrenchment of avian flu. Recent developments around avian flu include evidence of cases in the
“poultry belt” of the Northwestern Frontier Province (NWFP) and a documented chain of humanto-human
transmission, which was luckily un-sustained [18]. To date, however, resources of the PEI
have not contributed to strengthening other health information systems, nor have they been
strategically used to consolidate country capacity for generating and utilizing evidence.
Pakistan’s PEI has also supported the creation of a grassroots workforce, which has capacity to
deliver services door-to-door throughout the country. The creation of this capability, by itself,
constitutes a significant outreach capacity—one that can be leveraged for episodic nation-wide
activities. However, effective harnessing of this capability is dependent on the capacity of
monitoring and oversight agencies, which is where the actual gap lies. The incentive structures
created for this workforce have been criticized as undermining the functioning of routine
immunization. However, gaps in the performance of the routine immunization programme may be
more directly attributable to issues inherent to Pakistan’s health system, as described below; in
comparison, the role played by incentive structures is marginal. In addition, the availability of
expanded resources through the PEI has improved infrastructure in terms of cold chain equipment
and transport.
129

Some negative effects are also attributable to the PEI: the most important are the time implications
of organizing and implementing the supplementary immunization activities (SIAs) and the manner
in which these divert the attention of district administrative and service delivery workforce from
routine responsibilities. In addition, district authorities are often reported to abuse the mode of
payment used to conduct SIAs.
Determinants of failure to achieve the eradication goal
Recent resurgence of polio in Pakistan coincides with political turmoil and a worsening law and
order situation in many parts of the country in recent years. More than 12% of the country’s
territory in the Federally Administrated Tribal Areas (FATA) and in the northern zones of the NWFP
province has been in the grip of conflict for some time now. Recent escalation of violence has
rendered this territory outside of the remit of the PEI. Talibanization and the unfortunate
misconstrued interpretation of religion has led the clergy in these areas to campaign against polio
vaccination on a wide scale, effectively orchestrating refusal by parents to vaccinate children on
the mistaken grounds that vaccination is Haram [forbidden] by the religion. With the spreading
wave of Talibanization outside FATA to areas in the direct control of the state, this unfortunate
notion has also spread to many other conflict-ridden parts of the country. Cross-border movement
of nomadic populations between Afghanistan and Pakistan compounds these factors.
The resurgence of polio in areas of the country far removed from the western border, such as in
Punjab province, indicates that weaknesses in the delivery of services and broader issues of
governance at a health systems level are also a major factor in the failure to achieve the polio
eradication goal. In Pakistan’s mixed health system [16], the triad of insufficient funding for the
public sector, a poorly regulated private sector and lack of transparency in governance act together to
compromise the quality of public services and defeat the equity objective through a number of
mechanisms, as illustrated in Figure 3. As a result, forms of institutionalized malpractice, primarily
geared to pilfering resources from the system, become ingrained. Collusion between service
delivery staff and inspectors fosters deliberate inattention to staff misconduct. Consequently, staff
remain absent from duty, do not run field operations and pilfer vaccines for use in private facilities.
Service delivery is undermined both qualitatively and quantitatively, and charges are levied for
services that are supposed to be provided for free. Vaccinators may also engage in petty thefts in
the field at various levels—by charging money for vaccination cards and syringes that are
disseminated to them free and by selling part of the vaccine stock to private hospitals. The recent
Open-Vial Policy, where the field vaccinator has the prerogative to open the multi-dose vial even
for one child, in order to maximize vaccination coverage, is being particularly abused in this
connection. These issues, which are already prevalent in the system, have grown more serious in
recent months due to worsening governance, notably preferential treatment in staff deployment
and rapid turnover of programme managers.
130

Figure 3: The three governance-related determinants of weaknesses in mixed health systems
Less funds for the social sector
Clouding of Business
Environment
Collusion in
contracting and
Unofficial economy
L
Low investor
confidence
Predominant role of
the private sector
Lack of
transparency
and poor
implementation
of regulations
Dual job holding & deterioration
of public infrastructure
Purchasing
public
positions
State
capture by
the
corporate
Unethical
practices in
marketing
Misappropriati
on of talent
Selective
benefits
High costs of
care
Source: Nishtar S. Politics of health systems: WHO’s new frontier. Lancet 2007;370(9591):935-6.
Low quality of public
services
Equity objective
defeated
The determinants of failure to achieve polio eradication also demonstrate the importance of other
determinants outside of the health sector in influencing health status. Weaknesses in other state
institutional processes can impact polio control efforts. As a result of poor sanitation and lack of
clean water, diarrheal diseases are the third commonest cause of death in children [14] and can
interfere with the uptake of polio vaccine. With an average of eight hours of electricity
loadshedding, maintaining the cold chain may become a problem, although there is as yet no
direct evidence of an impact on PEI results.
Although all these factors may be significant, by far the most important reason for failure to break
polio transmission in Pakistan is “failure to vaccinate” during SIAs: SIA coverage in critical areas is
not high enough to bring children in these areas to a level of immunity sufficient to interrupt
transmission, and then to maintain that achievement. In sum, two factors stand out as
determinants of failure: (1) severe conflict-related limitations in reaching children in the tribal areas
of NWFP: and (2) organizational, managerial and operational problems in achieving sufficient
coverage, even in accessible areas.
Conclusion
By and large, the impact of the PEI on health systems in Pakistan has been positive. The PEI can be
credited with making OPV available for each of the 33 million children in the country over the last
nine years, creating a sensitive disease surveillance system and building capacity for door-to-door
service delivery outreach in the country. The inflow of resources directly attributable to the
programme has been unprecedented.
Despite this, polio eradication as a goal has not been achieved. The reasons for this failure include
factors stemming from weaknesses in Pakistan’s health systems and the country’s geo-political,
security and broader governance challenges. In many ways, a relatively well-resourced initiative
such as the PEI has exposed critical weaknesses in Pakistan’s health system and its inability to
deliver on targets, even when resource availability is not an issue.
The case of polio eradication also highlights the importance of factors outside of the health sector
in influencing health status. These issues help to demonstrate that, whereas GHIs can bring value
131

to country health systems by way of increasing fiscal and technical inputs and better coordination,
and can increase outputs and expedite processes, their ultimate impact in reducing mortality and
morbidity is deeply dependent on the performance of existing health systems, overall
effectiveness of governance in a country and prevailing macro-economic and geo-strategic
stability in the environment. Before labelling a GHI as unsuccessful in achieving its objectives, the
impact of these factors should be carefully assessed.
132

References
[1] Ministry of Finance, Finance Division, Government of Pakistan. Pakistan Economic Survey, 2007.
Islamabad, Government of Pakistan, 2008 (http://finance.gov.pk/finance_economic_survey.aspx; accessed
30 May 2008).
[2] Federal Bureau of Statistics, Government of Pakistan. National Education Census, 2005. Islamabad,
Government of Pakistan, 2005.
[3] Population Census Organization, Statistics Division, Ministry of Economic Affairs and Statistics,
Government of Pakistan. Population Census, 1998. Islamabad, Government of Pakistan, 1998
(http://www.statpak.gov.pk/depts/pco/statistics/statistics.html; accessed 6 May 2008).
[4] A literate person has been defined as someone who can read a newspaper and write a simple letter in any
language.
[5] Federal Bureau of Statistics, Statistics Division. 50 years of Pakistan. Islamabad, Government of Pakistan,
1998.
[6] State Bank of Pakistan. Annual Report of the State Bank of Pakistan, 2007. Islamabad, Government of
Pakistan, 2008 (www.sbp.org.pk/reports/annual/arfy07/index.htm; accessed 2 January 2009).
[7] Ministry of Law and Parliamentary Affairs, Justice Division. The Constitution of the Islamic Republic of
Pakistan (as modified up to March 1987). Islamabad, Government of Pakistan, 1987
(www.pakistani.org/pakistan/constitution/; accessed 30 May 2008).
[8] Government of Pakistan. Ordinance Number VI of 2001. Local Government Ordinance, 2001.
(www.nrb.gov.pk/publications/SBNP_Local_Govt_Ordinance_2001.pdf ; accessed 30 May 2008).
[9] Nishtar S. Choked Pipes: reforming Pakistan’s mixed health system. Gateway Paper III. Islamabad,
Heartfile, 2009. Forthcoming.
[10] Federal Bureau of Statistics, Statistics Division, Government of Pakistan. Pakistan Social and Living
Standards Measurement Survey (PSLMS), 2007-08. Islamabad, Government of Pakistan, 2008.
[11] Ministry of Health, Government of Pakistan (http://202.83.164.26/wps/portal/Moh; accessed 11 March
2009).
[12] Ministry of Population Welfare, Government of Pakistan (www.mopw.gov.pk/; accessed 10 September
2008).
[13] Nishtar S. Health Indicators of Pakistan – Gateway Paper II. Islamabad, Heartfile, Health Policy Forum,
Statistics Division, Government of Pakistan, World Health Organization, Ministry of Health, Government of
Pakistan, 2007.
[14] National Institute of Population Studies and Macro International Inc. Pakistan Demographic and Health
Survey, 2006-07. Islamabad, National Institute of Population Studies and Macro International Inc, 2008.
[15] Mahbub-ul-Haq Human Development Centre. Human Development in South Asia 2007. Islamabad,
Oxford University Press, 2008.
[16] Nishtar S. Politics of health systems: WHO’s new frontier. Lancet 2007;370(9591):935-6.
[17] National Surveillance Cell, Expanded Programme on Immunization, Ministry of Health, Government of
Pakistan, Islamabad, May 2006.
[18] World Health Organization. Human cases of Avian influenza-A (H5N1) in NWFP, Pakistan during the
period October-November 2007. Weekly Epidemiological Records, October 3, 2008
133

Peru: Effects of the Implementation of Global
Fund-supported HIV/AIDS Projects on
Health Systems, Civil Society and
Affected Communities, 2004-2007
Abstract
Carlos F. Cáceres, Maziel Girón, Clara Sandoval, Roberto López, José Pajuelo, Rocío
Valverde, Patricia Vásquez, Alfonso Silva-Santisteban, Ana Maria Rosasco 21
Because the Global Fund grants for HIV-related projects in Peru are substantial, they must be
closely monitored in order to ensure reasonable impact. This paper describes the effects of such
monitoring and support on key actors involved, decision-making processes, health sector
divisions, policies and funding sources, equity of access, and discrimination of vulnerable and
affected populations. Data were collected through interviews with key informants, discussion
groups with affected populations, and a review of secondary data.
Multisectorality, encouraged by the Global Fund is in its initial stages with centralist proposals with
limited consultation, a lack of consensus and short preparation times, prevailing. No effective
accountability mechanisms operate within the Country Coordinating Mechanism (CCM). Global
Fund-funded activities have required significant input from the public sector, sometimes beyond
the capacity of its human resources. A significant increase in HIV funding, in absolute amounts and
in fractions of the total budget, has been observed from several sources including the National
Treasury, but it is unclear whether this has reduced the budget for other priorities. Patterns of
social exclusion of people living with HIV/AIDS (PLWHA) are diverse: children and women are less
excluded, while transgender persons and sex workers are often more excluded.
Background
Peru is bordered by Ecuador and Colombia to the north, Brazil on the east, Bolivia on the
southeast, Chile on the south, and by the Pacific Ocean on the west. It is home to 27.9 million
people and averaged a 1.2% annual population growth rate between 2001 and 2007 [1]. Peru
ranked 79 th out of 179 countries on the UN Human Development Index in 2006 [2].
Official Development Assistance to Peru was US$ 467.9 million in 2006 [3]. Peru remains heavily
indebted but has been excluded from international debt relief agreements. The country’s total
external debt was US$ 28.17 billion or 30.2% of GDP [1, 3].
As of 2006, a considerable number of Peruvians (about 25%) faced severe constraints in access to
health care, in large part due to the high cost of health care [4]. In 2006, out-of-pocket
21 All authors are affiliated to the Health, Sexuality and Human Development Unit at Cayetano Heredia University School
of Public Health. The research for this study summary was funded by The Alliance for Health Policy and Systems
Research, based at the World Health Organization. The study summary was produced with the support of the Global
HIV/AIDS Initiatives Network (GHIN; www.ghinet.org).
134

expenditures accounted for one-third and external resources accounted for 1.6% of total
expenditure on health [5]. With funding from the World Bank, the Peruvian government has been
attempting to improve the equity of and access to health care services for its poor rural population
since 2000 [6].
The majority of reported HIV/AIDS cases are concentrated in Peru’s large urban areas, and sexual
transmission accounted for 97% of cases in 2007 [7]. That same year, there were 76 000 people
living with HIV/AIDS in Peru [8].
Despite a relatively low HIV prevalence in the general population, Peru’s HIV epidemic has
mobilized social actors, economic resources, and the Peruvian health sector [9,10]. In recent years,
the country has become the largest recipient of HIV funding from the Global Fund in Latin
America, with over US$ 77 million received or committed for projects implemented between 2004
and 2012 [11].
This magnitude of investment in a middle-income country with a concentrated epidemic is
significant, and it will likely have an impact on the relationships among the actors involved, the
organization of the response to the epidemic, the redistribution of public resources, and the
quality of care offered to PLWHA. Analysis of its effects on the health sector, civil society and
affected communities may provide significant lessons at multiple levels.
Table 1 Basic Socioeconomic, Demographic, and Health Indicators*
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 27,898 2007 World Bank
Geographic Size (sq. km) 1,280,000 2007 World Bank
GDP per capita, PPP (constant 2005
international $)
7,403 2007 World Bank
Gini index 49.55 2006 World Bank
Government expenditure on health (%
general government expenditure)
13.8 2007 WHO NHA
Per capita government expenditure (current
US$)
99 2007 WHO NHA
Physician density (per 10,000) 12 1999 WHO SIS
Nursing and midwifery density (per 10,000) 7 1999 WHO SIS
Maternal mortality ratio (per 100,000 live
births)
240 2005 WHO SIS
DTP3 coverage (%) 80 2007 WHO SIS
Estimated adult HIV (15-49) prevalence (%) 0.5 (0.3-0.6) 2007 UNAIDS
Estimated antiretroviral therapy coverage (%) 48 (36-62) 2007 WHO/UNAIDS/UNICEF
Tuberculosis prevalence (per 100,000) 136 2007 WHO GTD
Estimated malaria deaths 128 2006 WHO WMR
135

Table 2 Global Health Initiative Investments*
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 2, HIV/AIDS 22,166,497 21,619,940
Round 2, TB 25,552,603 25,198,382
Round 5, HIV/AIDS 12,867,465 8,700,781
Round 5, TB 32,306,243 12,814,527
Round 6, HIV/AIDS 24,153,662 15,140,430
Round 8, TB 15,178,076 0
TOTAL: 132,224,546 83,473,060
Objectives and Methodology
This study aims to assess the impact of interaction among Peruvian HIV/AIDS stakeholders on the
development and submission of HIV/AIDS projects, as well as on the implementation of such
projects upon approval. Analysis during the first phase of the study, presented here, was focused
on the following objectives: I – Identify the effects that participation in the Global Fund processes
has had on institutional actors involved in work on HIV/AIDS, and on the interactions among those
actors in policy decision-making and project implementation; II – Assess the effects of the
interaction with the Global Fund on the structure and functioning of the Ministry of Health (MOH);
III – Understand the impact of access to these funds upon public and private resources and
policies to fund the response to the AIDS epidemic; and IV – Assess equity in access to project
benefits, and impact on stigma and discrimination affecting PLWHA and vulnerable groups.
The study was implemented between September 2006 and February 2009. Data collection
measures included interviews with key informants 22 ; in-depth interviews and focus groups with
vulnerable and affected populations; a review of secondary data from the health sector and
nongovernmental organizations (NGOs); analysis of existing databases; analysis of a package of
communications materials prepared as part of Global Fund activities; and a review of previous
studies (including baseline and follow-up studies measuring indicators established by the Global
Fund projects). For each component, analysis was guided by four to five “best-case scenario”
hypotheses that formulated ideal conditions to which empirical data could be compared.
Results
Leadership and Governance
Although Peru’s Country Coordinating Mechanism (CCM) is responsible for implementing the
country’s National Multisectoral Strategic Plan (MSP), the multisectoral character of the CCM was
still evolving in 2006-2009 [12]. Various interpretations of its meaning existed among stakeholders,
and achieving consensus on the roles of different actors within the CCM was difficult, given the
lack of other truly multisectoral experiences in the past. Important absences were apparent early
on. For example, representatives of vulnerable communities were notably lacking. Commitment of
22 Key informants include: officials in the public sector and in cooperation agencies, organizations of people living with
HIV/AIDS, NGOs, churches, academics, the Principal Recipient (PR) and Sub-Recipients (SR)
136

sectors other than health was weak, largely because the epidemic was understood mainly as a
“health sector problem.” The CCM needed reinforcement through a stronger representation of
organizations participating in the national HIV/AIDS response, including public, civil society and
regional organizations.
Despite efforts to include varied actors in Peru’s HIV/AIDS response, a centralist view was
established within the CCM during the body’s first years; efforts were focused on Lima, where the
epidemic was most apparent, and many regions were left out of the national response. In an effort
to address this problem, the proposal submitted to the Global Fund sixth round took into account
the regions and created Regional Coordination Mechanisms (RCMs) to decentralize decisionmaking
and take action according to regional HIV/AIDS needs. No formula was created to reconcile
the regional interests with the interests of the National Response as a whole. RCMs continue to
operate in a legal vacuum, and no channels of responsibility have been established between the
CCM and the RCMs.
Peru’s CCM exhibits a number of notable shortcomings. There are no mechanisms of
accountability for CCM representatives, nor channels for communication between representatives
and those represented. The body’s efforts at monitoring and evaluation are weak. And while the
majority of institutions that form the CCM are also sub- recipients of Global Fund project activities,
there are no policies within the CCM to prevent, define or manage conflicts of interest.
Effects on the structure and functioning of the Ministry of Health
The public sector is responsible for the implementation of Peru’s treatment programme; as such,
recent Global Fund activities have demanded a sizable response from the sector which has not
always been able to perform, given its limited human resource capacity. In the first stages of
execution, the lack of clarity on the roles, functions or competencies of the various actors involved
in Global Fund activities – the MOH, the CCM, the Principal Recipient (PR), and Sub Recipients (SR)
– generated project delays. Additionally, the implementation of Global Fund-supported activities
has forced the MOH’s National HIV/STI Sanitary Strategy (NHSS) to dedicate time to additional
administrative tasks, rather than to improve the coordination within the MOH offices and with
other public institutions and regional governments. Although Global Fund activities have required
some investment by Peru’s public sector, they have also offered resources for a significant number
of activities, which are expected to be continued by the MOH and other public institutions.
The MOH has adopted a number of objectives from Global Fund projects. In so doing, the MOH has
bypassed some of the lines of intervention emphasized in its own 2001-2004 National HIV/AIDS
Plan, such as the prevention and treatment of other STIs, and the prevention of parenteral
transmission. While the NHSS at the MOH has improved its technical capacities, existing demands
tend to prevent Ministry personnel from taking a proactive role and completing the development
of missing regulations.
Financing
Peru has received three Global Fund grants for HIV/AIDS, in Global Fund Rounds Two, Five, and Six.
The PR for all three grants is CARE, an international NGO. According to public sector budgetary
data, significant changes have occurred in the national response to the epidemic due to a
considerable increase in funding for HIV/AIDS, both in absolute amounts and in amounts relative
to the total budget. Increases in HIV/AIDS public budgetary levels reflect new activities in
137

HIV/AIDS, mainly around the Comprehensive Care Programme (see Table 4). Interestingly, only
11% of HIV funding in the three projects was allocated to HIV treatment, and this amount was to
be expended in the first phase, the first two years of the project, in 2004-2005. All other expenses
were focused on prevention, plus some training, equipment purchases, and overhead expenses.
ART expenses have been entirely covered by the National Treasury since 2006.
Budgetary information in the public sector is not regularly organized by specific diseases.
According to another study [13], HIV/AIDS budgetary increases have not implied cuts in other
health programmes.
Table 3: Total investment for HIV/AIDS in Peru, by funding source, 2005 and 2006 (US$)
Funding source 2005 (US $) 2006 (US $)
GFATM 4 644 784 1 768 372
Ministry of Health 2 191 648 7 040 128
Other – Public Sector 801 969 881 912
Other – Private Sector 606 500 852 100
Information
Large volumes of information are channelled to the NHSS where existing infrastructure is
insufficient to organize and use it for decision-making. The MOH lacks a comprehensive
monitoring and evaluation system able to provide feedback to Global Fund activities. Serious
difficulties were apparent regarding access among affected communities to information
generated by Peru’s HIV/AIDS projects. Informants thought that information was unclear and
focused on processes important to the relationship between the PR, SR and the Global Fund.
Internal discussions about dissemination of information on HIV/AIDS do not include other key
actors that could play a role in the social oversight of the national response.
Communities/Civil Society
Communities of PLWHA have become active in Global Fund projects, but this involvement has
sometimes undermined their original affiliations and patterns of collaboration. Before the Global
Fund entered the country, organizations of PLWHA were committed to building a National
Coordinating Organization (Peruanos Positivos). However, in late 2003, Global Fund guidelines
required consortia bidding to become Global Fund project implementers to include PLWHA
organizations in their activities. This created competition among different organizations, and in
2006 16 organizations left Peruanos Positivos to become independent groups that could
participate in consortia applying to become implementers of Global Fund Projects.
Effects on Access, Outcomes and Equity
The main change in the response to the epidemic is increased access to care. The government-led
National HIV Treatment Programme, which started in May 2004 with Global Fund funding and
became fully funded by the National Treasury in 2006, appears to have succeeded in reducing outof-pocket
care expenses dramatically. In 2000, it was estimated that out-of-pocket payments by
PLWHA accounted for more than 90% of spending on treatment. In 2007, the relevant rates,
including total household spending on treatment, were much lower. This indicates that the Global
Fund-supported roll-out of ART has permitted reductions in out-of-pocket spending (see Table 4).
138

Table 4: Household expenses and expenses in HIV care by city, 2007 (average monthly expenses in nuevos soles)
City Total (cases) Household expenses Expenses in HIV
care and tests
Lima 273 753,8 6,1
Callao 55 733,6 9,4
Chimbote 59 648,7 0,2
Huancayo 28 751,8 0,7
Iquitos 58 284,5 0,9
Total 473 680,7 4,8
Source: Survey among PLWHA in 4 Peruvian Cities, 2007 (CARE Peru & Universidad Peruana Cayetano Heredia, 2007)
The most important effect of implementation of Global Fund projects seems to be reduction of
morbidity and mortality among PLWHA. While no specific study of morbidity and mortality has
been conducted, both the survey and monitoring statistics from the National Treatment
Programme reveal low incidence of complications and of treatment failure.
The socioeconomic distribution of PLWHA in the National Treatment Programme is not different
from that of the population in general, and the rates of poverty among PLWHA are similar to the
national distribution (i.e., 58.2% among PLWHA, 54.4% in the general population). However, it is
not possible to rule out a problem of inequity in access to ART, since there is no information on
income distribution amongst the total population of PLWHA.
The economic productivity of PLWHA does not generally appear to have been affected by the
infection. Faced with a loss of employment because of HIV, PLWHA have been able to reinsert
themselves in the labour market, in part since most of them are independent workers (see Table 5).
Table 5: Loss of employment due to HIV/AIDS by employment status, Peru, 2006 (% PLHA and % cases)
Ever lost employment due to
HIV?
Total
Employment Status
Employed Unemployed Not EAP
Total 473 68.3 5.5 26.2
Yes 186 74.7 3.8 21.5
No 287 64.1 6.6 29.3
Source: Survey among PLWHA in 4 Peruvian Cities, 2007 (CARE Peru & Universidad Peruana Cayetano Heredia, 2007).
Discrimination and Vulnerable Populations
According to key informants, actions targeting “vulnerable populations”, i.e. men who have sex
with men (MSM) or female sex workers, are fairly limited, and HIV/AIDS programmes in Peru
devote little attention to the social roots of vulnerability. While earlier programmes included a
large peer-education component, as well as a medical care programme, peer education has since
been abandoned, and resources have been reallocated to recruit subjects for a programme of
periodic medical check-ups. As a consequence, Peru’s HIV/AIDS strategy is primarily biomedical,
139

without substantial emphasis on preventive information and interpersonal resources for risk
reduction.
Responses to questions exploring attitudes towards PLWHA reveal discriminatory attitudes in the
general population. Importantly, differences in the prevalence of such attitudes were identified
across geographic regions, but not across educational levels. Attitudes were most positive in
Iquitos, a city in the Peruvian Amazonia, with a culture generally more open to sexual diversity.
According to qualitative findings, health workers’ perceptions about PLWHA vary on the basis of
the characteristics of individuals infected. Perceptions of “guilt” or innocence in contracting the
disease play a clear role in how PLWHA are treated [14,15]; children and women (i.e., “innocent
victims”) tend to be more valued, while transgender people and sex workers are, generally, socially
excluded.
A review of communication materials prepared for specific target populations of the Global Fund
projects showed diverse messages and perspectives. Some materials still present frightening
messages about HIV directed to adolescents and young adults, which may produce
misinformation and contribute to stigma and social marginalization of PLWHA.
Among PLHA responding to the survey, no differences were reported on the quality of care at
public clinics based on gender and sexual orientation. Interpretation of this finding, however, is
limited by potential differences in expected quality of care based on pre-existing social exclusion
(see Table 3) [16,17].
Discussion
The Global Fund-funded projects in Peru have defined a new relationship between the public
sector and civil society with many positive aspects, although several issues remain to be resolved.
Multisectorality implies fairness in decision-making by all actors. The process of instituting
multisectorality within Peru’s Global Fund activities is still new, and has meant new challenges
overall for the MOH, NGOs, organizations of PLWHA, and vulnerable groups.
Generally, the study has revealed a multifaceted process that has changed the ways in which
HIV/AIDS work is conducted since the introduction of funds from Global Fund. The Global Fund’s
top-down call for a CCM has required multiple sectors as well as both public and private actors to
articulate a consistent, national response to the epidemic. This has introduced the opportunity for
broader commitment and, to some extent, a more democratic process. The Global Fund-funded
projects in Peru have configured a new relationship between the public sector and civil society.
However, more work is needed to ensure fairness and genuine inclusivity in decision-making
processes [18,19].
Organizations of vulnerable populations, while not involved in the CCM originally, have recently
been invited to participate. However, their needs are still not considered adequately in the design
and implementation of the projects. To achieve equity in input to Peru’s HIV/AIDS strategy, the
health sector and other sectors must recognize that the HIV epidemic is a broad social problem,
rather than only a health problem. Activities funded by the Global Fund should improve the level
of equity in access to project benefits and counter stigma and discrimination directed towards
PLWHA and vulnerable groups. It is also clear that other public sectors still find it difficult to
understand their role in the response to a problem that is perceived merely as a health problem. A
well-defined strategy is needed to mainstream the public response to HIV/AIDS in ways that really
strengthen its scope and impact [20].
140

Within the health sector, the main overall conclusion is that the implementation of Global Fund
projects has emphasized the role of non-state organizations, with limited mechanisms for transfer
of funds to the MOH. However, the new processes related to implementation of Global Fundfunded
projects have demanded a substantially increased time commitment from the NHSS
Technical Team, and seem to have led to a loss in their capacity to guide and oversee the health
sector components of the national response in more proactive ways. Logistic systems have also
experienced problems in meeting project demands, and the NHSS’s capacity to process the
substantial amount of information generated has also been limited.
Instabilities introduced by the size of HIV/AIDS investments in relation to other health investments
must be addressed through appropriate funding mechanisms and through operations research
[21]. There is also a need to create mechanisms of accountability within the CCM for the
organizations it represents, with regard to the CCM itself and to Peruvian society as a whole
[22,23]. Moreover, the CCM must fulfil Global Fund requirements with regard to the adoption of a
code to prevent and manage conflicts of interest, as well as the creation of effective mechanisms
for monitoring and evaluation [24-26].
141

References
[1] The World Bank Group. Peru at a glance. (http://devdata.worldbank.org/AAG/per_aag.pdf;
accessed 20 March 2009).
[2] United Nations Development Programme. Human Development Reports: Peru. Proprietary
Online Database. ( http://hdrstats.undp.org/en/2008/countries
/country_fact_sheets/cty_fs_PER.html; accessed 20 March 2009).
[3] The World Bank Group. World Development Indicators (WDI) Online. (http://ddpext.worldbank.org.ezp-prod1.hul.harvard.edu/ext/
DDPQQ/member. do?method=getMembers;
accessed 20 March 2009).
[4] Voice of America News. Many Peruvians Struggle to Gain Health Care
Access. (http://www.voanews.com/english/archive/2006-05/2006-05-03-voa41.cfm)
[5] World Health Organization. WHO Statistical Information System
(WHOSIS). (http://www.who.int/whosis/en/; accessed 4 March 2009).
[6] The Financial. Peru/WB: US$15 Million to Improve Health Care for Poor Rural Populations.
(http://www.finchannel.com/index.php?option=com_ content&task=view&id=29990&Itemid=9;
accessed 4 March 2009).
[7] USAID: Peru. HIV/AIDS Health Profile; 2008.
[8] UNAIDS. 2008 Report on the Global AIDS Epidemic 2008.
[9] Cáceres C. La Pandemia del SIDA en un mundo globalizado: Vulnerabilidad, subjetividad y los
diálogos entre la salud pública y los nuevos movimientos sociales. En: Cáceres C, Cueto M, Ramos
M y Vallenas S. (Eds): La Salud como Derecho Ciudadano: Perspectivas y Propuestas desde América
Latina. Lima: UPCH. 2003.
[10] Cueto, Marcos. 2001. Culpa y Coraje. Historia de las políticas sobre el VIH/Sida en el Perú. CIES,
UPCH, Lima, 2001.
[11] Global Fund to fight AIDS Tuberculosis and Maleria. Portfolio of Grants - Peru & the Global
Fund to Fight AIDS, Tuberculosis and Malaria.
(http://www.theglobalfund.org/programs/portfolio/?lang=en&countryID=PER; accessed 20 March
2009).
[12] UNAIDS. 2008. Country Situation: Peru. Available in:
(http://data.unaids.org/pub/FactSheet/2008/sa08_per_en.pdf).
[13] Cáceres C, Petrera M, Portocarrero A, Sandoval C. (2009). Evolution of Public HIV Expending in
Peru in the Context of GFATM Funding - Findings of the Peru HIV and Globalization Study. Abstract
submitted to IAS 2009.
[14] Ogden J, Nyblade L. 2005. Common at its core: HIV-related stigma across contexts. Report
prepared by the International Center for Research on Women (ICRW). 2005. Available in:
(http://www.icrw.org/docs/2005_report_stigma_ synthesis.pdf).
142

[15] Parker R, Aggleton P. HIV/AIDS-related stigma and discrimination: a conceptual framework
and agenda for action. Report prepared by the Population Council, May 2002.
(www.popcouncil.org/pdfs/horizons/sdcncptlfrmwrk.pdf )
[16] Girón JM., Segura E., Salazar V., Valverde R., Salazar X., Cáceres C. 2007. Percepciones de las
personas viviendo con VIH/SIDA sobre los servicios de salud y el tratamiento antiretroviral de gran
actividad: Un estudio transversal en 5 ciudades del Perú. Rev Peru Med Exp Salud Pública, jul/set
2007, vol.24, no.3, p.211-217.
[17] Llanos Zavalaga, Luis Fernando; Mayca Perez, Julio y Navarro Chumbes, Gian Carlos. 2006.
Percepciones de las personas viviendo con VIH-SIDA (PVVS) sobre la calidad de atención
relacionada con TARGA. Rev Med Hered, jul./set. 2006, vol.17, no.3, p.156-169.
[18] Caceres CF, Cueto M, Palomino N. Policies around Sexual and Reproductive Health and Rights
in Peru: Unveiling False Paradoxes. Global Public Health 2008, 3(S2): 39-57.
[19] Cunill Grau, Nuria. 2005. “La intersectorialidad en el gobierno y gestión de la política social”.
Trabajo elaborado por encargo del Diálogo Regional de Política del Banco Interamericano de
Desarrollo, BID.
[20] UNDP & UNAIDS. 2005. Contribution towards reversing the HIV/AIDS epidemic in the context
of the United Nations system strategic plan for HIV/AIDS for 2001-2005. Available in:
(http://www.undp.org/execbrd/word/dp02-3.doc).
[21] Buse K, Walt G. Aid coordination for health sector reform: a conceptual framework for analysis
and assessment. International Health Policy Programme, London School of Hygiene and Tropical
Medicine, London, UK. Health Policy. 1996; 38:173-187.
[22] Caceres CF, Mendoza W. The National Response to the HIV/AIDS epidemic in Peru.
Accomplishments and gaps. In press, JAIDS.
[23] ONUSIDA. Informe sobre la Epidemia Mundial del Sida 2006. Resumen de Orientación-Edición
especial con motivo del décimo aniversario del ONUSIDA. Ginebra (Suiza), 2006.
[24] The Global Fund to Fight AIDS. Tuberculosis and Malaria. 2003. Explanatory Document on
Conflict of Interest; Fifth Board meeting report of the Governance and Partnership Committee;
Geneva 5-6 June 2003.
[25] The Global Fund to Fight AIDS, Tuberculosis and Malaria. 2004. Policy on Ethics and Conflict of
Interest for Global Fund Institutions; Seven Board Meeting, Geneva 18-19, March 2004.
[26] The Global Fund to Fight AIDS, Tuberculosis and Malaria. 2006. Report of the Global Fund
Partnership Forum 2006. Celebrating Successes and Overcoming Challenges. Final Draft; 1-3 July,
2006, Durban South Africa.
143

Rwanda: The Impact of Global Health Initiatives on
the Health System: A mixed methods
analysis
Abstract
Mukherjee JS, Jerome JG, Sullivan E, May MA, Mayfield A, Lambert W,
Dhavan N, Carney N, Rhatigan J, Ivers LC
With acknowledgment to Ruzindana K, Wroe E, Kersten E.
Increasing attention is being paid to the impact that GHIs such as the Global Fund and PEPFAR
have on health systems as a whole. We aimed to evaluate the impact of these two GHIs on the
health system in Rwanda.
We used mixed qualitative and quantitative methods: semi-structured interviews, observation,
examination of documentary material, and collection of quantitative data. Fifteen key informants
from NGOs, international organizations and the Ministry of Health (MOH) participated in semistructured
interviews.
The Global Fund and PEPFAR were largely associated with strengthening of the health system in
Rwanda, with clear credit for the success of the programmes attributed to the strong national
framework and to coordination by the government of Rwanda.
GHIs such as the Global Fund and PEPFAR can successfully strengthen health systems if this
objective is included in planning and design, and if leadership and coordination are ensured.
Background
The Rwandese Republic is a small, landlocked country in East Africa. Bordered by the Republics of
Burundi, Uganda and Kenya, and the Democratic Republic of Congo, it has a population of 9.7
million and is expected to average a 2.7% annual population growth rate between 2005 and 2010
[1]. After decades of colonial rule by Germany and Belgium, Rwanda gained independence and
held its first parliamentary elections in 1962. Violent outbreaks between the majority ethnic group,
the Hutus, and the minority Tutsis, characterized the following decades. In April 1994, ethnic
tensions between the extremist Hutu-led government and the Tutsis culminated in a 100-day
genocide in which approximately 800,000 Tutsis and moderate Hutus were killed. In July 1994,
Rwanda Patriotic Front (RPF) troops, led by Commander Paul Kagame, defeated the extremist
regime.
In the years following the genocide, a single-party coalition government was established, and a
new constitution was adopted in 1995 [2]. Rwanda held its first post-genocide local elections in
March 1999, and its first presidential and legislative elections in August and September of 2003,
respectively. In the parliamentary elections of 2003, the RPF won 33 of 53 seats and became the
ruling party of Rwanda. Later that year, Kagame was elected President by a direct popular vote.
144

Since 1994, Rwanda has pursued a policy of decentralization, granting local entities some degree
of administrative and fiscal autonomy, while allowing the Central Government to retain
responsibility for policy formulation, monitoring, and analysis [3]. Power has shifted from the
country’s five provinces, to its 30 districts, 416 sectors, and finally, 2080 local cells [4].
Rwanda - Table 1 Basic Socioeconomic, Demographic and Health Indicators*
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 9,734 2007 World Bank
Geographic Size (sq. km) 24,670 2007 World Bank
GDP per capita, PPP (constant
2005 international $)
818.24 2007 World Bank
Gini index 46.68 2000 World Bank
Government expenditure on
health (% of general
government expenditure)
18.8 2007 WHO NHA
Per capita government
expenditure on health
(current US$)
17 2007 WHO NHA
Physician density (per 10,000)

Rwanda was US$ 584.9 million in 2006 [10]. GHIs have contributed substantially to health
financing in Rwanda. Since 2003, the Global Fund has approved US$ 292.3 million and disbursed
US$ 224.3 million in funding for HIV/AIDS, HIV/TB, malaria, TB, and health systems strengthening
(HSS) to Rwanda. PEPFAR has disbursed US$ 394.6 million in funding for HIV/AIDS prevention,
treatment, and care since 2004. In 2006, 92% of the US$ 87.6 million spent on HIV/AIDS was
incurred by donors [11]. That year, 48% of the total health sector budget came from Rwanda’s
national budget and the remaining 52% came from international partners [9]. Rwanda received
US$ 62 in aid per capita in 2006, compared with US$ 48 in aid per capita for sub-Saharan Africa
overall [5].
Table 2 Global Health Initiative Investments*
(*) Full data sources for all indicators are provided in Annex 1.
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 1, HIV/TB 14,641,046 14,641,046
Round 3, HIV/AIDS 56,646,465 52,501,904
Round 3, Malaria 38,597,403 29,827,296
Round 4, TB 17,027,672 10,298,215
Round 5, HSS 33,945,080 27,647,900
Round 5, Malaria 39,649,362 39,149,502
Round 6, HIV/AIDS 31,563,456 30,196,743
Round 6, TB 7,426,750 2,438,357
Round 7, HIV/AIDS 63,978,011 27,590,463
Round 8, Malaria 58,926,734 0
TOTAL: 359,401,979 234,291,426
PEPFAR (in US$)
Year Amount Allocated (in US$)
2004 39,240,985
2005 56,909,487
2006 72,102,434
2007 103,041,870
2008 123,468,840
TOTAL: 394,763,616
GAVI (in US$)
Disease Priority Amount Approved (in US$)
Pentavalent vaccine 44,268,000
Pneumococcal vaccine 23,785,000
Vaccine introduction grant 217,000
Injection Safety 369,500
Immunization services support 3,788,700
Health system strengthening 5,605,000
TOTAL: 78,032,988
146

World Bank MAP
Title FY Approved/Closing Date Commitment (in US$)
Multisectoral HIV/AIDS Project 2003/2008 40,500,000
Methodology
Qualitative case study
The data collection for the Rwanda case study included semi-structured interviews, observation,
examination of documentary material, and collection of quantitative data. Key informant
interviews used a standard semi-structured interview template that was designed to address how
PEPFAR and the Global Fund funding have interacted with Rwanda’s health system. We addressed
the following elements of the health system: governance, financing, health workforce, monitoring
and evaluation (M&E), health technologies, and communities and civil society WHO considers
these elements to be the building blocks of the health system. To limit the scope of the initial ninemonth
research programme, we limited our embedded units of analysis to HIV and TB
programmes funded by PEPFAR and the Global Fund.
The sample strategy for key informants used purposive sampling. Most key informants had welldefined
roles in the country’s health system as well as significant experience with both PEPFAR
and the Global Fund. The Rwanda study had a sample size of 19; follow up interviews were
completed with five key informants. The sample included people working in the government at
the national level (size = 11), NGO directors from both Rwanda-based and international NGOs (size
= 6), a USAID employee, and one referral hospital director.
Researchers recorded and transcribed interviews to ensure accurate data collection and facilitate
qualitative analysis, conducting six interviews in French and the remainder in English. All
interviews were transcribed in English for analysis. Interviewers took field notes based on their
observations and completed post-interview debriefings. Transcribed interviews were entered into
NVivo8, a qualitative data analysis program (QSR International, Cambridge, MA). A thematic
qualitative analysis approach and an iteratively developed set of codes were used to examine the
data. To further support the authenticity of findings and auditability of analytic processes, we
engaged in inter-rater reliability activities as we created and applied codes; wrote memos about
our analytic decision-making; and conducted participant validation exercises.
To triangulate the data collected in key informant interviews, we collected and reviewed publicly
available documents in each country (i.e. National Health Strategy, PEPFAR or Global Fund Country
Reports).
Quantitative Data
Quantitative data was collected from four health centres in rural Rwanda. Convenience sampling
was used to select facilities from departments that interviewees represented and to select facilities
with different degrees and types of GHI funding and different models of care. A facility analysis
tool was created based on combining the common elements of instruments of the WHO, the
147

International Health Partnership, the Clinton Foundation and other groups. 23 Data collection was
carried out in January and February 2009. Data were abstracted from paper registers, monthly or
annual site reports, or from the regional ministry of health office. Complete definitions of
indicators were provided to the data abstraction teams. Data were entered in Boston, after a data
dictionary was created and rules of coding were established. Data were entered into MS Access,
and analysed in R version 2.7. Analysis of this data is ongoing.
Results
Leadership and Governance
GHI impact on the role of the Ministry of Health
The MOH is committed to leading the national response to health, and informants universally
agreed that the MOH had the capacity to do so effectively, providing direction and leadership. The
role of GHIs was seen as enabling and strengthening the national response. While the Global Fund
has allowed the MOH to define its role and priorities, informants felt that PEPFAR’s model of
working through NGOs did not acknowledge the government’s leadership in health system
strengthening.
GHI impact on health care managerial and leadership capacity
While informants mentioned some investment from the GHIs in improving managerial and
leadership capacity, most felt that more support was needed in these areas. Some Global Fund
funding was used to build administrative capacity within the MOH, but generally the Global Fund
was seen as “simply a funding system.” PEPFAR provided more technical support, which has been
valuable in improving the management capacity of the Rwanda Drug, Consumables and
Equipment Central Procurement Agency (CAMERWA).
Financing
GHI funding impact on resource allocation within national or local health budgets
While Global Fund money flows directly to the MOH, and PEPFAR money is channelled to NGOs,
both the Global Fund and PEPFAR contributed greatly to the resources available for health,
particularly for HIV care. The GoR has significantly expanded its HIV programming as a result, and
has tried, when possible, to scale-up general health services as well. Several informants described
the GoR as using its national plan for health to align available funding with its priorities. It has had
to negotiate with GHIs around this strategy, but generally been able gain their support. One
informant explained:
23 A referral score was created from variables that evaluated the facility’s ability to function within a health network. One
point was assigned for each of the following: a higher level health care facility is within two hours by car from the health
centre location; the health care centre has the ability to communicate using a radio, phone, or the internet, with a higher
level health care facility; the health centre has access to an ambulance; the health centre provides the patient with a
transportation fee to facilitate referral; the health centre uses referral forms when referring patients from facility to
facility; the patient who is referred is physically accompanied to the referral site during part or all of the journey and/or
physically accompanied during the encounter at the referral site; and the health centre has referral forms for community
health workers to refer patients to the facility.
148

As much as it is always easy to use the funds from the national budget for
integrating whatever you have to integrate, it is not obvious when the funds are
coming from PEPFAR, from Global Fund, from bilaterals, even multilaterals,
because generally, they come through, for vertical programmes. It is difficult to
convince the donors to integrate. But we have been very lucky. We have been
working with people who are very understanding, and the representatives –
account representatives – they were very understanding, and I’m sure that they
have been advocating for that toward their headquarters.
Both GHIs, in particular the Global Fund, were seen as broadening the scope of activities they
allowed over time and demonstrating increasing willingness to support the general health system.
One informant commented, “Obviously, the Global Fund had more impact because their
mechanism and the funds were fully managed by Rwandans and Rwandan organizations unlike
PEPFAR.” Another commented:
But the other thing that we have, that is good for having the Ministry of Health as
the [Global Fund] Principle Recipient, is like this money comes in to support our
national budget. That implies that what we have put down as the priority, we,
the government of Rwanda, have put down as priority, to be financed – that is
what gets financed. And this is how we have managed to get this money into the
health system.
Informants were mixed in their assessment of whether PEPFAR funding aligned with national
priorities. Some described it as having more rigid funding restrictions, while others thought that
PEPFAR was increasing primary care service availability in the NGO sector. One informant
reflected, “Even since I’ve been here, I’ve seen changes in PEPFAR’s willingness to see itself as a
larger part of a development programme, and not very narrowly focused on HIV.”
Transparency of resource allocation decisions
The Global Fund and PEPFAR both engaged the GoR in grant planning. When the MOH
decentralized HIV services, moving the financial unit from the national level to the central level,
both GHIs agreed to the new system. An informant explained:
Global Fund funding… is the first resources that we transferred in cash to health
centres for staff salaries, and the results showed that they were well-managed;
people used to think that health centres could not be able to manage them.
…today even PEPFAR accepts to send funds in cash to health centres and also to
the government.
The Country Coordinating Mechanism (CCM) made Global Fund resource allocation decisions in
Rwanda, controlling how Global Fund funding was allocated both in terms of programming within
the MOH and among NGO partners. Continued Global Fund funding was dependent on results,
and informants felt that this system created accountability for both the GoR and the Global Fund.
149

PEPFAR’s funding practices were less transparent to respondents. Many stated that there the
distribution of funding was not as centralized, there were too many overheads, and some felt that
working through NGOs created inefficiencies. One informant commented:
American NGOs have different rules; the contract they sign with the American
government they sign it as American NGOs, and they manage the resources the
American way, and their context is a context of a rich country. A rich country
spends a lot in accordance to its riches. The staff get higher salaries. As managers
we cannot afford the salaries they get. Any cost related to their activities,
logistics and all that is very high compared to ours. That is why like they say their
overhead is very high that is why the level of their operations is higher than ours.
To me that is the difference.
A few others commented that the GoR could not evaluate the efficiency of the PEPFAR-funded
NGOs because the funding information was not shared with it.
Perceptions of the PEPFAR steering committee were mixed. Some felt that the committee was
collaborative and valuable, while others thought that PEPFAR’s priorities still dictated final
decisions. Some commented that it was less flexible in its funding requirements than the Global
Fund, but not necessarily inflexible.
Impact of GHI funding process on recipients
As mentioned, both the Global Fund and PEPFAR’s funding processes have improved over time,
but there are still shortcomings with each one. Some data suggested the Global Fund occasionally
changed its requirements suddenly. While the Global Fund initially approved the MOH as head of
the CCM, it later said that its leadership represented a conflict of interest, and requested that a
different body take the lead. The National Council for the Fight against Aids (CNLS), which works
closely with the MOH, took over leadership of the CCM, and the MOH remains the Principal
Recipient. Because the MOH remained closely involved with the Global Fund’s funding process,
there was general approval of its funding decisions. Sudden changes to the Global Fund M&E
requirements have also been difficult for recipients:
What I can just say it’s that reporting is a good thing, but Geneva always changes
rules and procedures and they are very tiring… What is good today will not be
good tomorrow. It just comes from the sky, we don’t know why. Many of those
changes are absolute. Doesn’t fit in our national policy. Disturbs us. It doesn’t go
for what work in our country, and we are obliged to go with that. But it’s not the
best way to do. That’s my point of view. Probably as it’s Global Fund it’s global
and it’s difficult to do things that will fit in all national policies and strategies, but
they should be more generic and more based on results than on process and
they’re very focused on process.
150

In terms of PEPFAR, respondents suggested the one-year funding cycle was too short. Many
resources went into completing the application each year, and there was not sufficient time to
evaluate the previous grant before submitting the next proposal:
This [PEPFAR monitoring and evaluation] was a nightmare! At the beginning you
would receive money to be spent in one year, but this would come way past the
middle of the year, around July or August. This I think was as a result of the
schedule of the American Congress. The funds would come in when the state
budget is already running, and you would try to disburse it. But before you are
even half way through it, around the month of December, you are asked to
submit project proposals for another round of funding before you even find out
the results of round one. We tried to set up a steering committee, but it failed to
streamline the working modalities. The reporting of PEPFAR was only
understood by one or two people.
Limitations of GHI funding models
Informants repeatedly commented that there was a significant unmet need for investments in
infrastructure. The expansion of the Mutuelle de Santé programme was an important step in
expanding access and increasing demand for services (including voluntary counseling and testing
[VCT] opportunities), but more investment of the system-level sort was needed.
In particular, PEPFAR’s restriction on renovation of facilities (as opposed to construction of new
facilities) was seen as a significant impediment to implementing HIV programmes in some
locations. In addition, its stipulation of the percentages of funding that had to be spent on
different programme components (i.e. prevention, treatment, support) was considered limiting.
Health Workforce
GHI funding impact on staffing levels, training and retention
GHI funding helped Rwandan health facilities achieve their target levels of staffing at public
facilities, largely by increasing the total number of doctors working in the health system. PEPFAR
and the Global Fund have increased the availability of training within Rwanda for community
health workers and doctors. The Global Fund allowed Rwanda to use some of its training funds on
non-target diseases, strengthening the overall health system. The informants linked increases in
workforce retention rates with Global Fund training. Informants appreciated the Global Fund’s
outside consultants conducting trainings, crediting them with building capacity in the public
sector. One informant commented, “Global Fund found good consultants who trained the
personnel, and the personnel learnt a lot from them. And that is why the reports we now submit to
CCM in Geneva are quite well made because there are people who got trained in doing it, and they
are also training others.”
Some informants commented that the quantity of PEPFAR trainings had resulted in frequent
absenteeism in health facilities. PEPFAR’s training had also extended to Rwanda’s national supply
chain management entity (CAMERWA), from which it had sent staff to several other countries for
training.
Salary increases made possible by GFATM monies generally had a positive impact on recruitment
and retention in the public sector. Respondents cited the Global Fund as supporting the GoR in
hiring district level staff. However, some informants commented that they lost public sector
151

medical staff to higher-paying administrative positions at the Global Fund, exacerbating the
human resource shortage. As one informant reported:
Doctors and nurses have quit the patient’s bedside in a scramble for ‘AIDS’
money, as cynics like to call it. These medical people leave the mainstream
clinical work to take up administrative jobs in Global Fund coordination offices.
This has escalated the scarcity of medical professionals even further.
In addition to the movement of clinical staff to administrative Global Fund positions, PEPFAR was
also seen as pulling human resources out of the public sector and into NGOs, as the latter were
often able to offer higher salaries. While many informants viewed this as a negative consequence
of PEPFAR’s funding system, others commented that overall it increased human resources within
Rwanda, as it prevented practitioners from seeking jobs abroad.
GHI funding impact on workforce capacity
GHIs have built significant capacity through their investments in training. In particular,
respondents saw great potential in the willingness of GHIs to recognize community health workers
as an important cadre of human resources for health. One informant stated, “If they reinforce
community health workers, everything will be achieved.” Since community health workers are
rarely limited to addressing HIV at the community level, capacity building within this population
was seen as a form of health system strengthening, as it could increase access to services for the
general population.
Data showed PEPFAR increased employment opportunities and built capacity in the Rwandan
workforce, but not in the public sector given that it funded NGOs. In contrast, the Global Fund
contributed to capacity building of all personnel. Global Fund funding could be used for measures
that would sustainably enhance the workforce, such as investing in nursing schools.
Quantitative analysis demonstrated increases in the number of doctors (non-specialized), nursing
staff, non-degree laboratory technicians, non-degree pharmacists, and community health workers.
Altogether, facilities in Rwanda had no doctors before GHI funding, and 1.45 after, all of whom
were non-specialist practitioners.
GHI impact on human resource strategies
GHIs have helped with the human resource shortages by contributing to salaries for doctors and
nurses. Global Fund monies were used to increase salaries in rural areas, leading physicians to
accept jobs and remain in areas where positions had previously not been able to be filled.
Retention also increased in the public sector overall and the number of health practitioners
seeking opportunities abroad fell.
In addition, the GHI’s decision to support the GoR’s performance-based financing system has
enabled payment structures that are more effective and that provide incentives to health care
professionals. However, despite these changes, informants often stated that the government
could not compete with the salaries that NGOs offered. Some described NGO salaries as
“destabilizing” to a market where there were so few doctors and nurses.
152

Medical Vaccines, Products and Technologies
GHI funding impact on health system infrastructure
Most informants felt that more GHI investment in infrastructure was needed despite significant
contributions. The Global Fund contributes to the renovation and construction of health facilities
to deliver Prevention of Mother to Child Transmission (PMTCT), Multidrug-Resistant Tuberculosis
(MDR-TB) treatment, and VCT, in addition to the general laboratories in health centres and the
creation of a national MDR-TB referral centre. In most cases, these projects improved the care of all
patients, not just patients with HIV or TB. One respondent described infrastructure improvements
as follows:
I mentioned that they built these 100 plus health centres – buildings – to allow
capacity for, really HIV treatment centres. Again, they aren’t dedicated only to
HIV treatment centres. They’ve done a number of other things for these facilities
that are HIV treatment centres. They’ve provided generators. They’ve helped
build up the national laboratory. They’ve helped fund the [TRAC] Centre….
PEPFAR monies were used only for the renovation, not construction, of health centres and the
percentages of funding that had to be spent on different programme components (i.e. prevention,
treatment, support) were considered limiting. One PEPFAR-funded NGO was thought to have
renovated more than 30 facilities. In addition, PEPFAR supported the installation of solar power at
many facilities.
The quantitative analysis demonstrated significant increase in physical space: on average, physical
infrastructure increased by 1138 square feet per facility in Rwanda, a growth of 83%.
GHI impact on access to innovative technologies
The GHIs contribute to procuring equipment for hospitals and laboratories. Global Fund funding
has been used to purchase microscopes, x-ray machines, ambulances and other equipment. Global
Fund monies also improved the referral system between health facilities and community health
workers, leading to improved coordination of care. PEPFAR’s investments in equipment and
materials are accompanied by technical support, which is beneficial.
GHI materials and equipment benefited those with non-target illnesses as well as those with the
target diseases. In particular, great gains were seen in primary care and family planning. One
respondent explained, “We equip antenatal clinics, we equip delivery rooms. All those
infrastructures and equipments serve not only for malaria or for TB or for HIV itself, but for the
overall population, and among them people suffering from TB, malaria, and HIV.”
Supply Chain
Both GHIs have invested significantly in Rwanda’s supply chain and logistics system. All drugs,
supplies and consumables purchased with Global Fund monies and most purchased with PEPFAR
monies are handled by CAMERWA. In addition, all antiretrovirals (ARVs) distributed nationally are
warehoused by CAMERWA. Antiretroviral therapy (ART) sites submit their forecasted needs
through the National Quantification Committee, in which both the Global Fund and PEPFAR
participate. The Quantification Committee approves the total quantities of medications needed
and divides the costs among the donors. CAMERWA then receives all the medications at the
national level and distributes them to the sites through the coordinated procurement distribution
153

system (CPDS), in which many stakeholders participate. CPDS is hailed as an example of true
partnership making sustainable improvements:
I think it’s [CPDS] a really strong example of SCMS [supply chain management
system], Global Fund, Clinton Foundation, UNICEF [the United Nations Children’s
Fund] working together, but working together with the actual country – so again
trying to reduce the issue of vertical programmes, but also ensure sustainability
of the supply chain. And really, the CPDS governs the whole quantification and
procurement process.
With this national distribution system, the distribution of drugs occurs based on need, not on
available, site-specific funding (i.e. PEPFAR sites often provide first-line drugs paid for by the Global
Fund and non-PEPFAR sites often provide second line drugs that PEPFAR purchased). In addition,
CAMERWA is the main supplier of essential medicines and supplies in Rwanda, so GHI investments
have spillover into the functionality of the general supply chain. GoR’s insistence that GHIs work
together with CAMERWA helped avoid the creation of parallel systems and resulted in great health
system strengthening. As one respondent described:
I think Rwanda’s fairly unique in that they have been quite aware of this [the
dangers of parallel systems], so they’ve tried to prevent that. So I would say
they’re actually an example of how not to set up parallel systems. But some have
been created. But I think it’s minimal.
The Global Fund’s contributions to CPDS were primarily financial; it helped finance a warehouse,
the CPDS, and staff. In addition, it paid general overheads to CAMERWA, increasing the amount as
CAMERWA transitioned to an active distribution system for all medicines and supplies ordered
with Global Fund money. The Global Fund helped CAMERWA adjust to its growth by reinforcing
inventory and distribution systems, resulting in fewer stock-outs, and providing transportation for
distribution.
PEPFAR funded the procurement of large quantities of ARVs, second-line ARVs, treatment for
opportunistic infections, machines and reagents. In contrast to the Global Fund’s model of
allowing CAMERWA to purchase the goods directly, PEPFAR has CAMERWA choose a PEPFARapproved
supplier and place the purchase order, then pays the supplier directly (i.e. CAMERWA
does not have access to the funding). These restrictions sometimes resulted in uncertainty for
CAMERWA about the status of orders. The technical assistance PEPFAR provided through SCMS
was invaluable and included on-site support with major activities including staff training,
warehouse modernization, and installing information systems, such as MAX, an electronic
warehouse stock management system. One informant commented:
I’m getting quality TA [technical assistance] from SCMS. At the moment I’ve got
about 5 consultants that are helping me to modernize my warehouse, develop
systems at CAMERWA, and also work on distribution, active distribution. So for
these two years I’ve had a wonderful relationship with them.
Other PEPFAR partners also provided support at the district level, including salary support and
training for pharmacists and other workers.
154

Information
GHI influence on reporting mechanisms
GHI funded improved M&E systems, although some informants were concerned that HIV systems
were more sophisticated than other systems and not integrated. The MOH developed and put in
place the Global Fund M&E plan. An M&E group ensures the plan’s execution, reviewing strategic
plans, making field visits to collect data, and revising the plans when appropriate. In addition to
this group, the Global Fund funded the installation of electronic information systems to facilitate
reporting, and other measures to improve the quality of data collected, including hiring additional
district-level staff. The challenges in meeting Global Fund M&E requirements did not necessarily
stem from a lack of funding, but from a lack of capacity. One informant said, “I remember even in
my office we used to fail to finish [Global Fund] money allocated for monitoring and evaluation
[M&E] because you had to find the right people to do that monitoring and evaluation [M&E] from
the national level up to the community level.”
PEPFAR has worked with the government on its M&E system and requirements as well. The
PEPFAR requirements are quite extensive and require sophisticated systems to be in place, which
has at times been a burden on implementers. Some informants criticized PEPFAR’s requirements
because PEPFAR did not want the national indicators included in the same system as the PEPFAR
indicators. Currently, PEPFAR is working to harmonize multiple information systems.
Many informants suggested that all GHIs synchronize their indicators and M&E plan, including
their reporting calendar. As one person expressed:
… my personal impression is that the information systems, and data collection
and monitoring and evaluation [M&E] in Rwanda have been very vertical – very
separate for the malaria programme, for the HIV programme, and probably. . . as
a result of, or definitely supported by the vertical programmes that we’ve
implemented.
While presenting some challenges, the GHI monitoring and evaluation systems did facilitate the
transition to performance-based financing by making data to evaluate performance more reliable
and available.
Service Delivery
Integration of GHI-funded programmes and the health system
As mentioned, the GoR developed an integrated approach to health delivery and built system
strengthening components into GHI grants as much as possible. As one respondent reported:
It [the health system] was integrated; there are no HIV health facilities. We have a
microscope, and we have the capacity to test HIV. With this capacity, we also test
other diseases, like syphilis or other. The material is there. The people are trained,
so they are not just doing HIV. But we have through the opportunity of HIV
money to buy the microscope or for the machine, etc.
Other examples of health system strengthening efforts include the integration of PMTCT
programmes in maternal health programmes and the purchase of ambulances for all those
155

needing medical transport. Both GHIs provided funding for Mutuelles de Santé. To achieve
integration between the GHI-funded programmes and the health system, the CCM funded
organizations that would invest in health system strengthening. System strengthening was seen
as a necessary part of developing targeted responses, and the Global Fund approved funding
being used in this way.
Overall, the Global Fund and PEPFAR investments supported the national response to HIV, TB, and
malaria. Both GHIs supported the national decentralization of HIV services and the MOH’s national
HIV treatment guidelines so that all patients accessing HIV treatment would receive the same
medications, regardless of the site’s funding source. One respondent talked about a similar means
of integration for the TB programme:
The [Global Fund] project has allowed us to integrate the money in the national
plan for response to the disease; it has been a success. Also what is good is that
all the funds for tuberculosis have been aligned to the same plan and the same
programme. So, in one word, we have applied what we call the three one’s. That
means one action plan, one monitoring and evaluation [M&E] plan, and one
unique authority to lead people around that plan and that monitoring and
evaluation [M&E] plan. That’s how we managed. It has help also to, to provide
the same level of services with geographic equity.
Treatment programmes for HIV and TB have also been integrated with one another at the facility
level.
The MOH dictates where its partners (NGOs providing HIV services) - who are also required to
provide a minimum package of primary care services - implement programmes and therefore
where GHI-funded programmes work.
Several coordination bodies exist between GHI and the government. These include the GHIspecific
groups, such as the CCM and the PEPFAR steering committee, but also entities like the
coordinated procurement and distribution system, which organizes the ARV supply chain for the
country (see the supply chain section below for more information).
Respondents perceived PEPFAR’s funding model as resulting in less integrated programmes than
the Global Fund’s model. The number of NGOs in the health sector can result in lack of
coordination or duplication of activities. As one informant noted:
Again this [impact on non-target diseases and access] depended entirely on the
NGOs involved. It depended on the goodwill of the country representative or the
ladies and gentlemen who were managing them. We could get some good
impact only when they accepted to integrate their actions with ours. Otherwise
our systems run parallel. The Government’s interest was to integrate them into
the system but theirs was to stay isolated.
Impact of delivery models on access and coverage
GHI investment in Rwanda had a significant impact on geographic equity of health delivery,
increasing the availability of ART and other services, particularly in rural areas. The Global Fund
enabled the government to implement community-based delivery models, including malaria
management and DOTS, which increased the accessibility of care and treatment for many. In
addition, investments in Mutuelles de Santé - covering the fees for those who met the criteria for
poverty (regardless of HIV status) - helped reduce financial barriers to care for those without target
diseases as well.
156

The quantitative portion demonstrated a more robust referral system temporally associated with
the arrival of GHI monies. All of the facilities in Rwanda experienced a referral score i increase after
the introduction of GHI funding, which was statistically significant overall (p=.04). The average
change in referral score for the facilities in Rwanda was 3.5 points. Out of a possible score of 7, the
average referral score for Rwanda was 3.25 before GHI funding and 6.75 afterwards.
Community/Civil Society
GHI impact on health seeking behaviour
GHI support has increased health-seeking behaviour. Funding for Mutuelles de Santé contributed
to making health care accessible to poor and vulnerable populations, alleviating the financial
barrier. Patients who previously would have had to travel long distances to receive care are able to
find care closer to their homes given the increased presence of physicians in health centres. Both
GHIs have made significant efforts to sensitize the population about HIV and educate them on
prevention and treatment. With the availability of free HIV medications, some informants felt the
stigma surrounding HIV had decreased as communities saw individuals living with HIV retain or
regain their health. Demand for services, including primary care and timely presentation for TB
treatment, increased, along with the opportunity to screen for HIV and other diseases. In particular,
the increases of services in the health centres has impacted the demand for prenatal and delivery
services. As one informant explained:
Healthcare services are taken close to the population and the patients seek
treatment in time. Pregnant women deliver at health centres, family planning
methods are taught at health centres, and we also become able to easily teach
family planning methods in the community and men and women attend
teaching sessions at health centres for education and communication sessions, it
is also an advantage.
GHI impact on the focus and role of NGOs
Monies from both GHIs flow to NGOs. The CCM in Rwanda favours local and more established
NGOs and ensures that NGOs adhere to national priorities and coordinate their activities with
other providers. Some informants saw NGOs as a means of keeping the government accountable
as well; for example, one informant reported:
The NGO is also there to, I think, help make sure the government fulfils its
promise to the people, the rights of the people – the right to healthcare and
other rights. …The NGO is there to make sure the free market doesn’t screw up,
and the NGO is there to make sure, or help the government, let’s say, assist the
government, to provide these rights to people. And I think that PEPFAR and [the]
Global Fund have significantly allowed these NGOs to play a bigger role because
of these resources. But I think there’s probably also some negative effects, and
that you have to be careful. We don’t want to create parallel systems where an
NGO provides, in my opinion, a separate clinic and weakens the public health
sector.
PEPFAR, while it has the MOH verify that its overall plan is aligned with national priorities, makes its
own decisions about who to fund, and many informants felt that recipients of PEPFAR funding
were not all of high quality.
157

Some informants felt that local NGOs benefited greatly from GHI funding and partnerships with
international NGOs that had more resources and experience to share:
Discussion
Local NGOs, they benefited from the experience of international NGOs. They
learnt from them. It is like small soccer teams. When a small soccer team from the
village plays with a big experienced team, it does have the fear to lose the game,
however the small team learns from the big one. For sure local NGOs have learnt
a lot; we have noticed that in the area of how they now work, plan, and how they
organize themselves. This strengthens both the Rwandan society and the health
system. And we strengthen more the lower level institutions, it is very important
for us.
Rwanda is an example of how, with strong coordination at the national level, the money from GHIs
can strengthen health systems. Specific examples of the linkage between the stewardship of the
Rwandan government and the utilization of GHIs that were demonstrated by the qualitative
analysis are: the unified procurement system CAMERWA with medications beyond targeted
diseases funded by Global Fund with technical capacity building by PEPFAR. Both GHIs were
called upon to contribute to the national insurance system of “mutuelle de santé” which increased
access for targeted and non-targeted diseases. The national plan to spread and decentralize
services across the country was supported by both of the main GHIs in Rwanda. Lastly, the
Rwandan government successfully coordinates even international NGOs to work within the
national framework through the PEPFAR steering committee that was created by the government.
This strong national framework with donor collaboration with the clear intent of designing a
health system as a national priority and, in that context to deliver equitable, well distributed HIV
care, is a critical lesson to both national governments, international NGOs and donors. This
approach highlights the importance of government oversight and planning and stewardship in
the long-term success of such projects.
158

References
[1] World Population Prospects: The 2008 Revision Population Database [online database]. New
York; United Nations Population Division. 11 April 2009.
[2] Country Brief: Rwanda. World Bank, 2009
(http://web.worldbank.org/WBSITE/EXTERNAL/COUNTRIES/AFRICAEXT/RWANDAEXTN/0,,menuPK:
368714~pagePK:141132~piPK:141107~theSitePK:368651,00.html; accessed 7 April 2009).
[3] Rwanda Decentralization Assessment. United States Agency for International Development,
2002 ( http://pdf.usaid.gov/pdf_docs/Pnacr570.pdf; accessed 12 April 2009).
[4] Sekabaraga C. Rwanda Primary Health Care and Decentralization of Health System. In:
International conference on primary health care and health systems in Africa: towards the
achievement of health millennium development goals. Ouagadougou, Burkina Faso, 2008.
[5] World Development Indicators Online [online database]. World Bank. 8 April 2009.
[6] United Nations Development Programme Rwanda. Turning Vision 2020 into Reality: From
Recovery to Sustainable Human Development. National Human Development Report, 2007.
[7] United Nations Development Programme. Human Development Report 2007/2008: Fighting
Climate Change: Human Solidarity in a Divided World. New York: Palgrave Macmillan, 2007.
[8] Development Economics LDB Database [database]. Rwanda at a Glance 2008. The World Bank.
19 April 2009.
[9] Republic of Rwanda. Rwanda Service Provision Assessment Survey 2007. Kigali, National
Institute of Statistics, Ministry of Finance and Economic Planning, 2007.
[10] World Health Statistics 2008 [database]. World Health Organization. 19 April 2009.
[11] Country Situation: Rwanda. Geneva, Switzerland, UNAIDS, 2008
(http://data.unaids.org/pub/FactSheet/2008/sa08_rwa_en.pdf; accessed 19 April 2009).
159

Senegal: The effects of Global Health Initiatives on
the health system
Abstract
Papa Salif Sow, * Fatou Francesca Mbow, † Aliou Diallo, ‡ Demba Dione, ‡ Marième Ba ,‡
Senegal depends heavily on foreign assistance. In the health sector, however, Senegal is less
dependent on donor resources than many other countries in sub-Saharan Africa (SSA). Qualitative
and quantitative research on the interaction of GHIs and the health system in Senegal found an
important joint effort to ensure national ownership and to promote equitable access through
support for subsidies and promotion of increased service coverage. There are important GHIsupported
programmes that balance the weaknesses of the health system. Moreover, the funding
they provide is relatively independent of political changes and helps promote accountability by
linking funding to performance (although assessment of performance requires a sound
monitoring and evaluation (M&E) system which the researchers found to be weak). In some areas,
however, the interaction between GHIs and health systems was found to be less positive.
Promoting health system quality requires that clients be able to access physicians’ expertise if
desired. GHIs should also be better integrated with other global initiatives with significant
implications for population health, such as the environment.
Background
Senegal is bordered by Mauritania to the north; Mali to the east; and Guinea and Guinea-Bissau to
the south. It is home to 12.4 million people and averaged a 2.6% population growth rate from 2001
to 2007 [1]. Senegal ranked 153rd out of 179 countries on the UN Human Development Index in
2006 [2]. Since 1997, the country’s real gross domestic product (GDP) growth has averaged 4.5%
[3]. Official development assistance to Senegal was USD 824.9 million in 2006 [4]. The country’s
total external debt was USD 1.98 billion or 21.4% of GDP [3],[4].
Senegal has one of the lowest rates of HIV prevalence in sub-Saharan Africa (SSA), due in part to
the quick response from government and civil society after the first AIDS case was diagnosed in
1986. In 2007, there were 67,000 people living with HIV [5].
* Infectious Diseases Department, Fann’s Teaching Hospital, Dakar, Senegal (Study Coordinator)
† Independent Consultant, Dakar, Senegal (Senior Research Assistant)
‡ Ministry of Health, Dakar, Senegal (Research Assistant)
160

Senegal depends heavily on foreign assistance, which in 2007 represented about 23% of overall
government spending [6]. However, in health, Senegal is less dependent on donor resources than
are many countries in SSA.
Over the last ten years, government contributions to health system financing have become
increasingly important as has the contribution of Senegalese households through user fees.
Figure 1: Trends in Senegal’s health system funding - billions of XOF (1 USD= 500 XOF)
(by source of funding, 1998 to 2007).
Source: Draft PNDS- 03, November 2008.
Senegal’s public health system has swung between horizontal and vertical programming ever
since the French colonial administration began providing free medical services to all “indigenous
populations” in 1905. For most of the last decade the National Health and Social Development
Plan (“Plan National de Développement Sanitaire et Social,” or PNDS-1) has governed national
health strategy. Covering the years 1998-2007, PNDS-1 adopted a “horizontal,” integrative model
of health action with payment for health services as a part of its financing strategy. To complement
this policy, Senegal put in place a set of subsidy initiatives to reduce barriers that prevented access
for specific groups and service areas. The adoption of the Millennium Development Goals (MDGs)
within the two Poverty Reduction Strategic Papers (PRSP) produced by Senegal (2003-2005 and
2006-2010) paved the way for the 2009-2018 Plan National de Développement Sanitaire (PNDS-2).
PNDS-2 is strongly results-oriented, aiming toward the achievement of the MDGs through specific
vertical national programmes (including those for HIV, TB, malaria and vaccination). Funding for
some of these programmes is heavily dependent on support from GHIs.
Senegal’s performance on the achievement of the health-related MDGs shows some important
improvements, especially in child health. Reproductive health care remains the weakest area. HIV
prevalence has increased, but this could be due to decreased AIDS mortality in an era when
antiretroviral therapy coverage is now estimated at 68.8 % (it was 29.3% in 2004) [7]. TB indicators
reveal weaknesses in the national TB programme that only started benefiting from Global Fund
support in 2008.
161

Table 1: Basic Health System and Epidemiological Indicators*
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 12,400 2007 World Bank
Geographic Size (sq. km) 192,530 2007 World Bank
GDP per capita, PPP (constant 2005 international $) 1,572.90 2007 World Bank
Gini index 39.19 2005 World Bank
Government expenditure on health (% of general
government expenditure)
12 2007 WHO NHA
Per capita government expenditure on health (current US$) 29 2007 WHO NHA
Physician density (per 10,000)

President’s Malaria Initiative (in US$)
Year Amount Disbursed
2006 2,200,000
2007 16,7 00,000
2008 15,800,000
2009 15,700,000
TOTAL: 33,700,000
MAP (in US$)
Years Amount Disbursed
2002 30,000,000
TOTAL: 30,000,000
GAVI (in US$)
Disease Priority Amount Disbursed
Pentavalent vaccine 4,725,720 (2005) and 5,145,500 (2007)
Tetravalent vaccine N/A
Vaccine introduction grant 795,104 (2005)
Injection Safety N/A
Immunization services support 370,575(2004); 837,978 (2006); 507,843 (2007)
Health systems strengthening N/A
TOTAL: 12,382,722
Objectives and Methodology
This study aimed to better understand the nature and perceptions of GHI activity in different
settings and to document how GHIs have interacted with existing health system institutions.
The data collection for this study included semi-structured interviews, field observations, and the
examination of documentary material and collection of quantitative data. The study employed the
common data abstraction tool used across Maximizing Positive Synergies (MPS) sites. Data were
collected from various points in time both before and after the initiation of GHI funding. Purposive
sampling and snowball sampling were both used. Key informants included representatives of
major GHIs, relevant United Nations (UN) and government agencies, and personnel of private
sector institutions working on health at the central, regional and health district level (both forprofit
and non-profit). Within the very limited period of study, we secured interviews with 33
stakeholders. In some cases follow-up interviews were conducted.
Research assistants conducted field visits in each district, accompanied in some cases by the Senior
Research Assistant. A summary of the interview was provided to the Senior Research Assistant
after each interview and thematic areas were identified and reviewed on an ongoing basis. A
thematic qualitative analysis provided the bulk of the qualitative information documented in the
163

case study. Data analysis was conducted both within cases (health district) as well as across cases.
In view of the limited amount of data available no software was needed to make the analysis. An
exemption certificate was provided to the MPS team by the National Ethical Committee.
Given time constraints, the documentary review attempted to better understand the relationship
between wealth and health in the interaction between GHIs and health systems by selecting three
regions on the basis of relative affluence (one poor, one moderately poor and one rich region).
Selection criteria also embedded other important areas of analysis (for example the impact of
disease prevalence).
Table 3: Focus Health Districts
Health District Ziguinchor Fatick Dakar
Selection Criteria Highest poverty levels
Results
Highest HIV
prevalence rate
Moderately poor region Richest region
Majority of private and public health
structures.
Health policy makers as well as donors
heavily represented
Leadership and Governance
There are various mechanisms by which GHIs in Senegal work synergistically with the health
system to ensure national ownership of the funding process:
1. The Country Coordinating Mechanism (CCM) of the Global Fund.
2. The specific requirement that proposals for GHIs must be aligned with disease-specific
national strategies (all GHIs).
3. National Programmes as Principal Recipients.
4. Bilateral funding for AIDS is now channelled through the Global Fund (European Union
and French and Canadian Cooperation). This can be seen as strengthening national
ownership as it erases the political dimension of bilateral funding.
GHIs are seen as credible partners, on which national programmes can rely because their
disbursement mechanisms are not influenced by changes in the political arena.
164

Performance-based funding
In 2005, the Global Fund cut malaria grants in Senegal worth $7.1 million over systemic issues that
resulted in poor performance. The Fund later approved a grant proposal for malaria projects
submitted in Round 4.
The GAVI alliance receives yearly proposals and only funds those that show improvements from
the previous years. For every new child vaccinated, US $20 is added to the previous year’s funding.
Financing
GHIs began operating in Senegal in 2002. Until 2004, Senegalese households financed around 50%
of total expenditure on health through out-of-pocket spending at the point of consumption (the
most regressive and inequitable financing mechanism). In 2004, government and donors began
subsidizing access to key health services.
This policy change had a significant impact on equity. Voluntary Counselling and Testing for HIV
(VCT) were free from the outset, but scale-up in a number of sites really took off in 2003. Between
2003 and 2004 sites increased 4.5-fold and between 2004 and 2007, 3-fold; they are now in all
regions in Senegal. Research at the district level showed that in Ziguinchor, - the region and health
district most affected by the HIV epidemic (with twice the national HIV prevalence) - HIV testing in
the district increased nearly 12-fold between 2004 (166 HIV tests performed) and 2008 (1918 tests
performed).
Senegal was the first country in Africa to introduce antiretroviral treatment in 1998. Yet it was only
when ART became fully free of charge (without income-based contributions from users) that the
number of treatment sites and the number of patients on treatment peaked: from 20 sites to 70
sites between 2003 and 2007, and from around 20,000 clients on ARV in 2003 to around 70,000 in
2007 [8].
Service delivery
In 2005, nine sexually transmitted infection (STI) services specifically directed at men having sex
with men (MSM) were created within existing health care services in Senegal. This number rose to
12 in 2006 and 18 in 2007, and now covers 10 out of the 11 regions in Senegal. Syphilis and
Hepatitis B testing are provided free of charge since GHIs (Global Fund and MAP) started
supporting the Conseil National de la Lutte contre le SIDA (CNLS - National Council for the Fight
against AIDS).
This has increased the capacity (equipment, staff) of laboratories all over the country and
supported the scale-up of diagnosis of these illnesses in the country. For example, 1,478 syphilis
tests were performed in 2008 in Ziguinchor (the health district in Senegal with the highest HIV
prevalence); none were performed in 2004. DTP vaccination coverage increased from 52% in the
year 2000, to 87% in 2004 [9] and 88% in 2008. [10] GAVI has supported free DTP3 coverage in
Senegal since 2002.
“Parrainage” or mentoring programmes, whereby Dakar-based “ART mentors” from a tertiary level
treatment structure are responsible for providing technical support to each region in Senegal, was
key in Senegal’s impressive decentralization efforts in ART coverage [11].
Between 2000 and 2004, a household survey with standardized sampling and measurement
methods to compare four countries at two points in time, showed that in Senegal the percentage
165

of respondents aware of treated nets rose from 70% to 97.3%, and the number of households
owning a bednet rose from one third (33.6%) to half (56.1%) [12].
Information
Although all key informants felt that monitoring and evaluation were strengthened through GHIs,
national programmes were unable to provide researchers with district-level data. This lack of data
availability shows an existing weakness in the M&E system for GHIs too, and focuses attention on
the quality of the data published at national level.
One of the most disappointing, yet instructive, research outcomes was the poor performance of
the country monitoring and evaluation system. Few data trends could be gathered from
Ziguinchor (2004-2008), no significant data trends could be discerned for Mbao (2007-2008), and
no data trends at all for Fatick (only 2008 data were available). This prevented researchers from
analysing the relationship between median income or regional disease prevalence on the
performance of GHIs. The results of this study are therefore mostly based on the qualitative
interviews, the data gathered during those interviews and the extensive review of the literature
performed by the MPS team.
Discussion
The outcome of the quantitative data collection process at district level was disappointing: only
one of the three focus Health Districts, Ziguinchor, provided data which allowed some trend
analysis on selected MPS project indicators. Nonetheless, the qualitative interviews of a diverse
range of individuals involved in GHIs at all levels of the health system, as well as the extensive
literature search on GHIs and health systems in Senegal, yielded a wealth of information on how
GHIs and health systems interact in this country.
In Senegal, the interaction between the GHIs and the health system included an important effort
to ensure national ownership of projects and programmes funded through GHIs (particularly by
the Global Fund); and improvements in equity of access through subsidies and the expansion of
service coverage. Here, the strengths of the GHIs balance the weaknesses of the health system,
providing funding that is relatively independent of changes in political administration and linking
funding to performance. An adequate performance assessment, however, requires a sound M&E
system, which the MPS research highlighted as a weak component of the health system in
Senegal.
Some outcomes of the interaction between GHIs and health systems were less positive. While
GHIs have encouraged task shifting, it would be desirable to promote the training and recruitment
of physicians in Senegal. Efficiency requires quality and the Senegal MPS team advocates for the
development of a health system where clients can access physicians’ expertise if they wish to.
Efficiency also calls for maximizing the use of opportunities of synergistic funding, and the Senegal
MPS team proposes that GHIs create better links with other global priority issues (such as the
environment), for which substantive funding has been earmarked and for which the linkages with
health are clear.
166

References
[1] The World Bank Group. Senegal Data-at-a-Glance 2008.
[2] United Nations Development Programme. Human Development Reports: Senegal.
(http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_SEN.html).
[3] The World Bank Group. Senegal at a glance.
(http://devdata.worldbank.org/AAG/sen_aag.pdf).
[4] The World Bank Group. World Development Indicators (WDI) Online
[5] UNAIDS. 2008 Report on the global AIDS epidemic 2008.
[6] US Department of State. Background Notes: Senegal
(http://www.state.gov/r/pa/ei/bgn/2862.htm).
[7] Senegal UNGASS Report 2008
(http://data.unaids.org/pub/Report/2008/senegal_2008_country_progress_report_fr.pdf).
[8] UNAIDS, Senegal Country Report 2008
(http://data.unaids.org/pub/Report/2008/senegal_2008_country_progress_report_fr.pdf).
[9] WHO Global Summary 2006 (http://www.who.int/vaccinesdocuments/GlobalSummary/GlobalSummary.pdf).
[10] Personal Communication, GAVI Focal Point, Ministry of Health, Dakar, Senegal(4/30/2009).
[11] Papa Salif Sow. Country Review: Senegal. Chapter 24 of the “A Decade of HAART” Edited
by Jose M Zuniga, Alan Whiteside, Amin Ghaziani and John G Bartlett (Oxford University Press).
[12] Carol A Baume and M Celeste Marin. Gains in awareness, ownership and use of insecticidetreated
nets in Nigeria, Senegal, Uganda and Zambia. Malar J. 2008; 7: 153.
(http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=2527013&blobtype=pdf).
167

South Africa: The effects of Global Health Initiative
funding for HIV/AIDS on the health
system
Abstract
Thomas Bisika, 24 Eric Buch, 24 Thubelihle Mathole, 25 Annie Parsons, 25
David Sanders 25
The study seeks to understand how GHIs operate in South Africa and their effects on the health
system. GHI funding mostly focuses on HIV/AIDS; since apartheid, the South African health system
has aimed to ensure equal access to health services.
Preliminary results presented in this paper rely on interviews conducted as part of qualitative data
collection at national level. Secondary data sources were used where available. Data analysis was
framed by the WHO building block thematic areas.
External financing centres on HIV/AIDS and is considered additional to state spending. The
government is concerned about sustainability as such funding is not part of its long-term budget
plans.
GHIs have positively contributed towards HIV service scale-up by providing trainings and staff.
However, health workers in the public sector are generally struggling with increased workloads.
Historically, donors and government have failed to coordinate activities, with donors competing
for attribution of outcomes. This is improving, however, with common indicators being developed
to minimize parallel reporting systems.
HIV/AIDS is a challenge to South Africa, but well-meant interventions should not undermine
government efforts. Effectively utilized, GHIs can increase health care coverage and improve
access to health care in South Africa.
Background
The Republic of South Africa is bordered to the north by the Republics of Namibia, Botswana, and
Zimbabwe; to the east by the Republic of Mozambique and the Kingdom of Swaziland; while the
Kingdom of Lesotho is surrounded by South African territory. South Africa is home to 47.6 million
people and averaged a 1.0% population growth rate from 2001 to 2007 [1]. South Africa ranked
125 th out of 179 countries on the UN Human Development Index in 2006 [2]. The country has
experienced substantial economic growth in 15 years of democracy following the end of apartheid
24 University of Pretoria
25 University of Western Cape
168

(4.8% real GDP growth in 2007), and is seen as a major emerging economy [3]. However, high rates
of unemployment (38.8% in 2005) and income inequality pose major social and economic
challenges [4,5].
First identified in South Africa in 1982, HIV/AIDS has had a major impact, and has helped fuel the
spread of TB, including multi-drug resistant TB (MDR-TB). Recent years have seen a growing
incidence of extremely-drug resistant TB (XDR-TB) [6,7].
Forty percent of total health care expenditure in South Africa flows through the public health
sector, which in turn serves around 80% of the population [8]. The current South African health
system features free health care for vulnerable groups (particularly pregnant women, children
aged less than six years, the disabled and the elderly), waivers for the poor, and free primary health
care for all [9]. In 2006, 66% of South African doctors worked in private practice, with the remaining
third working in the public sector [10]. In 2007, 26% of the government’s HIV/AIDS budget was
externally financed though total donor aid was less than 1% of South Africa’s overall health budget
[11].
Official development assistance to South Africa was US$ 717.8 million in 2006 [12]. The country’s
total external debt was US$ 35.5 billion or 13.9% of GDP [12].
Table 1 Basic Socioeconomic, Demographic and Health Indicators (*)
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 47,588 2007 World Bank
Geographic Size (sq. km) 1,214,470 2007 World Bank
GDP per capita, PPP (constant 2005
international $)
9,191.38 2007 World Bank
Gini index 57.78 (†) 2000 World Bank
Government expenditure on health (% of total
government expenditure)
9.1 2006 WHO NHA
Per capita government expenditure on health
at average exchange rate (in US$)
176 2007 WHO NHA
Physician Density (per 10,000) 8 2004 WHO SIS
Nursing and midwifery density (per 10,000) 41 2004 WHO SIS
Maternal mortality ratio (per 100,000 live
births)
400 2005 WHO SIS
DPT3 coverage (%) 97 2007 WHO SIS
Estimated adult HIV (15-49) prevalence (%) 18.1 (15.4-20.9) 2007 UNAIDS
Estimated antiretroviral therapy coverage (%) 28 (22-36) 2007 WHO/UNAIDS/UNICEF
Tuberculosis prevalence (per 100,000) 998 2006 WHO GTD
Estimated malaria deaths, all ages 146 2006 WHO WMR
† 2000 Gini data provided to facilitate cross-country comparison. South Africa’s Gini figures have increased since 2000.
South Africa’s Gini coefficient stood at 0.73 in 2005 (Source: Statistics South Africa. Income and expenditure of households
2005/2006: Analysis of results. Pretoria, Statistics South Africa, 2008).
169

Table 2 Global Health Initiative Investments
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 1, HIV/TB 62,476,536 49,771,823
Round 1, HIV/TB 20,226,665 20,226,665
Round 2, HIV/TB 24,400,220 12,579,554
Round 3, HIV/AIDS 66,501,629 62,190,178
Round 6, HIV/AIDS 55,071,906 24,927,005
TOTAL: 228,676,956 169,695,225
PEPFAR
Year Amount Allocated (in US$)
2004 89,272,988
2005 148,187,427
2006 221,539,430
2007 397,777,008
2008 590,897,685
TOTAL: 1,447,674,538
Objectives and Methodology
Little is known about how the shift from traditional funding mechanisms towards global
approaches has impacted on health systems in recipient countries, nor how donor harmonization
initiatives have interplayed with GHIs at country level. The study therefore seeks to understand
how GHIs operate in South Africa and their outcomes in terms of the functioning of South Africa’s
health system. This study is part of a multi‐country study being conducted in five African countries
(Angola, Burundi, Lesotho, Mozambique and South Africa), with support from three European
institutions in Ireland, Belgium and Portugal. Additional support was received from the WHO
Maximizing Positive Synergies Group and Harvard University.
The study was descriptive in nature and used a qualitative research methodology, which enabled
the research team to uncover any unforeseen concerns of the study population and obtain a
deeper understanding of the subject under study. Data were collected through document analysis
and individual interviews after obtaining ethics approval from the University of Western Cape
Senate Research and Ethics Committee. Three levels of data collection are planned to reflect
national, provincial and facility-level diversity. Data presented in this paper reflect work done at
national level.
Individual interviews were conducted with 15 participants at national level. Participants were
purposefully selected from among key staff involved in GHIs at national level as well as amongst
in‐country staff of donors and recipient nongovernmental organizations (NGOs). Senior
government officials and policy makers at the National Department of Health (NDOH)
headquarters were interviewed. An attempt was made to maintain a balance across the different
categories.
170

The collection of information from different categories of respondents from diverse organizations
helped the team to assess whether there was consensus on the issues raised. The interviews
followed a question guide, with the scope of the questions focused on participants’ experiences
and views. The interviews lasted between 45 and 90 minutes, differing slightly as the interviewer
probed according to what was said by each interviewee. The interviews addressed both what had
been implemented with GHI support but also stakeholder perspectives on likely implications for
the broader health care system.
The purpose of the interview and the freedom to participate or withdraw if they so chose, were
verbally explained to each participant. Most interviews were carried out at the interviewees’ offices
while others were telephone interviews. All interviews were in English. Some of the interviews
were tape recorded, but where this was not possible detailed notes were taken. The interview
format was refined periodically in order to probe and explore leads on key issues that were not
clear.
Data analysis was an ongoing process using interpretive description as described by Thorne and
others [13]. Audio taped interviews were transcribed and analysed, with key themes identified
through this process. The consistency and reliability of the analysis process was improved through
the involvement of more than one analyst [14].
Results
Leadership, Governance and Coordination
South Africa’s position as a middle-income country with some of the highest HIV prevalence rates
in the world places it in the unique position of being able to fund its own health care system, while
receiving large amounts of external funding for HIV-specific services. Current GHI and donor
activity largely focuses on HIV/AIDS, though reliance on the state to coordinate and monitor
activities has had varied results. Most funding centres on HIV‐related areas such as home based
care, training and support of community health workers, supporting the rollout and scale‐up of
ART services, and the funding of community‐based HIV prevention organizations. Many clinically
based interventions are operationalized in conjunction with or through national and provincial
departments of health.
State health‐related policy guidelines in South Africa are generally strong, with strategic plans
regularly crafted to cover various areas [11]. Government policies currently include amongst
others: the HIV and AIDS and STI Strategic Plan 2007‐2011; the National Infection Prevention and
Control Policy for TB, MDRTB and XDRTB; and the Comprehensive HIV and AIDS Care, Management
and Treatment Plan for South Africa. In addition to specific health‐related policies, state planning
follows the principles of the Reconstruction and Development Program (RDP) and other
macro‐economic strategies. South Africa’s federal governance system gives provinces the ability to
interpret national policy according to their own priorities, with funding from the national level
allocated according to need. Post-apartheid restructuring of the public health sector has focused
significantly on promoting equity.
Donors are also expected to follow state guidelines, and many do seek to align with government
priorities and policies. However, the imperative of addressing focal diseases, along with local
political conditions, has led to most donor‐funded projects being concentrated in specific regions
171

of South Africa (GHI2, ND6). 26 Donors have also tended to concentrate on supporting existing
projects rather than financing new initiatives: as a result, few rural areas have health services other
than those publicly provided by the state.
Disparate government viewpoints on HIV/AIDS have historically affected the willingness of
donors to follow state guidance. Former high‐ranking officials gained international notoriety in
2000 for questioning the link between HIV infection and the development of AIDS‐related
conditions. Though national and provincial departments of health continued to treat AIDS‐related
conditions as outcomes of HIV infection, this official unwillingness has been cited as an important
reason for the relatively slow roll‐out and scale‐up of ARV treatment in South Africa and an
estimated 365,000 AIDS‐related deaths [15]. An additional complication was that the NDOH saw
attempts to focus on a single disease as contrary to efforts to improve equity (D1).
Programmatic and funding oversight of externally funded projects by the South African
government relies heavily on local conditions: overall, the state lacks capacity, skills and
experience (ND3, ND6). One GHI respondent noted that the offer of external assistance requires
the recipient to provide the administrative capacity to handle such funding, and this is not a focus
of the NDOH (GHI3). In addition, NGOs and provincial government are not required to apply to the
NDOH for approval before submitting funding proposals or accepting funds, limiting national level
oversight (ND2). These issues provided an incentive for some external donors to work directly with
NGOs where possible and bypass government structures (GHI1, D1), though this is reported to
have improved in the last few years (ND3, ND5).
At times, different reports on similar programmes or different statistics on the same issues are
presented at the same international fora by agencies implementing projects in South Africa. In
some instances, the NDOH itself only gets to know on such occasions about programmes
implemented in South Africa (ND5, ND6). Competition among donors has on occasion led to
attribution becoming an issue (ND2).
Participants in the study cited the desire to see tangible results or outputs that could be
definitively matched to funding and resource inputs. This would require direct linkages between
proposals, implementation plans, and monitoring and evaluation. Strategic planning at a national
level for NDOH funded projects has been ongoing for a number of years (ND6), but coordination
with non-state actors has varied. State institutions such as the South African AIDS Council (SANAC),
the NDOH’s Donor Coordination Forum, the National Treasury’s International Development
Cooperation Unit (IDCU), as well as donor bodies such as the EU+ Forum are some of the initiatives
mandated to coordinate functions amongst the different GHIs, donors and government
institutions. Donor coordinating units are also located within provincial and local levels of
government. The actual operation of these initiatives varies widely.
Historically, the Donor Coordination Forum has not succeeded in meeting on a frequent basis.
SANAC lacks the legal powers to enforce its decisions as it was not set up by an Act of Parliament.
Most participants agreed that it was the responsibility of government to ensure SANAC had the
authority to address donor‐related concerns. The importance of a government institution that
holds everybody accountable at every level was highlighted. For example, it was decided two
years ago that ARV treatment should be integrated into general clinical services and that
stand‐alone ARV clinics should no longer exist. However, stand‐alone ARV clinics are still found in
some parts of the country (ND6).
26 The anonymity of each interview was ensured using the following codes: National Department of Health personnel
are ‘ND’, donors are ‘D’ and GHIs are ‘GHI’.
172

Both NDOH and donor participants observed a lack of joint activities or reviews amongst the major
donors operating in South Africa, though frequent meetings between the NDOH and HIV-related
donors have recently been revived. Donors have been asked by the NDOH to report on their
activities in different parts of the country and the NDOH is looking to build its liaison capacity with
GHI and donor assistance.
There was a general call from respondents in the study for a donor database for South Africa. Some
major GHI funds go directly to NGOs or CSOs without any interaction with government, and it is
therefore difficult to trace donor activities (ND5). Such donors have independent reviews and
evaluations, reporting directly to their head offices in or out of country (GHI3, ND6).
Financing
Most external financing for health‐related issues in South Africa is disease‐specific and centres on
HIV/AIDS. Though malaria and TB receive attention, malaria-related efforts are geographically
limited and TB efforts tend to relate to HIV/AIDS.
The overall health expenditure of the NDOH in 2007/2008 was roughly US$ 1,823 million
(approximately US$ 1 = SAR 7 in 2008). Around 19% was spent on the HIV/AIDS and STIs
programme, with US$ 1,3 million expended on TB control and management [16].
Donor expenditure on HIV/AIDS in South Africa in 2007 – excluding PEPFAR – was equivalent to
around a quarter of the total amount spent by the NDOH. PEPFAR’s allocated funding for South
Africa was US$ 1.4 billion from 2003 to 2008, with the figure for 2007 over 100% of the amount
spent by the NDOH in that year and equivalent to 83% of all government expenditure on
HIV/AIDS‐related projects [17].
Most donor support for clinical services is channelled through the government, except for PEPFAR.
As of November 2008, the South African government supported 630,775 patients across 259 sites
on ART as part of its regular health budget [18]. PEPFAR presently finances and supports ARV
treatment for 30,000 patients in South Africa independently of the South African government
(GHI1). No external financing is used for budget support, with all donor-funded projects in South
Africa regarded as additional monies (ND6).
External financing for malaria‐related programmes began in 2003 with a round 2 grant from the
Global Fund as part of a multi‐country initiative between South Africa, Mozambique and
Swaziland, with South Africa’s Medical Research Council (MRC) acting as the implementing agency
[19]. As of 2006, the NDOH is again the sole funder of MRC South African‐specific activities in
Limpopo, Mpumalanga and KwaZulu Natal (D3).
Donor support of TB activities tends to focus on capacity support for integrated HIV and TB
management. The Global Fund has funded monitoring and evaluation training and the setting up
of MDR TB centres, with the EU providing training to senior and middle management in
government and NGOs, as well as the training of staff on TB treatment defaulters. Belgium has
provided similar support (ND4). Research is another key area that has been supported by GHIs. For
instance, the CDC provides grants to the MRC and academic research institutions for studies on
HIV and TB. Some hospitals have also had equipment and infrastructure used in the diagnosis and
treatment of TB refurbished (ND4).
The NDOH, independently and in its role as principal recipient of Global Fund round 6 funds NGOs
that work on HIV and TB. However, there have at points been concerns about the ability of the
173

NDOH to disburse external funds in the timeframe required by donors, in particular for HIV‐related
activities. In December 2008, the NDOH failed to channel a received Global Fund disbursement to
NGO recipients, with the Global Fund citing lack of required development around disbursement
capacity within the department.
Health Workforce
While GHIs do not provide direct funding for human resources, this study’s preliminary results
show that their interventions have significant impact on human resources for health at all levels.
South Africa, like many other African countries, is currently experiencing a general shortage of
skilled personnel in the public health sector, which is exacerbated by the problem of a high staff
turnover. The scale‐up of services associated with HIV‐related programmes has in general led to
increased workloads for existing health care workers. A GHI’s offer of technical assistance may
often exacerbate existing capacity issues by drawing away staff for training or concentrating
facility efforts on one set of health issues at the expense of others (ND2). It was noted by one donor
that scale‐up had been further hampered by government backtracking on the agreement that
nurses would be able to initiative ARV treatment, particularly in areas with few physicians (D2).
Both government and some GHI representatives acknowledge the problem of insufficient health
workers. For example, six out of nine TB/HIV coordinators trained by the NDOH in 2006 were
reported to have left the department by February 2009. Such turnover usually disrupts the
management and implementation of projects, with additional time needed to orient and train new
staff. Some donors have also seconded and trained staff on TB and HIV management who later
leave and join the private sector.
Informants linked mobility from government to GHI or donor‐funded programmes, especially ART‐
related projects, to remuneration differences and frustration with government’s HIV/AIDS policies.
Though GHI‐funded projects pay the same salaries as the public sector, there are fewer deductions
and therefore the take‐home amount “looks” higher (GHI1, ND2). Poor working conditions have
contributed to dissent in the public sector.
To address high staff turnover, the NDOH has developed staff retention strategies that encourage
staff to stay and also make working in the public sector more attractive, especially in rural areas.
However, these are targeted at clinical workers and have failed to increase supervision capacity
(ND2). Some donors work closely with government to identify gaps and second staff to
government projects for a specific period of time. Government public sector policy allows for GHI
funds to be used for short-term consultants, but such funds cannot support public sector salaries
(ND2). GHIs will occasionally negotiate with state structures to ensure higher salaries for staff
within the government salary structure (D2).
Donor funding has also been used to provide community health workers to CSOs, either
independently of government or through state health structures (GHI1, ND6). Overall, the
provision of specific training by GHI‐funded projects is regarded by the NDOH as having had
positive effects on HIV/AIDS service delivery (ND3). However, GHI-provided training tends to focus
on specific topics and fails to contribute towards an integrated approach to health care (ND3).
In these circumstances, human resource (HR) planning becomes a critical requirement, with
forward planning a crucial skill in the context of ever increasing numbers of TB and HIV/AIDS
patients. The NDOH acknowledges that there is minimal ability in the department to plan and
forecast HR needs. One respondent described HR managers under NDOH jurisdiction as generally
personnel managers who are only interested in personnel issues such as salaries and leave days
174

(ND1). Some donors have identified districts without this expertise and respond by seconding staff
to those specific districts (GHI2).
The state also lacks proposal-writing capacity in applying for Global Fund grants, though it has
received technical support from UN agencies. No direct technical support has been forthcoming
from the Global Fund, with technical expertise being imported from other countries instead of
training local people in an effort to build capacity (ND5).
Medical Products, Vaccines, and Technologies
Donor funding for medical technologies has focused on the provision of disease‐specific
pharmaceuticals and equipment, as well as support for the strengthening of the pharmaceutical
regulatory system. In the past decade there has been a notable improvement in diagnosis and
treatment practices for malaria. Global Fund funding enabled the MRC in 2000 to introduce a
diagnostic test that reduced the time required for results from one or three days to ten minutes
(D3). This has improved treatment practices as patients receive a definite diagnosis and initiate
treatment in one visit. The enhanced malaria treatment and control measures resulted in large
reductions in malaria incidence rates in the highest malaria risk border areas of South Africa,
Swaziland and Mozambique. Malaria incidence in this region decreased from 250 per 1000 in the
1999/2000 baseline year to less than 20 per 1000 in 2006/2007 [20]. Malaria incidence has
dramatically reduced by over 99% in KwaZulu Natal and over 86% in Mpumalanga (D3).
Donors generally regard South African’s pharmaceutical regulatory system as strong, though
PEPFAR procures ARVs independently of state supply channels (GHI1, ND5). Both DFID and the
CDC fund training initiatives for state pharmaceutical services at national and sub‐national levels
in South Africa. DFID, together with the WHO, has supported pharmaceutical regulation at national
and provincial levels since 1997 through the South African Drug Action Program (D5). However,
strong policies do not necessarily translate into effective supply management.
Donors work with the South African government to improve its laboratory capacity, and in
introducing new diagnostic techniques (GHI1). These efforts are partly directed towards
diagnosing and treating MDR and XDR TB cases, with the collaborative Regional International
Training and Research Centre in operation since 2006.
South Africa is also one of the centres for HIV vaccine research in Africa, though much of the
funding for these initiatives is channelled through the MRC or NGOs, and not the NDOH.
Information
Health services in South Africa are under pressure to develop information systems that will
respond to the challenges of managing the public health care sector and ensure accountability.
Our data show that there is an increasing demand for health information to inform policies (D1),
priority setting (GHI2), resource allocation, and to guide impact assessments of health
programmes (ND4). The lack of clear national policy and guidelines, common indicators or
common reporting systems, and little feedback to staff at lower levels of health care were
mentioned as negatively affecting the quality of data. The NDOH is aware of these issues and
recently organized a meeting to develop common indicators for integrated TB/HIV management
(ND4). Informants identified severe shortages in health information skills and expertise as another
problematic issue.
175

The lack of effective monitoring and evaluation systems has meant difficulty in linking resources to
outcomes. In part, data collection is the result of GHIs and their implementing partners’ diverse
and independent information requirements. This has resulted in the duplication of data collection
and also adds to the burden of work, with potentially negative effects on the quality of data
produced. One major HIV-related donor only uses information collected by its own programmes at
district level (ND6), which it has recently begun sharing with the NDOH. One participant reasoned
that if all monitoring and evaluation were put together and everybody held accountable for their
work, actual project outcomes would be different (GHI3).
During the research, NDOH representatives expressed gratitude for donor support in improving
information systems and monitoring and evaluation systems. GHIs and donors have begun
working closely with government to develop health information systems. This involves developing
software, monitoring and evaluation systems, as well as the secondment and training of
information officers and monitoring and evaluation officers, along with supplying computers.
There has been a subsequent noted improvement in the quality of data collected through the
NDOH.
Discussion
The data presented in this paper are preliminary results at national level of an ongoing study on
the effects of GHIs on health systems in South Africa, based on qualitative methods and thus
relying on the perceptions of interviewees. The paper also draws on secondary data from
documents obtained during the data collection process. Data analysis and presentation for this
paper were guided by the WHO building blocks for health systems strengthening framework.
Increasing levels of funds are being spent by international bodies on health in under‐resourced
countries in amounts not previously seen. South Africa is unique in that the amounts coming in are
largely confined to addressing HIV/AIDS and do not dwarf the government’s own contributions to
health. However, the government remains concerned about the sustainability of such externally
financed projects as the money is not contained within its own budget. This reflects global
concerns around the potential long-term effects of such vertical funding on public health systems
in general [21].
Our data suggest that funding specific programmes without building overall capacity in the
recipient country by improving the public health system is not ideal. It is necessary to have a
well‐functioning health care system, align funds with the state, and ensure commitment from all
stakeholders.
The study found that there has historically been some level of competition amongst GHIs, which
view themselves as primarily accountable to their head offices and funding structures. Each GHI
therefore produces separate reports, often based on individual data collection by recipient
organizations. An evaluation of the district health information system in KwaZulu Natal found that
though there was some improvement in the collection of data at primary care level, the quality of
data was still poor and staff were not able to effectively use it [22]. Absent or weak accountability
practices, including poor information systems and weak monitoring and evaluation mechanisms
undermine effective implementation. Parallel reporting systems still exist, with duplication in the
reporting systems of different stakeholders at district level. Strong leadership is therefore needed
to enforce what Garrett describes as “cooperative thinking” and guidance [21]. This is not unique
to South Africa, with other countries in Southern Africa, such as Zambia, reporting coordination of
donor activities as a significant challenge [23].
176

It appears there is limited progress in translating global commitments on aid effectiveness into
concrete action at country level – principally in relation to reducing aid fragmentation, in line with
the Paris Declaration on harmonization and aid effectiveness. Even though South Africa has policy
guidelines on donor harmonization and alignment, the level of coordination among GHIs and
other donors is still low. Some do not work closely with government in planning, setting priorities,
implementing, monitoring and evaluating their programmes. Little attention has been paid to the
administrative and programme costs of either coordination or non-coordination. Parallel reporting
systems still exist, with duplication in the reporting systems of different stakeholders at district
level.
That said, there have been noted improvements of late in efforts at harmonization, alignment and
health system strengthening. These latest developments should be commended and encouraged,
with more support given to those spearheading the process. Bigger GHIs such as PEPFAR and
Global Fund should be encouraged to lead by example, by considering joint activities at country
level and examining how to supplement each other’s efforts. The prioritization of coordination by
government might be better served by integrating it as a line function of senior officials in the
NDOH.
One common concern amongst GHIs, other donors and government is the lack of sufficient health
care workers in South Africa, particularly in the public sector. Lack of health workers has been
recognized as a major constraint in the scale-up of HIV/AIDS programmes in both this study and
others [23, 24]. Competition for available labour by the public sector and the private sector,
including GHIs and other donors, as well as by overseas recruitment agents, increases high
mobility among more experienced, skilled labour in focal areas such as TB and HIV management.
The number of vacant posts in the public sector is still very high [17], though lack of HR forward
planning means this calculation is often based on out-of-date projections.
Despite severe staff shortages and migration of skilled labour in the health sector, there is no
available evidence of increased investment in overall human resources development from
government, GHIs or other donors, with HR efforts concentrated in specific focal areas. There has
been little support given to developing and producing more skilled labour, except for short-term
training in specific areas such as information systems, monitoring and evaluation and
management of TB and HIV/AIDS, among others. Evidence shows that the health workforce drives
a health system’s performance, and the production of more skilled labour as well as better
distribution and retention of existing workers are necessities [24].
GHIs and donors should aim to strengthen training institutions over the long-term, with shortterm
efforts concentrated on means of improving capacity without increasing workloads. Training
should include non-focal issues that indirectly impact on areas of concern, such as financial
management, HR forward planning, and the importance of monitoring or evaluation activities at
facility-level. Government should interact with such efforts to ensure comprehensive skills building
within facilities and in training institutions. Means of improving and sustaining effective
supervision should be investigated by the NDOH.
Weak information on the health workforce in South Africa hampers planning, policy and
programme development, and implementation [24]. Participants in this study acknowledged that
health workforce‐related information is still sparse and fragmented in South Africa and
emphasized the need for forward planning to develop HR plans that will accommodate contextual
challenges experienced by the country. The problems experienced by all parties demonstrate the
importance of partnership by government with all stakeholders involved in HR issues, including
GHIs and donors implementing different programmes across the country, as well as labour unions
and research institutions that collect significant amounts of data on HR. Therefore, a reliable and
regularly updated HR database for South Africa should be developed.
177

Despite these problems, GHI funding has helped vastly expand access to life‐saving ARV treatment
for thousands in South Africa. Of the total estimated number of people in need of treatment, 55%
were enrolled in the ART programme by the end of 2007 [25]. In addition, a steady increase in the
amount of health funding from both government and donors for malaria has led to a drastic
reduction in malaria incidence rates in the highest risk areas of South Africa. The cooperative
approach adopted by South Africa, Swaziland and Mozambique has reduced overall incidence
rates by more than 85% in the region [20].
Though South Africa has one of the most comprehensive private health sectors in the region, its
public health sector caters for 80% of the population and faces a number of challenges [17]. The
issue of equity in health care hence remains a challenge in South Africa. The inclusion of GHIs and
donors in government strategic planning could increase their ability to provide complementary
projects that strengthen comprehensive health services in poorly resourced areas while avoiding
the potential pitfalls of budget support. Our study concluded that, effectively utilized, donor and
GHI funding can provide a means of improving equity in access to health care for all South
Africans.
178

References
[1] The World Bank Group. South Africa at a glance. Washington, DC, World Bank, 2008
(http://devdata.worldbank.org/AAG/zaf_aag.pdf; accessed 19 May 2009).
[2] United Nations Development Programme. Human Development Reports: South Africa. New
York, UNDP, 2008
(http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_ZAF.html; accessed 19
May).
[3] Department of Health, Medical Research Council, OcrMacro. South Africa Demographic and
Health Survey 2003. Pretoria, Department of Health, 2007.
[4] Kingdon G, Knight J. Unemployment in South Africa: a microeconomic approach.
http://www.csae.ox.ac.uk/resprogs/usam/default.html; accessed 27 February 2009.
[5] South African Institute of Race Relations. South Africa Survey 2007/2008.
[6] AVERT. HIV and AIDS in South Africa (http://www.avert.org/aidssouthafrica.htm; accessed 27
February 2009).
[7] World Health Organization. Global Tuberculosis Database
(http://www.who.int/globalatlas/dataQuery/default.asp).
[8] SouthAfrica.info. Health care in South Africa
(http://www.southafrica.info/about/health/health.htm; accessed 27 February 2009).
[9] McIntyre D, Garshong B, Mtei G, et al. Beyond fragmentation and towards universal coverage:
insights from Ghana, South Africa and the United Republic of Tanzania. Bulletin of the World
Health Organization. 2008;86:871-876.
[10] Benatar S. Health Care Reform and the Crisis of HIV/AIDS in South Africa. New England Journal
of Medicine. 2004(351):81-92.
[11] OECD. 2006 Survey on Monitoring the Paris Declaration, South Africa. Paris, OECD, 2007.
[12] The World Bank Group. World Development Indicators (WDI) Online.
[13] Thorne S, Kirkham SR, O’Flyn-Maggie K. The Analytic Challenge in Interpretative description.
International Journal on Qualitative Methods. 2005, 3:1-21.
[14] Mayan MJ. An Introduction to Qualitative Methods; A training module for students and
professionals. Alberta, International Institute for Qualitative Methods, 2001.
[15] Dugger CW. Study Cites Toll of AIDS Policy in South Africa. New York Times. 25 November
2008.
[16] South African Government. National Department of Health. 2009.
[17] AIDSMark. Regional Lessons Learned: Southern Africa. Washington DC, PSI, 2007.
[18] South African National Department of Health. Treasury Budget 2009.
179

[19] The Global Fund for Aids Tuberculosis and Malaria. Portfolio of Grants - South Africa & the
Global Fund to Fight AIDS, Tuberculosis and Malaria
(http://www.theglobalfund.org/programs/portfolio/?lang=e&countryID=SAF; accessed 19 May
2009).
[20] Maharaj R. Annual LSDI report. Durban, MRC, 2007.
[21] Garrett L. The Challenge of Global Health. New York/Washington, DC, Council on Foreign
Relations, 2007.
[22] Garrib A, Stoops N, McKenzie A, Dlamini L, Govender T, Rohde J, Herbst K. An evaluation of the
District Health Information System in rural South Africa. S Afr Med J 2008, 98:549-522.
[23] Hanfield J, Musheke M. What impact do Global Health Initiatives have on human resources for
antiretroviral treatment roll-out? A qualitative policy analysis of implementation processes in
Zambia. BioMed Central. 2009, 7:8-11.
[24] Chen L, Evans T, Anand S, Boufford JI, Brown H, Chowdhury M, Cueto M, Dare L, Dussault G,
Elzinga G, Fee E, Habte D, Hanvoravangchai P, Jacobs M, Kurowski C, Michael S, Pablos Mendez A,
Sewankambo N, Solimano G, Stillwell B, De Waal A, Wibulpolprasert, S. Human resources for
health: overcoming the crisis. Lancet, 2004, 364: 1984-90.
[25] TAC. Key HIV Statistics (http://www.tac.org.za/community; accessed 6 March 2009.
180

Uganda: Expanding Targeted Services Into
Primary Health Care*
Abstract
Dr. Bernard Michael Etukoit1, Mr. Richard Wanyama1
1 The AIDS Support Organization
Several GHIs have funded programmes in Uganda since 2000, most recently the Global Fund and
PEPFAR. Overall, the GHIs have demonstrated system-wide benefits, but in some instances have
negatively impacted health systems. The study explored interactions between GHIs and health
systems. The research aim was to generate information to inform health policy and programming
from the knowledge, experience and perceptions of personnel at GHI-supported facilities. The
study was cross-sectional, employing both quantitative and qualitative methods.
GHIs were associated with improved service delivery in most facilities. They improved human
resources (HR) numbers and capacity with the exception of a few HR cadres. They led to overall
improvements in supply chain management and access to essential commodities (though
increased programme activity did in some instances overload supply, leading to stock-outs).
Leaders and governance structures were supported, but coordination remained a challenge. While
the increase in funding was significant, overall financing remained erratic. GHIs also raised a need
for improved data systems and capacity development/support; without such improvements,
reporting requirements led to staff overload. There also was conflict between infrastructure
development and programme expansion associated with many GHI-supported interventions.
Overall, there were positive synergies between GHI implementation and health systems in
Uganda. However, there were some negative outcomes as well.
Background
Uganda is a landlocked country in East Africa bordered by Sudan in the north, Kenya in the east,
the United Republic of Tanzania and Rwanda in the south, and the Democratic Republic of Congo
in the west. In 2007 Uganda had an estimated population of 30.9 million people with an average
annual population growth of 3.2% per year [1]. Uganda ranked 156 th out of 179 countries on the
Human Development Index in 2006 [2].
The HIV/AIDS prevalence rate in Uganda peaked in the 1990s at 25-30% in major urban areas, but
stabilized at 5.4% by 2007 [3]. About 940 000 people were living with HIV/AIDS by the end of 2007
[4].
An estimated 51% of households do not have access to medical services in Uganda, with health
care delivery especially poor in the northern areas affected by civil war [5]. While 60.89% of the
country’s physicians worked in cities in 2004, only 12.82% of the population lived in an urban area
[6]. General government expenditure on health as a percentage of total government expenditure
increased from 7.3% in 2000 to 10% in 2006 [7]. In 2006, 28.5% of the total health expenditure on
health came from external sources [7].
181

Official Development Assistance (ODA) to Uganda in 2006 was US$ 1.55 billion [8]. The country’s
total debt is 13.3% of gross domestic product (GDP) [1]. In 2006, The International Monetary Fund
(IMF) and the World Bank’s International Development Association provided the country with debt
relief of approximately US$4 billion, a reduction of about 90% [9].
Table 1 Basic Socioeconomic, Demographic and Health Indicators*
* Full data sources for all indicators are provided in Annex 1
Indicator Value Year Source
Population (thousands) 30,916 2007 World Bank
Geographic Size (sq. km) 197,100 2007 World Bank
GDP per capita, PPP
(constant 2005 international $)
1000 2007 World Bank
Gini index 43 2005 World Bank
Government expenditure on health (%
general government expenditure)
8.9
2007 WHO NHA
Per capita government expenditure on
health (current US$)
7 2007 WHO NHA
Physician density (per 10,000)

PEPFAR
Year Amount Disbursed (in US$)
2004 90,774,095
2005 148,435,327
2006 169,875,461
2007 236,626,415
2008 283,635,476
TOTAL: 929,346,774
GAVI
Disease Priority Amount Approved (in US$)
Pentavalent vaccine 175,310,000
Vaccine introduction grant 100,000
Injection Safety 1,385,000
Immunization services support 9,230,520
Health system strengthening 19,242,000
TOTAL: 205,267,157
World Bank MAP
Title FY Approved/Closing Date Commitment (in US$)
HIV/AIDS Control Project 2001/2006 47,500,000
Objectives and Methodology
This evaluation study was designed to generate information that can inform health policy and
programming from the knowledge, experience and perceptions of personnel at health facilities
implementing GHI-funded programmes. Specific aims were: (1) To elicit perceptions about GHIs in
Uganda; (2) to find out how GHIs have strengthened health systems in Uganda; (3) to establish
how GHIs have interacted with health systems at different levels in Uganda; (4) to generate
questions for further analysis and study within the health context in Uganda; (5) to identify
knowledge and data gaps in the Ugandan context.
The study was conducted in 23 districts selected from the four geo-political regions of Uganda.
Research was conducted at 35 sites: Four regional (Provincial) referral hospitals, 11 district
hospitals, three private sector hospitals, 15 Health Centre Grade IV facilities and two nongovernment
health facilities. Eighty-two percent of the sites surveyed were operated by
government; 9% by missionary organizations; and 9% by other nongovernment agencies. All
participating agencies had implemented programmes funded by GHIs, especially Global Fund and
PEPFAR. The study population comprised district health officers of the identified districts; medical
superintendents heading identified hospitals; medical officers in charge of identified health
centres; institution heads of identified private sector agencies; and opinion leaders identified and
contacted during the study.
183

The study employed both purposive sampling and snowball sampling. The study team selected
participants based on their roles and experiences in implementing projects/programmes
supported by GHIs in Uganda. The study aimed to interview as diverse a range of individuals as
possible, asking the interviewees identified through purposive sampling to nominate other people
they knew who would willingly participate in the study.
The evaluation was cross-sectional, employing both quantitative and qualitative methods
(administered structured questionnaires and conducted key informant interviews). It involved
review of documents relevant to GHI implementing partners in Uganda. Information from these
documents was used to complement findings from the other data collection methods. The
evaluation involved in-depth interviews with identified key informants. Interviews with the key
informants were conducted according to the key informant guide so that the interviews yielded
data responsive to the evaluation objectives. Key informants were also guided to provide
perspectives and perceptions on implementation of GHIs. Interviews were recorded following
consent of the informant, and field notes were taken.
Authority to conduct the study at identified health facilities in Uganda was sought from the
Uganda National Council for Science and Technology (UNCST). Lower level permission and
consent was sought from the management of the respective health facilities participating in the
study. Researchers also sought the consent of the interviewees who completed forms signifying
their consent to participate in the study and provide required data. Participants were informed
about their freedom to refrain from the study. Participants were given an opportunity to ask
questions and/or seek clarification. Participants were also informed of guaranteed confidentiality;
this included altering personal details captured on audio tapes during transcription to ensure that
Key Informants were not directly linked to their comments.
Table 3: Increase in ANC Attendance in relation to availability of HIV counseling services at
UPHOLD Sites in Uganda
Baseline
2004
2005 2006
Number % Change Number % Change
Targeted (Counseling) 1198 14958 1148.5% 30268 102.4%
Non-Targeted (ANC) 1745 22001 1160.8% 42509 93.2%
Table 4: Availability of TB Management Services at Surveyed Health Facilities
Available TB Services % Facilities before GHI % Facilities Currently
TB Diagnosis 85.7% 97.1%
TB Treatment 88.6% 94.3%
DOTS 65.7% 85.7%
MDRTB Treatment 14.3% 40.0%
184

Results
Leadership and Governance
GHIs have stimulated development of various health policies geared towards improving health
status. GHIs also improved the participation of communities in health care programmes. Funds
provided by GHIs have improved workforce motivation and strengthened programmes for staff
retention and coordination at national and district levels. However, there was duplication of work
due to poor or inconsistent coordination and monitoring of programmes.
Health Workforce
Generally there was an improvement in human resources for health both in terms of numbers of
staff and skills following GHI implementation. Key informants at the health facilities surveyed
associated this change with GHIs. On the positive side, many cadres of health workers were trained
by GHI-funded programmes (lab staff, HIV/AIDS counsellors, provision of antiretroviral therapy
[ART], etc). However, trainings were criticized for being disruptive to service delivery (taking health
workers out of station for long periods, thus creating service delivery gaps at facilities). GHIs were
also criticized for attracting health workers to GHI-funded programmes and denying the public
health sector much needed human resources for health. All the key informants reported that they
had lost doctors to GHI-funded projects.
Table 5: Numbers and Changes in Personnel Cadres at Surveyed Health Facilities
No. Before GHI No. Currently % Change
Surgeons / Obstetricians 19 28 47.3
Medical Doctors (Specialists) 16 22 37.5
Medical Doctors (Non-Specialists) 34 24 (29.4)
Non-Doctor Clinicians 30 56 86.7
Nurses (All Categories) 1037 1466 41.4
Laboratory Personnel (All categories) 74 115 55.4
Pharmacy Personnel (All categories) 33 67 103.0
Counsellors (all categories) 175 428 144.6
Community Health Workers 290 441 52.1
Lay Service Providers 72 68 (5.5)
Other service providers 193 498 158.0
There was general increase in the number of health workers at all the facilities that were surveyed,
with a 145% increase in the number of counsellors and a103% increase in the number of pharmacy
personnel of all categories. Other service providers increased by 158%. There was however a
reduction in the number of non-specialist medical doctors by nearly 30% and the number of lay
service providers reduced by 6%. For the rest of the health workforce the increases were over 35%.
185

Figure 1: Number of Clinicians at Surveyed Health Facilities before and after GHI
100%
80%
60%
40%
20%
0%
Surgeons/Obs
28
19
Med.Doctors (Spec)
22
16
Med.Doctors (Non‐Spec)
24
34
Non‐Doc Clinicians
56
30
Nurses
1466
1037
LabTechs
115
74
PharmTechs
67
33
Post‐GHI Increase in Staffing
Level per Category
Pre‐GHI Staffing Level per
Category
Figure 2: Change in Number of Other Service Providers after GHI Implementation
The GHIs created an opportunity for capacity building for various cadres of health workers.
However the capacity building approach led to a number of health workers not being available at
their stations, thus impacting negatively on health care delivery. There was attrition of health
workers from poor-paying public health sector jobs to better paying GHI project jobs within and
without the country. There was also a tendency for staff working in GHI projects to change jobs
within the various projects due to the non-uniform compensation across projects.
186

Service Delivery
All the health facilities surveyed reported an increase in both the range of health care services and
the volume of service delivery associated with interaction between GHI implementation and
health systems at the facilities. This relationship was also noted between non-targeted services
and the targeted services.
For the majority of the population, there was significantly increased access to a wide range of high
quality HIV, tuberculosis (TB) and malaria services.. In addition, the GHIs brought about a
tremendous reduction in the cost of health care for the targeted conditions and, to a lesser extent,
the non-target conditions. GHIs had limited impact on infrastructural development and the focus
was narrow.
Although the range and volume of service delivery were increased, there remained several critical
gaps. The UPHOLD Final Evaluation found that Artemisinin-based combination therapy (ACT) was
prescribed in only 41% of cases of malaria, despite the fact that more than 90% of health workers
had been oriented and given policy guidelines on management of uncomplicated malaria. More
than 80% of the health workers in the UPHOLD supported districts had adequate skills in malaria
case management. However, more than 70% of these health workers were not communicating
effectively with patients regarding malaria and its treatment. Other problems included stock outs
of ACT in 27% of the supervised facilities [10].
Researchers found improvements in the areas of TB diagnosis and treatment with significant
increase in implementation of the Directly Observed Treatment, Short-course (DOTS) programme.
Access to Multidrug-Resistant Tuberculosis (MDR-TB) treatment more than doubled. Key
informants also reported significant improvements in supply chain management for TB drugs and
laboratory reagents. Laboratory technicians were trained and there was funded supervision of
health workers providing TB services. The national programme was also able to significantly scaleup
the provision of community-based DOTS services (CB-DOTS). There was a near double increase
in case detection rates for TB in some districts. However respondents felt that less money was
allocated to TB compared with HIV and malaria.
The health facilities surveyed indicated some positive synergies between supported services and
radiology services. However, some variables indicated a reduction in access to radiology services
despite implementing GHI at surveyed facilities. There was approximately 10% increased
availability of plain X-rays on site but no change in access within two hours. There was however a
reduction in fees charged. There was a 15% increase in availability of ultrasound services on site,
there was reduced access to ultrasound services within 2 hours and an increase in the fee charged
for ultrasound services.
Table 6: Referral Practices at Surveyed Health Facilities shows the referral mechanisms at the
surveyed health facilities before the GHIs and the current situation.
% Facilities Before GHI % Facilities Currently
Higher Level Facility within 2 Hrs 62.9 77.1
Phone, etc Link to Higher Level Facility 34.3 48.6
Ambulance Available 60.0 88.6
Fee Charged to Facilitate Referral 11.4 40.0
Referral Form used (Facility to Facility) 80.0 91.4
Referral Accompanied to Facility 48.6 68.6
Referral Form used by Community Workers 31.4 62.9
187

There were very significant increases in referral practices for all the variables that were assessed,
with a 50% increase in the use of referral forms by community health workers.
Medical Products, Vaccines and Technologies
All the surveyed facilities and key informants associated improvement in medical products and
technologies to interaction between GHI implementation and health systems. Positive synergies
included introduction of new and more effective medicines; availability and access to technologies
like insecticide-treated bed nets; safe water vessels and Waterguard; home-based HIV counselling
and testing; home-based antiretroviral (ARV) delivery; rapid HIV test kits; and access to emergency
prophylaxis. Besides these benefits, the facilities and key informants highlighted negative
interactions in medical products and technologies such as frequent stock-outs and expiry of
medicines and other commodities.
Health Infrastructure
Most of the facilities and key informants surveyed associated GHI implementation with a mismatch
between investment in health infrastructure and the level of programme scale-up. GHIs had very
little impact on infrastructure development. Some infrastructure development was initially done
by PEPFAR but this has since been significantly scaled down. All the key informants felt that GHIs
had not impacted positively on health infrastructure. This is because the two programmes (the
Global Fund and PEPFAR) did not have components for infrastructure development compared to
the World Bank MAP programme, through which health centres, wards, clinical rooms and staff
houses were created.
Monitoring and Evaluation
All the health facilities and key informants surveyed associated GHI implementation with increased
need and emphasis for data and information management systems. All the health facilities and key
informants expressed challenges to service delivery associated with the demands and pressures of
data and information systems.
The AIDS/HIV integrated model district programme (AIM) Final Evaluation in Uganda found that
the project interventions had created extra workload which was not matched by staff increases,
the records assistants were generally overloaded, the data collection tool was very demanding
(requiring significant information), the programme created more reporting forms (resulting in
multiple/parallel reporting formats) and there was no effort to address this problem [11].
Financing
All the facilities and key informants surveyed associated increased health financing with GHI
implementation. Overall, the facilities and key informants reported that health financing had
remained erratic and uncertain.
Global Fund and PEPFAR have made supplementary funds available for HIV/AIDS, TB and malaria,
areas where government lacked adequate funding. However, this was observed to impact
negatively on prioritization and allocation of resources for health at both national and district
levels. Both levels of governance assume that GHIs funding is adequate to address national and
188

district-level health care needs. These funds were also observed to be erratic with little flexibility.
The application requirements were very rigid and there was a general feeling within the public
health sector that GHIs should incorporate their funds into government funding mechanisms.
Discussion
Global Health Initiatives (GHIs) have significantly increased access to a wide range of
HIV/TB/Malaria services of high quality and at no direct cost to the users. There is increased access
to health care services in Uganda as a result of interactions between GHIs and health systems.
There are significant improvements in the areas of TB diagnosis and treatment. The availability of
GHI funding enabled scale up and implementation of the CB-DOTS programme. In the last five
years, during which time GHI funds have been available, access to MDRTB treatment has more
than doubled.
Improvements were also observed in supply chain management for TB drugs and laboratory
reagents. Funding from GHIs also helped to train laboratory technicians and to improve
supervision of health workers providing TB services. Case detection rates for TB in some districts in
Uganda almost doubled. However, most of the funding went to HIV/AIDS, with comparably less
money allocated to TB and malaria.
There was a 10% increased availability of plain X-rays on site; however, ‘access within two hours’
did not change. There was a reduction in fees. Availability of ultrasound services on site increased
by 15%, but ‘access to ultrasound services within two hours’ was reduced and the fee charged for
ultrasound services increased.
There was marked improvement in referral practices for all the variables that were assessed, with a
50% increase in the use of referral forms by community health workers. There was also a great deal
of focus on children, as evidenced by the number of children accessing HIV/TB/malaria services.
There were significant reductions reported in the cost of health care for the targeted conditions,
and to a lesser extent, the non-target conditions.
However, GHIs had limited impact on infrastructural development and the focus of the funding
was limited to the three target diseases. The programmes had both positive and negative impacts
on the health care workforce. Positively, many cadres of health workers were trained (lab staff,
HIV/AIDS counsellors, provision of ART, etc). However, these trainings were criticized for being
“class room type” – taking health workers out of their duty stations for long durations, thus
creating service delivery gaps at public facilities. GHIs were also criticized for attracting health
workers to GHI funded programmes, denying the public health sector the much-needed human
resources for health. Most public health facilities had lost doctors to GHI funded projects within
and outside the country.
There was a general increase in the number of health workers at all the facilities that were
surveyed, with 145% increase in the number of counsellors and 103% increase in the number of
pharmacy personnel of all categories. Other service providers increased by 158%. There was,
however, a reduction in the number of non-specialist medical doctors by nearly 30% and lay
service providers were reduced by 6%. For the rest of the health workforce the increases were over
35%.
189

GHIs created an opportunity to conduct capacity building for various cadres of health workers.
There was a tendency for staff working in GHI projects to change jobs within the various projects
due to non-uniform compensation across projects.
Reporting mechanisms are multiple and complicated and draw frontline staff away from service
provision to servicing reporting requirements of the different GHIs because of their essentially
vertical nature. The data collected may not have been in line with the host country data
requirements. There was limited capacity building in the areas of logistics and stores management
systems although PEPFAR did more in this area than the Global Fund. GHIs greatly improved
logistics management, especially for ART and distribution of anti-malaria commodities (Coartem
and insecticide-treated bed nets). However, stock outs of commodities increased, sometimes
lasting over a month.
190

References
* Conducted by The AIDS Support Organization (TASO) Uganda Limited, in partnership with the
World Health Organization (WHO) and Harvard Medical School
[1] The World Bank Group. Uganda at a glance (http://devdata.worldbank.org/AAG/zmb_aag.pdf;
accessed 19, February 2009).
[2] United Nations Development Programme. Human Development Reports: Uganda
(http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_UGA.html; accessed 19
February 2009).
[3] United Nations General Assembly Special Session. Uganda Progress Report 2007 2008.
[4] UNAIDS. 2008 Report on the global AIDS epidemic 2008.
[5] United Nations Office for the Coordination of Humanitarian Affairs. Uganda Humanitarian
Country Profile. (http://www.irinnews.org/country.aspx?CountryCode=UG&RegionCode=EAF;
accessed 18 February 2009).
[6] World Health Organization. Global Atlas of the Health Workforce 2008.
[7] World Health Organization. WHO Statistical Information System (WHOSIS).
(http://www.who.int/whosis/en/; accessed 19 February 2009).
[8] The World Bank Group. World Development Indicators (WDI) Online. (http://ddpext.worldbank.org.ezp-prod1.hul.harvard.edu/ext/DDPQQ/member.do?method=getMembers;
proprietary database, accessed 19 February 2009).
[9] United Nations Office for the Coordination of Humanitarian Affairs. Uganda: Debt relief frees
up funds for poverty reduction; 11 May 2006.
[10] Final Evaluation of Uganda Program for Human and Holistic Development (UPHOLD), 2007,
Kampala
[11] AIM Project Evaluation – Final Report, 2007, Kampala.
(http://pdf.usaid.gov/pdf_docs/PDACL388.pdf).
191

Ukraine: Effects of the Global Fund on
the health system
Abstract
Tetyana Semigina 27
Ukraine has one of the most rapidly growing HIV/AIDS epidemics in Europe, with estimated
numbers of people living with HIV/AIDS (PLWHA) reaching 400,000 in 2008. Since 2003, the Global
Fund has committed to providing US$ 243 million towards the control of HIV/AIDS in Ukraine. This
study examines the impact of Global Fund financing on HIV/AIDS services and the health system in
Ukraine. Research was conducted in Kyiv, Odessa and L’viv.
The research finds that Global Fund resources have supported a dramatic scale-up of services for
PLWHA and at-risk populations. There has been a substantial increase in the number of PLWHA
receiving antiretroviral therapy (ART). New laboratory, diagnosis, prevention and support services
have been developed. The Global Fund grant has strengthened service delivery by training health
workers and supporting the development of HIV/AIDS surveillance systems.
Whilst the Global Fund has had a positive impact on governance and leadership, for instance by
promoting transparency among government health service providers and improved management
practices, cooperation between government and nongovernmental service providers is limited,
and coordination structures are weak. Many services financed by the Global Fund are provided by
nongovernmental organizations (NGOs) heavily dependent on continued Global Fund support,
highlighting issues of sustainability.
27 School of Public Health and School of Social Work, Kyiv-Mohyla Academy, Kyiv, Ukraine. The research for this study
summary was funded by The Open Society Institute, New York. The study summary was produced with the support of
the GHIN Network (www.ghinet.org).
192

Background
General Context
Table 1: Basic Socioeconomic, Demographic and Health Indicators*
* Full data sources for all indicators are provided in Annex 1
Indicator Value Year Source
Population (thousands) 46,383 2007 World Bank
Geographic Size (sq. km) 579,380 2007 World Bank
GDP per capita, PPP (constant 2005 international $) 6,528.44 2007 World Bank
Gini index 28.239 2005 World Bank
Government expenditure on health (% general
government expenditure)
8.8 2007 WHO NHA
Per capita government expenditure on health
(current US$)
116 2007 WHO NHA
Physician density (per 10,000) 31 2006 WHO SIS
Nursing and midwifery density (per 10,000) 83 2006 WHO SIS
Maternal mortality ratio (per 100,000 live births) 18 2005 WHO SIS
DTP3 coverage (%) 98 2007 WHO SIS
Estimated adult HIV (15-49) prevalence (%) 1.6 (1.2-2.0) 2007 UNAIDS
Estimated antiretroviral therapy coverage (%) 8 2007 WHO/UNAIDS/UNICEF
Tuberculosis prevalence (per 100,000) 102 2007 WHO GTD
Estimated malaria deaths N/A N/A N/A
Table 2 Global Health Initiative Investment*
(other than Global Fund and World Bank)
PEPFAR * (in US$)
Year Amount Disbursed
2004 5,504,000
2005 7,074,000
2006 5,027,000
2007 6,744,000
2008 5,850,000
TOTAL: 30,199,000
*Not a PEPFAR focus country; above sums represent total allocations to PEPFAR
country programmes from bilateral U.S. sources including USAID, Department of
Health and Human Services, Department of Labor, and Department of Defense.
193

GAVI (in US$)
Disease Priority Amount Approved
Hepatitis B vaccine 3,212,000
Vaccine introduction grant 100,000
Injection Safety 792,000
TOTAL: 4,326,280
NB: Investments from Global Fund and World Bank are reflected in Figure 1, below.
Epidemiological context
Ukraine has the most rapidly growing HIV/AIDS epidemic in Europe. The number of registered
people living with HIV/AIDS (PLWHA) was 131 000 in 2008; the number has increased from
approximately 46 000 registered cases in 2002. Many PLWHA remain unregistered, and indeed,
estimates suggest that there may currently be about 400 000 PLWHA in Ukraine. According to
Ministry of Health (MOH) data, as of 1 January 2008, the highest levels of HIV prevalence were in
Odessa (414 cases per 100 000 population), Dnipropetrovsk (409) and Donetsk (390) regions.
Intravenous drug use accounted for most HIV transmission (approximately 40%), sexual (mainly
heterosexual) for 38%, and vertical mother to child transmission for 19% [1].
Funding for HIV/AIDS programmes in Ukraine
Several donors and GHIs provide funding for HIV/AIDS prevention, treatment and care programmes in
Ukraine. The largest GHI is the Global Fund, which is slated to provide US$ 243 million between 2003
and 2011 for HIV/AIDS projects (see Figure 1). The World Bank programme to fight HIV/AIDS and
tuberculosis (US$ 77 million) was launched in Ukraine in 2003. The country has also received additional
funding and technical assistance from UN organizations (WHO, UNAIDS, UNICEF, UNDP, ILO), bilateral
programmes (USAID and the Swedish International Development Agency [SIDA]), and NGOs. In 2006,
the total amount of funding for HIV/AIDS programmes in Ukraine was US$ 55.4 million, of which US$
28.1 million was provided by the state budget and US$ 20.1 million from Global Fund grants [2]. There
are two principal recipients of the current Round 6 Global Fund grant: the International HIV/AIDS
Alliance and the Network of People Living with HIV/AIDS [3]. Figure 1 summarizes Global Fund and
World Bank HIV/AIDS programmes in Ukraine.
Figure 1: Global Fund and World Bank HIV/AIDS funding in Ukraine
GF
1 round
$ 92M
6 round
$151M
WB
$ 77M
2003 2004 2005 2006 2007 2008 2009 2010 2011
Error!
Reference
source
not
found.
MOH +UNDP
International HIV/AIDS Alliance
MOH +State Department of Corrections
194
International HIV/AIDS Alliance + Network of
People Living with HIV/AIDS

Methodology
Between July 2006 and December 2008, a team from the School of Public Health and the School of
Social Work at the National University, Kyiv-Mohyla Academy, supported by a team from the
London School of Hygiene and Tropical Medicine, conducted the Ukrainian part of an international
research project on the national and sub-national effects of GHIs for HIV/AIDS on country health
systems. The research was funded by the Open Society Institute. The study focused on scale-up of
HIV/AIDS services; coordination structures and coordinated service delivery, strengthening the
health workforce and service delivery/access. This case study summarizes findings from the final
study report, based on 2008 data collection.
The focus of the study is on Global Fund-financed programmes in Ukraine and their
implementation in three regions of the country: Kyiv, Odesa and L’viv. The World Bank’s Ukrainian
HIV/AIDS loan was not assessed in detail as part of this research, since in April 2006 the World Bank
suspended its support, arguing that the government had not distributed the funds and
implemented the programme adequately. Support resumed in November 2006 on the condition
that Ukraine would improve the management and implementation of the project. The programme
was scheduled to end in mid-2008 but was extended until the end of 2008. Only a limited
proportion of the overall funds have, however, been used [4].
Studying the influence of GHIs in Ukraine is particularly important given the slow progress that the
country has made in stabilizing and reducing the spread of HIV/AIDS [5]. This is despite the
legislative, political, organizational and administrative measures that have been taken in response
to the epidemic, including the recent formation of the National Coordination Council for HIV/AIDS
and increased activities among NGOs providing HIV/AIDS services. Previous research has not
provided a systematic analysis of the GHIs that operate in Ukraine and their influence on HIV/AIDS
services and the health system.
Qualitative and quantitative research methods were used to collect the data. Researchers analysed
documents including: annual reports of the International HIV/AIDS Alliance in Ukraine and the All-
Ukrainian Network of People Living with HIV/AIDS; national progress reports on the United Nations
General Assembly Special Session (UNGASS) Declaration of Commitment on HIV/AIDS; documents
from international organizations; presentations from stakeholders’ meetings organized by the
Principal Recipients of the Global Fund grant; official statistics on the epidemic in Ukraine; and
policy documents, including documentation on the national programme to combat HIV/AIDS, as
well as laws and decrees adopted by the Parliament, President and Cabinet of Ministries.
Researchers conducted semi-structured interviews with national and sub-national key informants;
semi-structured and structured interviews with clients of HIV-service organizations; and structured
interviews with providers of medical and social services. Instruments for the study were developed
jointly by the Ukrainian team and the Global HIV/AIDS Initiative Network (GHIN) partners. The
study was approved by the London School of Hygiene and Tropical Medicine Ethics Committee
prior to being undertaken.
In 2008, 49 key informants were interviewed using a semi-structured topic guide. Interviewees
included national stakeholders (n=21) and sub-national stakeholders from Kyiv (n=12); Odessa
(n=10); and L’viv (n=6). Stakeholders are defined as individuals making decisions about HIV/AIDS
programmes or implementing those programmes at national and regional levels. They include
representatives from government and international organizations, from regional government
departments, and from national and regional NGOs, including members of national and regional
HIV/AIDS Coordination Councils. A sample of 50 HIV/AIDS service providers, including both
195

government and nongovernmental providers, were interviewed using a structured survey tool.
Twenty-five clients participated in in-depth qualitative interviews and nine organizations were
surveyed using a structured facility survey tool.
Results
Governance
Some participants reported that Global Fund-supported programmes have engendered a change
in mentality among government medical institutions, including higher levels of transparency
among government service providers. In addition, the Global Fund is seen as having promoted
improved management practices among both government and nongovernmental agencies
providing HIV/AIDS services.
Legislative reforms
The study found that the Global Fund programme in Ukraine has promoted gradual changes in
regulation and legislation. An important component of the first stage of implementation of the
Global Fund programme was the development and publication of national clinical protocols by
the International HIV/AIDS Alliance and the MOH. These include: protocols for antiretroviral
therapy (ART) for adults, teenagers and children, treatment of opportunistic infections in HIV/AIDS
patients and methodological recommendations for laboratory monitoring of HIV infection and
ART.
Coordination structures
In 2002, a national HIV/AIDS coordination council was formed, consisting of government
representatives, international development actors and civil society members. Since its formation,
the council has undergone several changes in structure, focus and membership, and a number of
parallel HIV/AIDS structures have been established and abolished. The current structure is called
the Coordination Council on HIV/AIDS, TB and Drug Addiction. Many key stakeholders consider the
creation of the Council to be a positive step, because it serves as a good example of cooperation
between government and NGOs and actively involves PLWHA. However, some have been critical
about the functioning of the Council, which meets infrequently and has limited powers beyond
financial issues. The study revealed that coordination structures at the sub-national level were
particularly weak and lacked decision-making authority in some regions.
196

Health Workforce
Increase in the health workforce
From 2004-2007, the number of personnel increased in most governmental and nongovernmental HIVservice
organizations.
Table 3: Changes in the number of staff working for organizations providing HIV/AIDS services
Organization name 2004 2007 Total increase
Kyiv city hospital AIDS centre 60 150 90
Odessa AIDS centre 12 37 25
Kyiv office of the All-Ukrainian Network of PLWHA 12 50 38
Odessa NGO “Faith. Hope. Love” 27 90 63
Interviews with service providers revealed that many had experienced an increase in their
workload. The key factors contributing to this pattern were increases in client numbers and in
administrative work. However, data from most facility surveys show that the number of client visits
per worker actually did not change, as the number of workers grew in proportion with the number
of clients. The 2008 survey also found that of 50 respondents, 38 considered they were motivated
(felt happy delivering HIV services and were sufficiently incentivised), 36 were satisfied with their
work and 33 liked working with HIV-positive clients.
Training for HIV/AIDS service providers
International HIV/AIDS Alliance data indicate that resources have been committed to developing
the skills of managers and workers who provide HIV/AIDS services and prevention programmes.
Training programmes in 2004 covered the following topics: data collection, monitoring and
evaluation (M&E), organization of effective prevention programmes for commercial sex workers,
before- and after-test counselling, and advocacy on the rights of vulnerable groups.
By 2007, the reach and the topics of training had expanded considerably. There were nine regional
HIV/AIDS information and resource centres, and these organized 255 training and seminar sessions
for almost 3800 people. The International HIV/AIDS Alliance also introduced training aimed at
strengthening regional NGOs, including sessions on financial management, grant management
and project management.
Of the 50 service providers that were interviewed, 37 had taken part in HIV/AIDS training over the
past 12 months. The most common sessions were voluntary counselling and testing (VCT), HIV
prevention and social support. In most of these cases, financing was provided by the Global Fund.
However, it should be noted that training was mainly short-term, lasting between one and three
days, more rarely between four and seven days, and only in some cases continuing between eight
days and a month.
Service delivery
Strengthening HIV/AIDS service delivery
The study revealed that the Global Fund has strengthened service delivery in a number of ways
including:
197

• The introduction of new services, including ART and opiate substitution therapy;
• Training of medical and NGO personnel providing HIV/AIDS services;
• Refurbishment of premises of HIV/AIDS services;
• Funding the purchase of equipment;
• Fostering some improvements in cooperation between government and
nongovernmental service providers;
• Systems of epidemiological surveillance have been strengthened.
Scale-up of HIV/AIDS services
Funding has been used to dramatically scale-up medical, prevention and care/support services for
PLWHA and the most at-risk population. The Global Fund programme has led to a substantial
increase in the number of PLWHA receiving ART. In 2003 only 53 people received ART, whilst in
2008 this number increased to 900, of whom 6000 received treatment financed by the Global
Fund. In addition, new laboratory and diagnosis services have been developed, and prevention,
care and support programmes have been created. These include harm reduction and drug
dependence reduction programmes, such as needle/syringe exchange and substitution therapy.
Table 4 illustrates scale-up in client numbers in the three regions of Ukraine that were selected to
be a part of the study.
Table 4: Total client numbers receiving key HIV/AIDS services in Kyiv, Odessa and L’viv
Kyiv Odessa L’viv
Services 2004/2005 2006 2004/2005 2006 2004/2005 2006
Preventative services 7843 23139 11214 28536 782 4305
Substitution therapy - 240 - 99 - -
Care and support 1111 2862 989 4658 29 221
ARV treatment 232 301 373 374 - -
Source: International HIV/AIDS Alliance in Ukraine
Communities/Civil Society
Many of the services financed by the Global Fund are provided by NGOs and community
organizations. These are sometimes based at, or in cooperation with, government organizations.
Since the Global Fund programme was implemented, NGOs have played an increasingly important
role in delivering prevention, care and support services.
This study suggests the peer-to-peer approach to providing services for vulnerable groups such as
drug users and sex workers is effective. This approach was valued by clients and helped to break
down the barriers between providers and clients, thereby improving access to services.
Respondents pointed out that the Global Fund has had an impact on the development of civil
society in a number of ways:
1. The Principal Recipients of the Global Fund grant are NGOs. In 2007 and 2008 the majority
of HIV-service organizations in Ukraine received financing through the two Global Fund
198

Principal Recipients: the International HIV/AIDS Alliance in Ukraine and the All-Ukrainian
Network of People Living with HIV. Both organizations have become important actors in
Ukrainian policy to combat HIV/AIDS.
2. A substantial part of funding has been disseminated among NGOs. The research suggests
that in Kyiv, Odessa and L’viv, between 60% and 85% of funding for NGOs is from the
Global Fund.
3. NGO staff have received training, including in topics related to service provision and
organizational management.
4. NGOs have become involved in decision-making processes relating to public health policy
and HIV/AIDS. This is partly the result of the Global Fund’s requirement that the National
Coordination Committee should include representatives from civil society organizations.
Discussion
The implementation of the Global Fund programme in Ukraine has led to an increase in the
number of government and nongovernmental stakeholders from different sectors involved in
HIV/AIDS-related activities, and to scale-up of HIV/AIDS services, including prevention, testing,
treatment, care and support.
The study highlighted the challenges to extending access to key HIV services among vulnerable
populations. Potential obstacles include: stigmatisation of HIV-positive people, the criminalization
of drug use, and limited provision of information about HIV/AIDS services.
The study suggests that the quality of services provided by NGOs varies and depends on the
individuals who implement projects. This is because there are no official quality standards for the
provision of HIV/AIDS services by NGOs. Moreover, there is a lack of commitment by the
government to fund NGOs, which are currently highly dependent upon international donors and
initiatives, particularly the Global Fund. This has implications for NGOs’ long term sustainability.
199

References
[1] Data are from the Ministry of Health of Ukraine, Kyiv, 2008.
[2] Ukraine: National report on Monitoring Progress toward the UNGASS Declaration on
Commitment on HIV/AIDS. Kyiv, Ukraine, Ministry of Health of Ukraine, 2008 (http://www.aidsukraine.com/de/upload/ukraine_2008_country_progress_report_en.pdf;
accessed 20 March 2009).
[3] Support for HIV and AIDS Prevention, Treatment and Care for the Most Vulnerable Populations
in Ukraine: Ukraine Proposal for GFATM Sixth Call for Proposals. Kyiv, Ukraine, 2006.
[4] Ukraine TB and HIV/AIDS Control project: Current status. In: 13th meeting of the parties
involved, 7 December 2007 (http://www.network.org.ua; accessed 20 March 2009).
[5] See the Context Mapping report and Interim Report from this project for an overview of the
epidemiological context in Ukraine and the policy and programmatic response.
200

United Republic of Tanzania: Strengthening the
health system to address
the burden of HIV/AIDS
and TB
Abstract
Thyra de Jongh, Rifat Atun 28
The United Republic of Tanzania is facing a massive burden of HIV/AIDS and TB. Addressing these
issues is complicated by a weak health care infrastructure and a severe human resources crisis. In
response, the country has enjoyed considerable donor support, including several grants from the
Global Fund. Although these grants are primarily intended to fund disease-specific activities, it is
recognized that strengthening of the overall health system is a prerequisite to reap the full
benefits of the investment. This case study examined the effects of focused investment on
Tanzania’s health system as a whole. A mixed-methods design was used, based on the Systemic
Rapid Assessment toolkit (SYSRA). We conducted interviews with key stakeholders at all levels. We
found that, in addition to significant contributions in the areas of TB and HIV, the Global Fund has
strengthened the health system through training of health care workers and investment in general
health facilities. It has, however, also led to some duplication of activities, particularly in the area of
monitoring and reporting, diversion of resources, and potential distortion of resource allocation.
Although promising steps have been taken to further align national and donor activities, scope for
improvement remains.
Background
The United Republic of Tanzania was formed in 1964 by the union of the mainland territory then
known as Tanganyika with the island archipelago of Zanzibar. The largest country in East Africa,
Tanzania shares a border with Kenya, Uganda, Burundi, Zambia, Malawi, Mozambique, Rwanda,
and the Democratic Republic of Congo. As of 2007, Tanzania’s population was 40.4 million, with an
estimated 2.4% annual growth rate [1]. In 2006, Tanzania ranked 152 nd out of 179 countries on the
Human Development Index [2]. Though Tanzania is one of the continent’s most politically stable
countries, it remains one of the poorest countries in the world despite a real annual GDP growth of
about 6.3% since 2000 [1]. Significant macroeconomic and structural reforms introduced since the
mid-1990s have not succeeded in pulling the country out of poverty. Tanzania’s economy is largely
dependent on agriculture and tourism.
Since the first three cases were detected in 1983 in Kagera Region, HIV/AIDS has taken its place as
one of the leading causes of death in Tanzania [3]. In 2007, 1.4 million people were living with
HIV/AIDS [4].
28 Imperial College London/ Global Fund to Fight AIDS, TB and Malaria
201

The health system struggles with a severe shortage of health care workers, particularly in the rural
parts of the country, and suffers from weak infrastructure. In mainland Tanzania, the public health
sector (56%) is complemented by a private sector comprising private for-profit groups (14%), as
well as non-governmental organizations (NGOs) and faith-based organizations (FBOs) (30%) [5]. In
Zanzibar, most health facilities are publicly run. Traditional medicine remains an important part of
Tanzanian rural life [5].
In recent years government expenditure on health has increased, reaching 13.3% by 2006, but this
figure remains shy of the Abuja target of 15% [6]. As of 2006, external resources made up 45.4% of
the total expenditure on health [6]. In 2002, 44.5% of physicians worked in urban areas, although
only an estimated 23% of the population lived in cities [7].
In the meantime, the overall health budget has grown significantly, principally as a result of large
amounts of official development assistance (ODA). In 2006, ODA was US$ 1.8 billion. Tanzania’s
total external debt in the same period was US$ 4.2 billion, 29.9% of GDP [1,8].
Table 1 Basic Socioeconomic, Demographic and Health Indicators (*)
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 40,432 2007 World Bank
Geographic Size (sq. km) 885,800 2007 World Bank
GDP per capita, PPP (constant 2005
international $)
1,140.96 2007 World Bank
Gini index 34.62 2000 World Bank
Government expenditure on health (%
of general government expenditure)
13.7 2007 WHO NHA
Per capita government expenditure on
health (current US$)
11 2007 WHO NHA
Physician density (per 10,000)

Table 2 Global Health Initiative Investments (*)
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 1, Malaria 78,079,834 48,139,110
Round 1, HIV/AIDS 5,400,000 4,647,000
Round 3, HIV/TB 83,466,904 54,798,490
Round 4, HIV/AIDS 283,092,248 181,090,483
Round 4, Malaria 76,086,764 75,086,764
Round 6, TB 16,498,948 15,173,156
Round 7, Malaria 20,707,304 5,161,417
Round 8, HIV/AIDS 121,144,902 0
Round 8, Malaria 111,289,909 0
TOTAL: 795,766,813 384,096,420
PEPFAR
Year Amount Allocated (in US$)
2004 70,745,574
2005 108,778,095
2006 129,967,925
2007 205,482,327
2008 313,415,559
TOTAL: 828,389,480
GAVI
Disease Priority Amount Approved (in US$)
Tetravalent vaccine 26,378,000
Vaccine introduction grant 100,000
Injection Safety 1,494,600
Immunization services support 7,990,880
TOTAL: 39,891,230
World Bank MAP
Project Title FY Approved/Closing
Date
Commitment (in
US$)
Multisectoral AIDS Project 2004/2009 70,000,000
203

Objectives and Methodology
This study aimed to investigate the impact of the Global Fund on the provision of HIV/AIDS and TB
services in Tanzania, as well as on the wider health system. It sought to delineate how the national
disease control programs are organized with respect to the health system and to understand how
the Global Fund affects and, in turn, is affected by these organizational structures. Both intended
and unintended effects on the health system were considered.
The data collection for this case study included semi-structured interviews, field observations, and
examination of secondary data sources. Key informant interviews were conducted in a semistructured
fashion using a set of interview questions based on SYSRA and adapted to the analysis
of integration of health interventions [9,10]. Our analysis focused on the HIV/AIDS and TB
programs at the central, regional and district levels.
The sampling strategy for key informants targeted a diverse range of implementers, policymakers
and health leaders. The study used two sampling methods: purposive sampling and snowball
sampling. We conducted a total of 30 individual and group interviews. Our sample included 19
representatives from ministries or coordination and oversight bodies at the national level, three
from implementation partners in the NGO/FBO sector, two from the private health care sector, and
three from development partners. A total of 16 programme coordinators and managers from five
distinct districts were interviewed in a group format.
If informants consented, interviews were digitally recorded. The recordings were then used to
produce a set of detailed notes. Relevant sections of the interview were transcribed verbatim. A
framework approach was used to ‘rough code’ these notes and to chart data to the health system
functions described in the analytic framework developed by Atun and others [9]. Additional
information on the interaction between GHIs and the health system in Tanzania was gathered and
organized according to the WHO framework for health systems strengthening [11]. Secondary
data sources (e.g. national strategy papers, evaluation reports, Global Fund proposals) were
analysed in a similar thematic way. Primary and secondary data sources were compared for
internal validation of findings.
Results
Leadership and Governance
For effective and sustainable implementation of health activities, national governments need to be
able to exercise leadership: to set national level priorities and to oversee the development of a
strategic plan for the health sector and for the national responses to HIV/AIDS, TB and malaria.
There has been concern that global health initiatives (GHIs) like the Global Fund have undermined
this leadership capacity by imposing rigid conditionalities and guidelines. In Tanzania,
development partners provide a significant proportion of all health sector spending and, as such,
have considerable leverage over the national agenda. Interviews with stakeholders in Tanzania
suggest, however, that the activities supported by the Global Fund are considered to be largely in
line with the government’s own strategic objectives. Rather than influencing national priorities,
these funds are in effect helping to close the funding gap for existing priorities. The heavy
emphasis on HIV/AIDS, although possibly reinforced by the influx of large amounts of donor
money, is considered by stakeholders to be warranted by the severity of the disease burden.
204

As an outcome of its socialist history, the Government of Tanzania has long been reluctant to
involve private for-profit health facilities in the provision of public health services. Over the last
decade this attitude has changed somewhat, and a large number of private facilities now offer TB
and HIV/AIDS services. The further development of Public-Private Partnerships is stressed in the
third Health Sector Strategic Plan as essential for effective implementation of the national
response. The Global Fund has promoted greater involvement of the private sector, and numerous
private facilities, both for-profit and not-for-profit, are grant sub-recipients. This has helped to
bring stakeholders from both sectors closer together. Other civil society organizations (CSOs) have
also benefited from greater inclusion in policy development and strategic planning through their
membership in the Tanzania Country Coordinating Mechanism and through capacity building
supported by the Rapid Funding Envelope to which the Global Fund contributes.
Financing
Funding of HIV/AIDS and TB activities in Tanzania is heavily reliant on ODA. It is estimated that as
much as 95% of total expenditure on HIV/AIDS comes from development partners. It is widely
acknowledged by stakeholders that this has prompted the Government of Tanzania to shift some
of its resources to other sectors, reflected by a decrease in the government contribution to total
health sector expenditure in recent years.
Although Tanzania has a Sector Wide Approach with an associated Health Basket Fund (HBF),
grants by the Global Fund do not contribute to the HBF. Instead, the Ministry of Finance is a
Principal Recipient on all grants, and funds are channelled through the treasury. In theory these
funds are fully absorbed into the budgets of Local Government Authorities (LGAs), but delays in
disbursements have at times led LGAs to exclude these funds from their regular budgeting
procedures. Furthermore, funds from the Global Fund and other development partners are usually
allocated to individual regions or districts identified in the grant agreement. These contributions
are, however, not taken into account in the national formula used to allocate resources to the LGAs
and thus potentially create distributional inequalities.
Health Workforce
Tanzania has struggled with a sustained crisis in human resources for health (HRH). On average
only about one in three positions is currently filled with qualified workers, and the situation is even
starker in remote and rural parts of the country. The crisis has been fuelled by a toxic mix of a lack
of skills in HRH management, insufficient training capacity, high rates of attrition, and poor
compensation packages for health care workers. In order to effectively respond to the HIV/AIDS
and TB epidemics, resolving the HRH crisis is a priority for the Government of Tanzania, requiring a
combination of policy reforms and extra investment in HRH. In Tanzania, GHIs have been at the
forefront of recruitment and training of health workers, particularly in delivery of HIV/AIDS
services. Although these efforts have been essential in the implementation and scale-up of
antiretroviral therapy (ART) in the country, several of the facilities visited reported an unintended
consequence: the outflow of health workers from other–already understaffed–clinical areas as they
moved into positions at HIV/AIDS Care & Treatment Centers (CTCs), sometimes attracted by top-up
fees or more attractive remuneration packages. It is not clear exactly how large this effect has been
or to what extent this has been driven by GHIs. However, those interviewed in the course of this
study generally expressed the feeling that the shift of health workers from other fields to HIV/AIDS,
whilst unfortunate, has been necessary because of the severity of the HIV epidemic in the country.
205

Nonetheless, efforts to recruit and retain health workers in other clinical areas are required to
prevent further strain on the system. Although most of the trainings supported by GHIs have been
given in the context of a particular disease control programme, certain skills are perceived as
transferable. For instance, communication skills that were taught as part of training in voluntary
counselling and testing (VCT) have been used in delivery of other services as well.
Medical Products, Vaccines, and Technologies
The Global Fund and other development partners have had a major impact on the pharmaceutical
landscape in Tanzania. Their support has, for instance, enabled Tanzania to offer antiretrovirals
(ARVs) and drugs for the treatment of opportunistic infections (OIs) free of charge at the point of
care. Furthermore, to enable the scale-up of ART in Tanzania, the Global Fund has had to invest in
strengthening the systems for procurement and supply of drugs in the public system operated by
the Medical Stores Department (MSD). The Global Fund has invested in increasing the storage
capacity at zonal warehouses, and in improving supply chain management systems and skills. The
Tanzania Food and Drug Authority, charged with monitoring the quality, safety and efficacy of all
drugs in the country, has also received support from the Global Fund to improve its systems for
drug testing and quality assurance.
Information
Like many other developing countries, Tanzania has struggled with its systems for monitoring and
evaluation (M&E). Consequently, national control programmes like those for HIV/AIDS and TB, as
well as development partners, have set up their own, parallel systems. This has triggered a
proliferation of indicators used to assess the performance and impact of these programmes. Data
with respect to activities supported by the Global Fund are collected with the aid of ‘executive
dashboards’: Excel-based reporting forms for data collection and analysis. Some of the data is
collected from routine systems such as the Health Management Information System (HMIS),
whereas other information is specific to a particular grant. As a result of these parallel systems,
programme coordinators and managers at the implementation level spend a significant amount of
time on data collection and reporting. Further harmonization of data collection systems, indicators
and evaluation efforts is currently considered a priority under the Third Health Sector Strategic
Framework.
Although GHIs have added to the workload of those involved in M&E activities, they have also
been important contributors to the analytical capacity of the health system. Facilities supported by
the Global Fund have been able to purchase computers and software needed for data collection
and analysis. Workers at the district and facility levels have also received training in essential M&E
skills. Although these trainings are often disease-specific in nature, basic analytical skills are useful
across the health system. Furthermore, these developments have resulted in a greater
appreciation of the importance of M&E at the lower administrative levels.
Service Delivery
Undeniably, the GHIs have had a major impact on the capacity of the Tanzanian health system to
provide basic services for HIV/AIDS and TB. For instance, the Global Fund has enabled large-scale
206

enovation of health care facilities, which were often in dire shape. With support from its
development partners Tanzania has, furthermore, been able to set up a network of CTCs for the
provision of HIV/AIDS services.
Funds have also been used to purchase laboratory equipment and reagents. Some of these
investments have spillover effects beyond their disease-specific primary purpose. TB patients, for
instance, are seen in general health facilities, and investments in waiting and examination rooms
have thus also benefited the wider patient population. Microscopes purchased within the context
of the TB programme are sometimes also being used in the diagnosis of malaria and other
illnesses. Expanding the range of services on offer at health facilities through strengthening of the
HIV/AIDS and TB programmes is assumed to have brought greater trust in the public health system
as a whole. In the longer term this could have a positive impact on the uptake of other services,
such as those for family planning or reproductive and child health.
Discussion
GHIs have been indispensable partners in the fight against HIV/AIDS and TB in Tanzania, making
care and treatment options available to a population in a country that suffers from a high burden
of disease and which is unable to finance such options itself. At times, the investment has been
narrow and has had limited positive impact on the health system as a whole. For instance,
investment in health facilities and laboratories has often concentrated solely on strengthening the
delivery system for HIV/AIDS activities. At other times these efforts have spilled over into the
health system as a whole, for instance through training in transferable skills. Greater emphasis is
needed to systematically build these positive spillover effects into the supported proposals.
Negative impacts are felt mostly in areas of financing, governance and M&E where the presence of
the Global Fund has to an extent resulted in the creation of parallel systems and in duplication of
activities. Significant steps have already been taken to address these issues, but further efforts are
required to bring greater harmonization between the Government of Tanzania and its
development partners.
207

References
[1] The World Bank Group. Tanzania at a glance. Washington, DC, World Bank, 2008.
(http://devdata.worldbank.org/AAG/tza_aag.pdf.; accessed 19 May 2009).
[2] United Nations Development Programme. Human Development Reports: Tanzania. New York,
UNDP, 2008. (http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_TZA.html.;
accessed 19 May 2009).
[3] Tanzania Commission for AIDS. UNGASS Country Progress Report Tanzania Mainland. Geneva,
UNAIDS, 2008.
(http://data.unaids.org/pub/Report/2008/tanzania_2008_country_progress_report_en.pdf.;
accessed 19 May 2009).
[4] UNAIDS. 2008 Report on the Global AIDS Epidemic. Geneva, UNAIDS, 2008.
[5] Tanzania Service Provision Assessment Survey 2006. Dar es Salaam, National Bureau of
Statistics and Macro International Inc., 2007.
[6] World Health Organization. WHO Statistical Information System (WHOSIS).
(http://www.who.int/whosis/en/).
[7] World Health Organization. Global Atlas of the Health Workforce 2008. Geneva, WHO, 2008.
[8] The World Bank Group. World Development Indicators (WDI) Online. Washington, DC, World
Bank.
[9] Atun R, de Jongh TE, Secci FV, Ohiri K, Adeyi O. Integration of Targeted Health Interventions
into Health Systems: A Conceptual Framework for Analysis. Health Policy and Planning. 2009
forthcoming.
[10] Atun RA, Lennox-Chhugani N, Drobniewski F, Samyshkin Y, Coker R. A framework and toolkit
for capturing the communicable disease programmes within health systems: Tuberculosis control
as an illustrative example. European Journal of Public Health 2004;14(3):267-273.
[11] World Health Organization. Everybody’s business: strengthening health systems to improve
health outcomes. WHO’s framework for action: Geneva, World Health Organization, 2007.
208

Zambia: Global HIV/AIDS Initiatives and Health
System Capacity to Cope with the Scale-
Up of HIV Services
Abstract
Phillimon Ndubani 29 , Joseph Simbaya 27 , Aisling Walsh 30 , Ruairí Brugha 28
Zambia has received substantial funding from three HIV/AIDS-focused GHIs over the last seven
years to scale up HIV/AIDS prevention, treatment and care activities. This study assesses the effects
of these GHIs on the health system in Zambia at national and sub-national levels. Two rounds of
qualitative and quantitative data were collected (early 2007 and mid-2008) at national and district
levels, including two urban districts (Lusaka and Kabwe) and one rural district (Mumbwa).
Significant scale-up of HIV services was reported in all three districts studied between 2004 and
2007. While all three GHIs, government and civil society participate in multi-sectoral national
coordination processes, stakeholder roles and responsibilities are often unclear. At district level,
coordination remains weak despite the best efforts of District AIDS Coordination Advisors (DACAs)
and the District AIDS Task Forces (DATFs), which struggle due to inconsistent funding and a lack of
recognition by some stakeholders. Increases in staffing levels have occurred between 2004 and
2007, but only in the two urban districts. Staff workload has increased as a result of scale-up in HIV
services, particularly in Mumbwa rural district. More staff received training for HIV/AIDS services
than for non-HIV services, which was often credited to GHIs. Incentives were more frequently
reported for HIV than for non-HIV services.
Background
The Republic of Zambia is a landlocked country in southern Africa. It is surrounded by eight other
countries: the United Republic of Tanzania and the Democratic Republic of Congo in the north;
Botswana and Namibia in the south; Malawi and Mozambique in the east; and Zimbabwe and
Angola in the west. In 2007, Zambia had an estimated population of 11.9 million people, with an
29 Institute of Economic and Social Research, University of Zambia
30 Royal College of Surgeons in Ireland.
The research for this study summary was funded by The Open Society Institute, New York. The study summary was
produced with the support of the GHIN Network (www.ghinet.org).
209

average annual population growth of 1.7% [1]. In 2006, Zambia ranked 163rd out of 179 on the UN
Human Development Index [2].
Official Development Assistance (ODA) to Zambia in 2006 was US$ 945 million. In 2005, the
International Monetary Fund (IMF) and the World Bank’s International Development Association
provided the country with debt relief of approximately US$ 6 billion [3].
In 2008, Zambia’s health sector budget was approximately US$ 295 million or US$ 30 per capita,
and donor funds represented 50% of total national health expenditure. In 2007, 65% of Zambians
lived in rural areas, but only 52% of all health workers worked in rural settings [4]. Thirteen districts
did not have a single doctor, while more than half of all doctors practiced in the capital, Lusaka [5].
HIV/AIDS is the leading cause of death for all ages in Zambia. In 2007, an estimated 1.1 million
people in Zambia were living with HIV [6], Prevalence, however, has slightly decreased in recent
years from 16% in 2002 to 14% in 2007 [7]. Prevalence is twice as high in urban settings as in rural
areas, at 20% and 10% respectively [7].
Zambia has received large amounts of external HIV/AIDS funding over the last seven years to scaleup
prevention, treatment and care activities [8]. Although Zambia receives funds from several
donor agencies, the major contributors are three global HIV/AIDS initiatives: the Global Fund;
PEPFAR; and World Bank MAP (table 3). The national multi-sectoral AIDS response in Zambia is
coordinated by the National AIDS Council (NAC).
The Global Fund allocates funding to four Principal Recipients in Zambia. Two of these are
government agencies: the Ministry of Finance and National Planning (MoFNP) and the Ministry of
Health (MOH). Two are non-government: the Churches Health Association of Zambia (CHAZ) and
Zambia National AIDS Network (ZNAN).
The World Bank MAP (from 2002-2008) has channelled its resources to support the Zambia
National Response to HIV/AIDS Project (ZANARA), which is housed in the Ministry of Finance and
National Planning. It allocates resources to: the Community Response to HIV/AIDS (CRAIDS);
workplace programmes in the line ministries; and the NAC. The Community Response to HIV/AIDS
is the only World Bank funds recipient in Zambia that sub-grants to implementing local
organizations.
Zambia was chosen as one of PEPFAR’s fifteen focus countries in 2003, and by 2006 PEPFAR
contributed 62% of total HIV/AIDS funding in Zambia [9]. The major recipient of PEPFAR funds is
the United States Agency for International Development (USAID), and implementing agencies
mainly include international nongovernmental organizations (NGOs) and faith-based
organizations (FBOs), which access funds directly from PEPFAR. The Country Operational Plan
(COP) is prepared each year and is the framework through which PEPFAR-supported activities are
undertaken.
210

Table 1 Basic Socioeconomic, Demographic and Health Indicators*
(*) Full data sources for all indicators are provided in Annex 1.
Indicator Value Year Source
Population (thousands) 11,920 2007 World Bank
Geographic Size (sq. km) 743,390 2007 World Bank
GDP per capita, PPP (constant 2005
international $)
1,282.86 2007 World Bank
Gini index 50.8 2004 World Bank
Government expenditure on health (%
general government expenditure)
10.8 2007 WHO NHA
Per capita government expenditure on
health (current US$)
25
2007 WHO NHA
Physician density (per 10,000) 1 2004 WHO SIS
Nursing and midwifery density 20 2004 WHO SIS
Maternal mortality ratio (per 100,000 live
births)
830 2005 WHO SIS
DTP3 coverage (%) 80 2007 WHO SIS
Estimated adult HIV (15-49) prevalence (%) 15.2 (14.3 –
16.4)
2007 UNAIDS
Estimated antiretroviral therapy coverage
(%)
46 (40-56) 2007 WHO/UNAIDS/UNICEF
Tuberculosis prevalence (per 100,000) 387 2007 WHO GTD
Estimated malaria deaths 14,204 2006 WHO WMR
Table 2 Global Health Initiative Investments*
Global Fund
Round & Disease Priority Approved (in US$) Disbursed (in US$)
Round 1, HIV/AIDS 90,325,778 81,859,517
Round 1, Malaria 39,273,800 38,673,791
Round 1, TB 47,337,256 29,883,503
Round 4, HIV/AIDS 236,318,738 97,700,670
Round 4, Malaria 42,721,807 27,725,056
Round 7, Malaria 17,715,924 2,422,691
Round 7, TB 3,882,948 1,208,954
Round 8, HIV/AIDS 129,368,645 0
TOTAL: 606,944,896 279,474,182
PEPFAR
Year Amount Allocated (in US$)
2004 81,662,410
2005 130,088,605
211

2006 149,022,153
2007 216,012,780
2008 269,246,552
TOTAL: 846,032,500
GAVI
Disease Priority Amount Approved (in US$)
Pentavalent vaccine 46,540,000
Tetravalent vaccine 8,812,000
Vaccine introduction grant 100,000
Injection Safety 771,000
Immunization services support 3,864,060
Health systems strengthening 6,605,500
TOTAL: 62,692,116
World Bank MAP
Project Title FY Approved/Closing Date Commitment (in US$)
Zambia National Response to
HIV/AIDS
Objectives and Methodology
2003/2008 42,000,000
In view of the injection of large amounts of GHI funding into a fragile health system, there is a need
to understand GHIs’ effects in Zambia at national and sub-national levels. Specifically, this study
tracks the effects of GHIs on: scale-up of services; human resources for health; coordination of
services; and harmonization of donor priorities and activities.
In late 2006, a national context mapping exercise was carried out to document HIV/AIDS services
and structures at the national level. Two rounds of data collection followed, at both national and
sub-national levels, in January/February 2007 and June/July 2008.
Three districts were purposively selected to represent urban (Lusaka and Kabwe districts) and rural
areas (Mumbwa district). In each district, a list of health facilities and NGO-run facilities that were
delivering HIV/AIDS services was compiled, and facilities were then purposively selected, including
all government and NGO facilities providing antiretroviral treatment (ART) and a random sample of
facilities not providing ART. Facility data (numbers of service episodes/clients, pharmacy records,
laboratory records and human resources data) were collected from health facilities (n=39): 12 each
from Lusaka and Mumbwa, and 15 from Kabwe. Inaccessibility during the rainy season led to two
government facilities being excluded in Mumbwa.
Where records were incomplete or unavailable, they were supplemented with data from District
Health Management Team (DHMT) reports. Health staff structured questionnaires (n=234) were
administered to frontline health workers: doctors, clinical officers, nurses, laboratory and pharmacy
staff and other staff delivering HIV/AIDS services that were available to participate at each facility.
212

In-depth qualitative interviews were conducted at national level (n=18) with key informants from
government, bilaterals, multilaterals, NGOs and GHIs; and at district level (n= 43) with members of
the DHMT, health facility managers, NGOs, DATF representatives and community-based
organizations (CBOs).
All quantitative data collected from interviews and facility records were entered in Epi-data and
analysed using both SPSS (Version 16.0) and SAS (Version 9.1) statistical packages. Qualitative
interviews were recorded, data transcripts were typed in Microsoft Word and coding and analysis
were carried out using Atlas.ti software. This was followed by thematic analysis of the qualitative
outputs. Analysis of both quantitative and qualitative data is ongoing throughout the first half of
2009. The study was approved by the University of Zambia Research Ethics Committee.
The study is funded by the Open Society Institute (OSI) with partners from the Royal College of
Surgeons in Ireland (RCSI) and London School of Hygiene and Tropical Medicine (LSHTM). This
study is also part of the Global HIV/AIDS Initiatives Network (GHIN).
Results
Leadership and Governance
The NAC, established in 2002 in advance of the arrival of substantial external funding for HIV/AIDS,
is the main HIV/AIDS coordinating mechanism in Zambia. It operates in parallel to the Country
Coordinating Mechanism (CCM) for the Global Fund. Qualitative interviews with key informants
credited all three GHIs with participating in multisectoral national coordination structures and
processes. The World Bank MAP has provided capacity support to the NAC secretariat, which in
turn gives secretariat support to the Global Fund CCM. Roles and responsibilities have been illdefined
between NAC, the MOH, other ministries and CSOs, and between the various sub-national
structures and actors, something that NAC itself has documented [8].
CSOs were recognized as playing a key role in district level coordination. ZANARA, funded by the
World Bank, has supported the community response to AIDS by financing community-based
organizations (CBOs) who also participate in DATFs and Community AIDS Task Forces (CATFs).
Recipients of Global Fund resources were also credited with participation in coordination at district
and sub-district level.
Some respondents credited each district with adapting and interpreting the National Strategic
Plan for HIV/AIDS to suit the district, according to need on the ground, while others saw the
process as being much more top-down. The importance of DATFs in ensuring that all the HIV/AIDS
services are well coordinated was recognized by many informants. The United Nations
Development Programme (UNDP) has funded District AIDS Coordination Advisors (DACAs) to
facilitate more effective communication and coordination between HIV/AIDS programmes and
activities on the ground. Informants reported widely that it has been difficult for DACAs to operate
successfully due to erratic funding from NAC for the DATFs. DACAs face additional barriers in
fulfilling their coordination roles, including an unclear mandate and – in some cases – lack of
recognition by all the stakeholders supporting and/or implementing HIV/AIDS services.
While PEPFAR participates in coordination structures at the national level, its recipient
organizations directly fund NGOs in the community and in many instances do not register with
DATFs. Consequently, the DACAs often lack full knowledge of all players providing HIV/AIDS
services in their districts. According to informants, lack of coordination at the district level not only
213

isks duplication of services but also creates difficulties in compiling accurate data on the numbers
of clients accessing services from the various delivery points. DATFs have begun to provide
registers of services in the districts, and Kabwe’s referral network is a good example of a setting
where district coordination is working. NGOs that participate in DATFs are credited with
improving communication amongst stakeholders at the district and community level.
Health Workforce
Total numbers of health staff increased in the sampled facilities by 77% between 2004 and 2007.
When broken down, this shows an increase of 63% for clinical staff (doctors, clinical officers and
nurses) and a 126% increase in pharmacy and laboratory staff. However, this increase occurred in
the urban districts only, and Mumbwa rural district actually experienced slight reductions in
numbers of nurses (from 70 in 2004 to 67 in 2008) and clinical officers (from 26 in 2004 to 20 in
2008). Interviews with key informants reveal a growing concern about staffing levels, despite
appreciation of the scale-up of HIV/AID services. Informants affirmed that, as a result of increased
patient/client load, health staff were overburdened by work, especially in rural areas. Several
respondents spoke of rural health centres with only one staff member who was rolling out ART in
addition to other routine services. Health staff ratios to catchment populations increased in urban
districts and fell in the rural district. Respondents reported that counsellors and volunteers have
helped to relieve some of the workload of clinical staff, but for HIV counselling only. Quantitative
findings show that HIV counsellors have increased by 173% across the three districts, with a higher
increase in urban areas.
Interviews reported a high turnover of health staff from the public sector to NGOs that offered
better conditions of service. Two national-level respondents attributed this pattern to PEPFAR. A
number of respondents in Mumbwa spoke of rural health centres having problems attracting
health staff due to a lack of accommodation, despite the introduction of the rural retention
programme, which provides a hardship allowance, housing rehabilitation and vehicle loans in turn
for three years of service at a rural health facility.
The study also explored issues of training and incentives for providing HIV versus non-HIV services.
Forty percent of health staff received training in ART between June 2007 and June 2008, compared
to 26% in child health, 13% in maternal health and 12% in malaria. There were no significant
differences in training by staff category or urban-rural location. Over half of the 234 staff surveyed
reported receiving incentives for delivering HIV services, compared to a quarter who reported
receiving incentives for delivering non-HIV services. The important contribution of GHIs to training
was noted by respondents, and all three GHIs were mentioned as contributing positively.
However, some respondents reported that staff were spending too much time on training,
resulting in an increase in workload for those who remained at post.
Service Delivery
Significant scale-up of HIV/AIDS services – ART, prevention of mother to children transmission
(PMTCT) and voluntary HIV counselling and testing (VCT) – has occurred in all three districts
studied from 2004 to 2007. In 2008, 89% of surveyed facilities in Lusaka, 80% in Kabwe and 39% in
Mumbwa were providing ART. All 39 facilities sampled across the three districts provided VCT, and
89% of sampled facilities in Lusaka provided PMTCT, as did 100% in Kabwe and Mumbwa.
214

Numbers of clients receiving ART increased consistently from 2004 to 2007 across all three
districts. The numbers of clients receiving VCT also increased, with most of the VCT delivered at
sub-district level across urban and rural districts. Numbers of women receiving PMTCT also rose,
primarily in urban areas. Qualitative interviews with key informants from the national and district
levels confirmed the scale-up of HIV/AIDS services and cited greater uptake as a result of the
increased availability of services. All informants attributed scale-up to the GHIs. Although some
respondents spoke of GHIs generally, each GHI was also specifically mentioned as contributing.
There were also marked improvements in availability and accessibility of community-level services
and support services. These include home-based care (HBC), lay community counselling and
treatment support groups for HIV/AIDS. There was an increase in the numbers of organizations
providing support to community level HIV/AIDS services. It was reported that the GHIs had not
only provided financial support, but also helped in raising the capacities of community-based
groups. CRAIDS – with funding from World Bank MAP – in particular was credited with building the
capacity of such groups.
Medical Products, Vaccines, and Technologies
Drug availability for HIV/AIDS was reported to have increased. In 2006, no sampled facilities
experienced stock-outs of first line antiretroviral (ARV) drugs. 2007 findings show that two facilities
in Lusaka and Mumbwa ran out of first line ARVs, and another two facilities in Lusaka ran out of
second line ARVs. Six out of 12 facilities in Lusaka, nine out of 15 in Kabwe and 10 out of 12 in
Mumbwa reported experiencing stock-outs of first line malaria drugs in 2007.
Discussion
Significant scale-up of HIV/AIDS services has occurred at national and district levels in Zambia in
the last five years. This study documented the scale-up of services in two urban districts and one
rural district. While it proved very difficult to obtain data on funding flows, and categorical
attribution for this scale-up to the GHIs is difficult, the large level of funding from GHIs has
undoubtedly been crucial.
Notwithstanding the positive effects of GHI funding and support, limitations have been observed.
There is evidence of rural-urban inequities in staff recruitment and placements. The scale-up of
services for HIV/AIDS has occurred without a corresponding increase in health staff. To date,
however, none of the GHIs has invested money in hiring new staff. This has inevitably resulted in
higher workloads and extra strain on health workers. These findings support national-level
evidence that health worker numbers are insufficient, particularly in rural areas. The Zambian
Government, through the MOH, has developed a Human Resources for Health Strategic Plan 2006-
2010 in consultation with donors, including the three GHIs. The strategies and activities outlined in
the Plan attempt to provide a framework to guide and direct interventions, investments and
decision-making in the planning, management and development of human resources for health
[10].
The ratio of physicians to population stands at 1:10,000, lower than the WHO minimum
requirement of one doctor per 5000 population and 25 clinical staff (doctors, nurses and midwives)
per 10 000 population [11]. Our results show that health worker density for Lusaka, Kabwe and
Mumbwa was less than the 7.9 per 10 000 reported nationally [12]. More than 50% of rural health
215

centres have only one qualified health worker [11]. The urban-rural breakdown in our study shows
that virtually all of the recent increases in clinical staff numbers were in urban health centres, with
a slight reduction in rural health centres. The Human Resources for Health Strategic Plan is
attempting to address these gaps by introducing a rural placement and retention package.
Despite support from GHIs, this scheme has been slow to roll out due to shortage of
accommodation and a short timeframe for retention allowances. In addition, until 2007, only
doctors were eligible for the scheme, but it has now been extended to include nurses and nurse
tutors. Our findings support other studies that reported that PEPFAR-funded NGOs have attracted
staff away from the public sector [13,14]. Health staff in Mumbwa – where PEPFAR funded
organizations do not operate – were less likely to receive financial incentives for delivering HIV
services. Given their success in contributing to HIV/AIDS service scale-up, GHIs now need to
support government to train and retain new staff. To date, such support has not been
forthcoming. PEPFAR has committed to training new health workers in its focus countries,
although this had not commenced in Zambia at the time of the study.
A multisectoral response, across ministries and including non-state actors, has been adopted
throughout the implementation of Zambia’s HIV/AIDS programme. The Zambian National
HIV/AIDS Strategic Framework 2006–2010 demonstrates commitment to coordination through
policies explicitly embracing multisectoral HIV/AIDS control, including NGO engagement in
decision making. Despite this commitment, findings suggest weak coordination at the national
and particularly at the district levels, which has been compounded by the increasing numbers of
stakeholders. Although some informants hailed the NAC for its coordination role, they believed
that some stakeholders undermine the mandate of the council. Some PEPFAR-funded
organizations have sidelined the government coordination structures in their dealings with
district- and community-based organizations. This is seen as a source of conflict, as it does not
promote the collective spirit of “Three Ones” principles: one national AIDS framework, one
coordination body, one monitoring and evaluation (M&E) plan.
It is evident that greater coordination is required at the district level, where implementation of
activities takes place, if duplication of services and difficulties in accounting for numbers of clients
are to be avoided. Limited resources allocated to DATFs, along with the limited devolution of
decision-making powers, acutely affect district and community-level structures.
While direct attribution of system effects to specific GHIs is difficult, it is clear that the significant
resources provided by the GHIs have had effects, both positive and negative, at the district level in
Zambia. Given the amount of resources available from the GHIs for implementing HIV/AIDS
services – the Global Fund and PEPFAR are still active, although the World Bank MAP grant ended
in 2008 – the multiplicity of stakeholders makes coordination difficult and essential. Together, the
GHIs, other donors and government need to focus on training health workers and managers and
building up weak human resource capacity at all levels. We pose two recommendations at this
stage:
1. While the three GHIs have made positive contributions to human resources and in
particular in-service training, the issue of health worker shortages has been exacerbated by
the scale-up of HIV services. The GHIs should invest in new health worker hire in alignment
with the government strategic plan for human resources.
2. The GHIs should support capacity development for DATFs as the main coordination body
at the district level. All GHIs – but PEPFAR in particular - should encourage their recipient
organizations to work with DATFs.
216

References
[1] The World Bank Group. Zambia at a glance (http://devdata.worldbank.org/AAG/zmb_aag.pdf.;
accessed 17 March 2009).
[2] United Nations Development Programme. Human Development Reports: Zambia
(http://hdrstats.undp.org/en/2008/countries/country_fact_sheets/cty_fs_ZMB.html; accessed 17
March 2009).
[3] The World Bank Group. Zambia Country Brief. Washington, DC, World Bank, 2008.
[4] Ministry of Health, Republic of Zambia. National Health Strategic Plan 2006-2010. Lusaka, 2005.
[5] Herbst CH, Gijsbrechts D. Comprehensive and Accurate Information on Health Worker Stock,
Profiles, and Distribution in Zambia: Analysis of the JICA Data. Conference presentation. Human
Resources for Health Research Conference. Lusaka, Zambia, 2007.
[6] UNAIDS. 2008 Report on the global AIDS epidemic. Geneva, UNAIDS, 2008.
[7] Zambia Demographic and Health Survey 2007. Lusaka, 2009.
[8] Zambia National AIDS Council. Zambia Country Report: Multi-sectoral AIDS Response
Monitoring and Evaluation Biennial Report 2006-2007. Lusaka, NAC, 2008.
[9] Ooman N, Bernstein M, Rosenzweig S. Following the Funding for HIV/AIDS: a comparative
analysis of the funding practices of PEPFAR, the Global Fund and the World Bank MAP in
Mozambique, Uganda and Zambia. Washington, DC, Center for Global Development, 2007.
[10] Ministry of Health, Republic of Zambia. Human Resources for Health Strategic Plan (Draft)
2006-2010. Lusaka, 2005.
[11] Ministry of Health, Republic of Zambia. Human Resource Crisis in Zambia. A Paper for The High
Level Forum Abuja, December 2004.
[12] Ministry of Health, Republic of Zambia. Joint Annual Review for 2007 Main Report. Lusaka,
2008.
[13] Hanefeld J, Musheke M. What impact do Global Health Initiatives have on human resources for
antiretroviral treatment roll-out? A qualitative policy analysis of implementation processes in
Zambia. Human Resources for Health, 2009, 7:8.
[14] Ooman N, Bernstein M, Rosenzweig S. Seizing the Opportunity on AIDS and health systems: a
comparison of donor interactions with national health systems in Mozambique, Uganda and
Zambia, focusing on the US President’s Emergency Plan for AIDS Relief, the Global Fund to Fight
AIDS, TB and Malaria, and the World Bank’s Africa Multi-country AIDS Program. Washington, DC,
Center for Global Development, 2008.
217

Policy analysis of the impact of
Global Health Initiatives on health systems:
Policies and lessons learned from donor programmes*
Anne Rossier Markus, Seble Frehywot, Amie Heap, Alan Greenberg
Abstract
Understanding the broader context under which GHIs operate and the flexibility they grant
countries that apply for funding for health system strengthening (HSS) either separately or as an
integral component of a disease focused plan, is key in understanding the types of synergies that
may take place at the local, regional, and national levels. GHIs express this flexibility in very broad
terms and communicate their flexibility in different ways and often independently from each
other. This broad guidance and variation in communication must be counterbalanced with how
these messages are perceived by countries as well as the need for countries to better understand
in which health system areas they may request funds from individual GHIs. The proper balance
must be found between each GHI’s mission, goals and overarching operating model, and the need
for countries to have increased clarity about funding options and the likelihood of having their
applications met with success. GHIs should strive to achieve this balance “internally” within their
own programmes and “externally” with other GHIs that fund similar, overlapping, and even
distinct activities. This paper summarizes key findings from a systematic review of four GHIs’ laws,
policies and guidance, key informant interviews, and funded country applications for HSS,
particularly in the area of health workforce production, distribution, and retention.
Background
Major GHIs, such as GAVI and the Global Fund, have been developed and implemented over the
past several years to help reduce disease-specific morbidity and mortality in low and middleincome
countries. Core principal policies from donor countries and organizations have guided to a
great extent the trajectory and implementation of GHI funds at the host country level. Specifically,
these policies may influence how GHI funds can or cannot be used to strengthen different
elements of the health system as defined by WHO, namely health workforce, health information,
service delivery, governance and leadership, health financing, and medical products and
technologies. Concurrently, an increasing number of host countries have developed policies at the
national and local levels to address the different elements of their health systems, the specific
needs of their countries as they relate to varying disease burdens, and the three health-related
Millennium Development Goals. Despite general guidance from GHIs of how and when countries
can use GHI funds for HSS, these policies have been interpreted differently by various host
countries and implementing partners. Therefore, to maximize potential synergies between GHI
funding and health systems, it is important to improve our understanding of these guiding GHI
policies, and how they are viewed and interpreted at the host country level.
218

Total resources in HIV/AIDS in developing countries have increased from approximately $300
million in 1996 to $10 billion in 2007 [1] representing a substantial and unprecedented increase in
total funding devoted to the disease. The GHIs, which are the focus of this study, have contributed
greatly to this increase in funding. This section briefly summarizes similarities and differences in
GHI funding priorities, eligibility criteria and application processes and in GHI emphasis on HSS.
GHIs in General
GHIs vary in their eligibility criteria for country applicants, the general authorized use of funds, and
the grant application review criteria used by panels of appointed reviewers, which can be
independent (GAVI, Global Fund) or within the GHI’s organization (WB, PEPFAR).
GAVI
The GAVI Fund provides support to countries that are eligible based on national income (i.e.
countries with Gross National Income per capita

countries” [6]. The TRP apply several general criteria to reviews of applications: “Soundness of
approach; Feasibility; and Potential for sustainability and impact” [6]. Each of these criteria is
spelled out in greater detail in the Terms of Reference of the Technical Review Panel.
The World Bank MAP
The World Bank provides low-interest loans, interest-free credits, and grants to developing
countries for a wide array of purposes that include investments in education, health, public
administration, infrastructure, financial and private sector development, agriculture, and
environmental and natural resource management [7]. Typically, the Bank requires applicants to
produce a strategic plan to demonstrate how they expect the financing to have a desirable effect.
In the area of health, the Bank (via IDA) supports systemic change building on priorities and needs
identified by recipient countries. Governments develop Country Assistance Strategies to
determine areas for IDA assistance. IDA funding can be used flexibly to complement other sources
and deliver sustained support to strengthen health systems. Similarly, for the MAP each country
project has to have a national strategic plan as its basis, which is negotiated and agreed upon
between each country and the Bank [3]. MAP focuses mostly on a national AIDS response, capacity
building and institutional strengthening, with strong monitoring and evaluation (M&E) built in.
Speed, scaling-up existing programmes, building capacity, "learning by doing" and continuous
project revisions has necessitated significant reliance on the immediate M&E of programmes to
determine which activities are efficient and effective and should be expanded further, and which
are not and should be stopped or would benefit from more capacity building [8].
PEPFAR
The original five-year financial commitment of PEPFAR was $15 billion dollars supported by
American taxpayers. The actual expenditure during the following five-year period (2003-2008) was
$18.8 billion dollars. On July 30 2008 the Tom Lantos and Henry J. Hyde United States Global
Leadership Against HIV/AIDS, Tuberculosis and Malaria Reauthorization Act of 2008 was signed
into law. This legislation authorizes a U.S. financial commitment of up to US$ 48 billion to combat
these three diseases over the next five years [9] . The United States Government (USG) uses the
Country Operational Plan (COP) to award funding, most of which goes to international, mostly US,
non-governmental entities and is heavily concentrated in a number of focus countries (15 until
2008 now expanding to more under the reauthorization of PEPFAR) [3]. The 2008 law removes
many of the earmarks, which previously determined allocation of funds, although it still requires
that over 50% of the funding be allocated to the treatment of HIV/AIDS (e.g. ARV drugs, ARV
services, laboratory infrastructure), rather than to prevention and care. In PEPFAR I, countries were
required to spend some funds on strategic information (up to seven percent) and policy analysis
and HSS (up to seven percent). In PEPFAR II, HSS, including the rational allocation of health tasks
among health care workers (“task-shifting”), has taken on a heightened priority and is an explicit
goal of the initiative [10]. Although criteria for awarding funds are not provided in general
guidance, recipients of funds are required to meet a number of expectations, such as adherence to
Emergency Plan policy, collaboration with the Global Fund, accountability and reporting [11].
GHIs and Health Systems Strengthening
GHIs and experts alike recognize that progress toward providing treatment to persons affected
with HIV/AIDS and other diseases has been achieved in countries receiving funding. However,
they also acknowledge today that limited progress will be achieved from now on without paying
special attention to, not only the country’s macroeconomic framework, but also its health care
system. Particular attention will also need to be paid to the country’s legal and policy framework
for health and health systems in order to achieve greater integration of disease specific funding,
220

oth national and external, with the broader system. “Stronger and robust health systems are
essential” in a successful response to HIV/AIDS, tuberculosis, and malaria [12]. They are also
essential in ensuring the long-term, sustainable growth of national health systems and the overall
health of country populations. Over the past few years, HSS has been viewed as a necessary
component of a comprehensive GHI approach to combating infectious disease. An understanding
of this principle has led the major GHIs to consider the place of health sector capacity building
from a strategic, policy, and funding perspective.
As previously noted, all four GHIs had already included HSS as a consideration for their funding.
(See Table 1, below, for some of the historical evolution of the concept.) However, there is no
general consensus definition of HSS among the different GHIs, although the WHO framework is
often referenced. In regard to Human Resources for Health (HRH), however, there is growing
consensus that HRH constitutes all areas of the health (direct, indirect) workforce as well as the
health workforce that is needed to work in WHO's other five elements of the heath system (e.g.
health managers, health policy analysts, health finance personnel, and HMIS personnel, among
other cadres).
Table 1: GHIs & Historical Emphasis on HSS
GAVI
GAVI since its inception (as expressed in its by-laws) has made HSS an integral and explicit component of its mission. In
addition, more recently, in an effort to promote sustainability, GAVI is implementing a new co-financing policy. The
results of a study commissioned by the GAVI alliance in 2004 showed that health system issues beyond the
immunization system alone constrained the majority of the developing partner countries trying to increase or maintain
high immunization coverage. The GAVI Alliance determined that overcoming some of these health system barriers was
a major priority for the initiative and that addressing them would improve access to other child and maternal health
services, thereby increasing synergy. 1 In 2006, the GAVI Alliance Board invested $500 million for HSS for the 2006 to
2010 period and issued and posted on-line HSS-specific guidelines. 2
Global Fund
The Global Fund began accepting separate proposals for HSS in 2005 during the Round 5 proposal session, although
disease-specific proposals could also include a HSS component if they met Global Fund criteria. In 2007, the Global Fund
Board made the decision to “encourage applicants, wherever possible, to integrate requests for funding for HSS actions
within the relevant disease components.”3 Proposals for Round 9, due in June of 2009, allow countries to apply for
funding necessary to overcome fundamental health system and human resource constraints to successful and sustained
scale-up of HIV, TB, and malaria interventions.4 This round will also fund portions of national health workforce
strategies.5 HSS activities, for which applicants may apply, can be tied to a particular disease or system-wide, crosscutting
activities that benefit not only a particular disease program but also a wide range of health priorities.6 Currently,
35% of the $4 billion of approved financing within the Global Fund is supporting key health systems elements.
Additionally, $186 million was approved in the Round 7 application process for cross-cutting HHS funding.7 A $290
million allotment for cross-cutting HHS funding was recommended during the Round 8 process.8 9
World Bank
WB IDA and MAP have as a premise “building systems and institutions to channel resources to affected communities and
bring public, private, nonprofit sectors together to delivery effective evidence based strategies and policies.” A focus on
HSS was initiated at the World Bank in 1999. 10 At that time the Bank developed a new strategy for responding to the
HIV/AIDS epidemic, with a focus on the African continent. 11 This strategy sought to create a flexible way to provide
resources quickly, both to governments and civil society, within sound strategic frameworks to begin implementing
national programs while strengthening institutions and accountability. 12 As of 2006 the World Bank MAP had
committed $534 million (approximately 41% of total funds) to systems strengthening and $223 million to the health
sector. 13
221

PEPFAR I & II
The newest PEPFAR legislation has committed to invest resources that would allow partner countries to provide both
health workforce training and capacity-building. 14 The legislation has a stated priority to increase training and retention
of health care professionals, paraprofessionals and community health workers. In-service and pre-service training of
140,000 new health care professionals and paraprofessionals with an emphasis on training and in country deployment of
doctors and nurses is a first time commitment in the new legislation. The amount of funding PEPFAR country teams
could use to support long-term pre-service training increased to $6 million or 3%. 15 As stated above, under PEPFAR II,
HSS including “task-shifting,” has taken on a heightened priority and is an explicit goal of the initiative. 16 Finally, PEPFAR
II calls for stronger collaboration with other donors, including GFAMT, WHO, WB, and UNAIDS, review of policies that
may be obstacles to HSS, and activities that “complement” other donors and other programs from the U.S., such as MCH
programs. 17
1 The GAVI Alliance. (2008). Health System Strengthening. Information retrieved on March 17, 2009 from:
http://www.gavialliance.org/resources/6_HSS_overview_EN.pdf
2 Idem
3 The Global Fund. (2009). Information retrieved on March 17, 2009 from:
http://www.theglobalfund.org/en/fundingdecisions/
4 The Global Fund. (2009). Clinic 1F: Capacity Development: Using Global Fund Grants to Strengthen Health Systems.
Information retrieved on March 17, 2009 from:
http://www.theglobalfund.org/documents/partnershipforum/2008/presentations/0812/C1/5PM/HSS%20and%20CSS%2
0vf-short%208-12-2008.ppt#256,1,Clinic 1F Capacity Development: Using Global Fund grants to strengthen health
systems
5 The TPR Terms of Reference list as a sustainability and impact criteria for proposals the demonstration of how they will
contribute to strengthening the national health system in its different components (human resources, service delivery,
infrastructure, procurement, and supply management).
6 Physicians for Human Rights. (2009). Toolkit for Using Round 9 of the Global Fund for Health Systems Strengthening.
Information retrieved on March 17, 2009 from: http://physiciansforhumanrights.org/hiv-aids/globalfund_round9.html
7 Idem
8 Idem
9 Physicians for Human Rights. (2009). Toolkit for Using Round 9 of the Global Fund for Health Systems Strengthening.
Information retrieved on March 17, 2009 from: http://physiciansforhumanrights.org/hiv-aids/globalfund_round9.html
10 The World Bank MAP. (2009). Overview. Information retrieved on March 17, 2009 from:
http://siteresources.worldbank.org/EXTAFRREGTOPHIVAIDS/Resources/717147-1181768523896/overview.pdf
11 World Bank, IDA at work: Health – Supporting Systemic Change in a New Global Context, March 2007
12 The World Bank MAP. (2009). Overview. Information retrieved on March 17, 2009 from:
http://siteresources.worldbank.org/EXTAFRREGTOPHIVAIDS/Resources/717147-1181768523896/overview.pdf
13 Idem
14 The President’s Emergency Plan for AIDS Relief. Information retrieved on March 17, 2009 from:
http://www.pepfar.gov/documents/organization/105844.pdf
15 The U.S. President’s Emergency Plan for AIDS Relief. Information retrieved on March 16, 2009 from:
http://www.pepfar.gov/press/fifth_annual_report/113720.htm
16 HR 5501, Pub L 110-293
17 Idem
222

Objectives and Methods
In order to produce this policy analysis, the general approach adopted was qualitative. As such,
GW researchers relied on a combination of documentary reviews and interviews with key
informants from the four GHIs included in this study as well as individuals knowledgeable about
the GHIs studied. Our main questions guiding the analysis were:
• What are the policies that guide the operations of each GHI and what are some of the
common and different traits shared by the four GHIs in this respect?
• What do these policies indicate about funding priorities related to HRH, one of the six
health system elements identified by WHO?
A typology was developed to guide the systematic analysis and extraction of language from
official GHI documents (e.g. laws, policies, guidances) and from a select number of country
applications (see Table 3, which displays the countries included for analysis for each GHI, which are
a subset of the MPS country case studies). Key domains of the typology were: 1) Education – preservice,
in-service, general workforce capacity strengthening, producing different types of health
workers; 2) Financial incentives – salary payment, top-ups, rural allowances, pay-for-performance,
general recruitment support; 3) Management – general improvements in infrastructure, training of
health managers, support for continuous professional developments, career paths. These domains
were examined for three types of health workforce – providers who are involved in direct clinical
care, providers who are involved in indirect clinical care, and workers who are non-clinical but
support the health system in some capacity (e.g. health managers, health planners).
Collection of all of the documents publicly available on the websites of each GHI, including
country applications, as well as other relevant publications from think tanks and other
organizations, occurred over a 1.5 month period and analysis of these documents took place over
the following 2.5 month period. While the documentary review was underway, key informant
interviews were conducted in person or by telephone. At least one representative from each GHI
agreed to participate in the interviews, but in one case we were unable to secure a date and time
due to the need to reschedule multiple times. Because key informants were assured
confidentiality, the information gathered is not directly quoted or attributed and was strictly used
to complement the document reviews in developing the policy analysis.
All of the language found in GHI official documents and country applications was extracted and
inputted in tables for easier reference and comparison (see Table 4 for several of these tables). The
George Washington University Institutional Review Board (GWU IRB) determined this study to be
exempt research under #010924.
Results
GHIs differed in how extensive their official, written guidance was to country applicants, but GHI
documents, informants, and country applications highlighted the inherent flexibility that exists
around the use of funding for health system-related activities, including interventions pertaining
to the health workforce. Guidance is generally, and not surprisingly, worded quite broadly and can
be interpreted to encompass many different aspects of strengthening the workforce in a given
country. It also varied in what aspect of HRH was emphasized. This section summarizes key
findings in three key areas: 1) types of fundable HRH interventions; 2) types of fundable health
workers; 3) comparison of approved country applications and GHI guidance.
223

Types of fundable HRH interventions
Table 2, below, shows where language was found that pertained to a specific aspect of HRH.
Language can be very broad and subject to an inclusive interpretation or alternatively quite
specific and directive. Some highlights:
• Overall, the four GHIs combined addressed in some respect most of the domains related to
HRH although they varied by the amount of on-line guidance available to countries.
• All GHIs had some level of general guidance on pre-service education for direct, clinical
health workers and three GHIs had guidance on in-service education for these workers,
and all GHIs had guidance on general workforce capacity strengthening for all health
workers as well as for general infrastructure improvements and training of health
managers.
• PEPFAR had explicit guidance on training different types of direct, clinical health workers.
• GAVI and WB more frequently addressed specific areas of financial incentives, such as topups,
pay-for-performance, and general recruitment support.
• None of the GHIs had explicit guidance in relation to the use of funding for rural
allowances.
• WB had explicit guidance on supporting continuous professional development and career
paths for all types of health workers.
Table 2. Guidance by GHI and HRH Intervention Categories (as of May 2009)
HRH Intervention Categories
EDUCATION
Pre-service education
GAVI (G) 1 , GFAMT (GF) 2, World Bank
MAP Program (WB) 3 , PEPFAR (P) 4
DIRECT PROVIDERS INDIRECT
Clinical Non-Clinical All
G, GF, WB, P G, GF, WB, P G, GF, WB
In-service education GF, WB, P GF, WB, P GF, WB
Producing different types of health workers P
General Workforce Capacity Strengthening
FINANCIAL INCENTIVES
G, GF, WB, P G, GF, WB, P G, GF, WB, P
Salary payment GF, WB, P GF, P GF, WB
Top-ups G WB G WB G WB
Rural allowances (P) 5
Pay for performance G WB G WB, P G WB
General Recruitment Support
MANAGEMENT
G WB G WB G WB
General infrastructure improvements G, GF, WB, P G, GF, WB, P G, GF, WB, P
Training of health managers G, GF, WB, P G, GF, WB, P G, GF, WB, P
Support for continuous professional
development, career paths
WB WB WB
1 GAVI (2009). Revised GAVI HHS Guidelines for: GAVI Alliance Health System Strengthening Applications (page 8).
Information retrieved on January 15, 2009 from:
www.gavialliance.org/resources/HSS_Guidelines___2009_UPDATE___Final.pdf
224

2 Healthy Systems 2020 (2008). Toolkit for Using Round 9 of the Global Fund Health Systems Strengthening.
www.healthsystems2020.org/files/2122_file_GF_HSS_Toolkit_FIN.pdf
3 The World Bank (2008). The World Bank’s Commitment to HIV/AIDS in Africa: Our Agenda for Action, 2007-2011.
http://wwwwds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2008/05/30/000333038_20080530023715/Rendered/
PDF/439600PUB0Box310only109780821374481.pdf
4 The Library of Congress (2008). Tom Lantos and Henry J. Hyde United States Global Leadership Against HIV/AIDS,
Tuberculosis, and Malaria Reauthorization Act of 2008. information retrieved on January 15, 2009 from:
http://thomas.loc.gov/cgi-bin/query/D?c110:1:./temp/~c110mj9jUg:
5 PEPFAR technical considerations, which are sent to the countries but are not included in the COP guidance, state that
support for retention strategies, including rural retention, can be supported by PEPFAR funds.
Types of fundable health workers
An additional and important component of GHI guidance is whether it spells out which types of
workers may benefit from GHI financing at the country level. At the GHI level, PEPFAR guidance is
the most specific: it explicitly lists health care professionals (doctors, nurses, and midwives),
paraprofessionals, and community health workers (with a minimum of 6 months of training) as
direct, clinical workers, in whose pre-service training countries may wish to invest. Similarly, it
explicitly includes doctors, nurses, and community health workers as potential targeted groups for
in-service training. Finally, PEPFAR guidance targets all groups – doctors, nurses, midwives,
pharmacists, technicians, and community health workers – when it comes to achieving an
appropriate mix of providers. In contrast, GAVI, World Bank, and the Global Fund do not have this
level of specificity but our review of country applications funded by each GHI reveals that funds
have been approved to support varied types of health workers depending on country need (see
next section for more details).
Comparing approved country applications and GHI guidance
Table 4 compares each GHI guidance with pertinent individual country applications (20 in toto
were reviewed). As these tables illustrate, language found in country applications covered more
domains and sub-domains than the GHI guidance, indicating that GHIs are flexible over what they
actually fund on an individual, case-by-case basis. In the area of HRH, countries most frequently
requested funding for in-service training, particularly for direct, clinical health workers.
GAVI
In five country applications out of a total eight reviewed, GAVI funded in-service training for,
among other types of cadres, district-level professionals, teams for supportive supervision, Primary
Health Care (PHC) staff trained on Maternal and Child Health (MCH) and Primary Health (PH)
services (Armenia), health district management teams (Central African Republic), Health
Extension Workers (Ethiopia), PHC workers (Nigeria), and village health teams (Uganda).
225

Global Fund
In two country applications out of a total of three reviewed, Global Fund funded in-service training
for, among other types of cadres, Health Surveillance Assistants (HSAs) and community nurses
(Malawi), and health professionals (Rwanda).
World Bank MAP
In one country programme documentation out of a total of one reviewed, WB MAP funded inservice
training for, among other types of cadres, public sector employees in planning,
management and monitoring (Rwanda).
PEPFAR I
In seven Country Operational Plans (COP) out of a total of eight reviewed, PEPFAR funded inservice
training for, among other types of cadres, medical doctors, nurses, nurse midwives,
laboratorians, TBA, ANC providers and Ob/Gyn (Ethiopia), TBA, midwives, doctors, counselors,
laboratory staff, and pharmacists (Cambodia), health care workers, doctors, nurses, midwives,
others such as clinical officers, and public health officers (Kenya), Prevention of Mother to Child
Transmission (PMTCT) providers, MCH workers, HSAs, nurses, clinicians, lab technicians, and
midwives (Malawi), Antenatal Care (ANC)/PMTCT staff, teaching faculty, Tecnicos de Medicina, and
MCH nurses (Mozambique), health care providers, pediatricians, doctors, nurses, social workers,
HIV case managers, and nutritionists (Rwanda), and health workers to provide PMTCT (Sudan).
Discussion
Based on the review of documents and discussions with key informants, we conclude that GHI
written guidance generally exhibits flexibility for country-level funding in the area of human
resources for health. However, the guidance in some instances is very broad - so much so as to
create uncertainty regarding what kinds of interventions might be permissible. GHIs also
communicate expectations beyond the written word via internal communications, site visits,
technical fora and other mechanisms, which vary by GHI and presumably in effectiveness. One
aspect that we were not able to probe was how countries, particularly those whose country
applications were reviewed, perceived the written and oral expectations from each funder. In
addition, several key themes started to emerge, which are summarized below.
Defining health systems strengthening
The four GHIs had a varied sense of what HSS means and often referenced the WHO framework
and its six components in describing their initiatives. Consensus also emerged that it is a useful
framework to begin the discussion but that it is much less practical for countries when they seek
funding or for each GHI when they need to issue their own guidance.
226

Clarity of GHI policies/guidelines/guidance for funding
The four GHIs provide different kinds of flexibilities, and also communicate their expectations, in
ways that may not be clear to host countries. Countries may benefit from more explicit guidance
on the range of issues for which they may request funds from each GHI. The proper balance must
be found between each GHI's mission, goals and overarching operating model, and the need to
provide countries with clarity on funding options for supporting health systems. Some GHIs
stressed the importance of being more explicit with countries, arguing it did not detract from the
GHI’s basic mission and purpose.
HRH as a barrier to successful implementation of programmes
The four GHIs recognize that the lack of sufficient HRH in countries constitutes a severe limitation
to achieving their policy goal and consequently are each pursuing initiatives to scale up HRH in a
number of countries. They also pursue these initiatives independently of each other in order to
address their own objectives. These separate approaches have resulted in duplication of GHIfunded
efforts at country and local levels or in inadequate alignment with national and local
priorities for the health system. While this study focused on the problem of HRH supply,
distribution, and capacity strengthening, other issues related to HRH and systems, e.g. the need to
beef up infrastructures, were also highlighted in interviews and country applications.
The effect of the global economic crisis on GHIs
The four GHIs are reportedly committed to their mission and purpose but also acutely aware of the
current reality of the financial markets and the effect this may have on the availability and certainty
of acquiring funds. They also acknowledge the importance of continuing to emphasize HSS
alongside disease-specific approaches.
Conclusion
In conclusion, GHI support for HSS can be communicated and coordinated more effectively, while
taking into account each GHI’s specific structure and political context, the types of flexibility each
GHI grants countries seeking funds for HSS (either separately or as part of a disease-focused plan),
and the degree to which GHI-funded programmes are harmonized and aligned with national and
local priorities. What is not well known other than anecdotally is how country officials applying
for funding perceive and understand the current way GHI funding operates and how they would
respond to improved communication and coordination, particularly within their own political and
legal framework. More research needs to be conducted to answer this key question.
227

TABLE 3: Selected Countries: MPS Case Studies and GW Reviews of Country Applications
PEPFAR I
COP Summaries/
Full Submissions
GFAMT
HSS Round 5
only
GAVI
HSS
Brazil
Botswana
Burkina Faso
Burundi X
Cambodia
Cameroon
X X X
Central African Republic
China
X
Ethiopia
Georgia
Ghana
Haiti
India
X X X
Kenya
Kyrgyzstan
X X
Malawi X X X
Mozambique X X
Nicaragua X
Nigeria
Pakistan
Peru
X X
Republic of Armenia X
Rwanda
South Africa
X X X
Sudan* X
Tanzania X
Uganda
Ukraine
X X
Zambia X
TOTAL (29) 8 3 8 8
Note: All Countries are MPS Case Studies
228
WB MAP

GAVI
HRH Intervention Categories
GAVI: GHI Level Documents
DIRECT PROVIDERS INDIRECT
Clinical Non-Clinical All
EDUCATION
Pre-service education X31 X32 X33 In-service education X X X
Producing different types of health workers
General Workforce Capacity Strengthening X
FINANCIAL INCENTIVES
Salary payment
Top-ups X X X
Rural allowances X X X
Pay for performance X X X
General Recruitment Support
MANAGEMENT
X X X
General improvements in infrastructure X X X
Training of health managers
Support for continuous professional
development, career paths
X X X
Source: GWU Analysis of GHI Laws, Policies, Guidances, Spring 2009.
31 GAVI (2009). Revised GAVI HHS Guidelines for: GAVI Alliance Health System Strengthening Applications (page 8).
Information retrieved on January 15, 2009 from:
www.gavialliance.org/resources/HSS_Guidelines___2009_UPDATE___Final.pdf
32 Idem
33 Idem
229
X
X

HRH Intervention Categories
GAVI: Country Level Documents
DIRECT PROVIDERS INDIRECT
Clinical Non-Clinical All
EDUCATION
Pre-service education X34 X35 X36
In-service education X X X
Producing different types of health workers X
X
General Workforce Capacity Strengthening
FINANCIAL INCENTIVES
Salary payment X X X
Top-ups X X X
Rural allowances
Pay for performance
X X X
General Recruitment Support
MANAGEMENT
X X X
General improvements in infrastructure X X X
Training of health managers X X X
Support for continuous professional
development, career paths
X X
X
Source: GWU Analysis of GHI Country Applications and Documents, Spring 2009.
34 GAVI Alliance (2008). Application for Health System Strengthening in the Republic of Armenia. Information retrieved
on January 15, 2009 from: http://www.gavialliance.org/resources/Armenia_HSS_2008.pdf.
35 GAVI Alliance (2008). Application for Health System Strengthening in the Republic of Armenia. Information retrieved
on January 15, 2009 from: http://www.gavialliance.org/resources/Armenia_HSS_2008.pdf; GAVI Alliance (2007).
Application form for: GAVI Alliance Health System Strengthening (HSS) Republic of Uganda. Information retrieved on
January 20, 2009 from: http://www.gavialliance.org/resources/14_Uganda_HSS_Proposal.doc; GAVI Alliance (2007).
Application Form for: Nigeria. GAVI Alliance Health System Strengthening (HSS) Application. Information retrieved on
January 15, 2009.
36 GAVI Alliance (2008). Application for Health System Strengthening in the Republic of Armenia. Information retrieved
on January 15, 2009 from: http://www.gavialliance.org/resources/Armenia_HSS_2008.pdf. Nigeria. GAVI Alliance Health
System Strengthening (HSS) Application. Information retrieved on January 15, 2009.
230

GLOBAL FUND
HRH Intervention Categories
GLOBAL FUND: GHI Level Documents
DIRECT PROVIDERS INDIRECT
Clinical Non-Clinical All
EDUCATION
Pre-service education X37 X
Source: GWU Analysis of GHI Laws, Policies, Guidances, Spring 2009.
38 X39 In-service education
Producing different types of health workers
X X X
General Workforce Capacity Strengthening
FINANCIAL INCENTIVES
X X X
Salary payments
Top-ups
Rural allowances
Pay for performance
General Recruitment Support
MANAGEMENT
X X X
General improvements in infrastructure X X X
Training of health managers
Support for continuous professional
development, career paths
X X X
37 Healthy Systems 2020 (2008). Toolkit for Using Round 9 of the Global Fund Health Systems Strengthening.
www.healthsystems2020.org/files/2122_file_GF_HSS_Toolkit_FIN.pdf
38 Idem
39 Idem
231

HRH Intervention Categories
GLOBAL FUND: Country-level Documents
DIRECT PROVIDERS 40 INDIRECT 10
Clinical10 Non-Clinical10 All
EDUCATION
Pre-service education (*) X41 X42 X43 In-service education X X X
Producing different types of health workers X X X
General Workforce Capacity Strengthening
FINANCIAL INCENTIVES
X X X
Salary payments X X X
Top-ups X X X
Rural allowances X X
Pay for performance X X X
General Recruitment Support
MANAGEMENT
X X X
General improvements in infrastructure X X X
Training of health managers X X X
Support for continuous professional
development, career paths
X
X
X
WORLD BANK
HRH Intervention Categories
WORLD BANK: GHI Level Documents
DIRECT PROVIDERS INDIRECT
Clinical Non-Clinical All
EDUCATION
Pre-service education X44 X
Source: GWU Analysis of GHI Laws, Policies, Guidances, Spring 2009.
45 X46 In-service education
Producing different types of health workers
X X X
General Workforce Capacity Strengthening
FINANCIAL INCENTIVES
X X X
Salary payment X X X
Top-ups
Rural allowances
X X X
Pay for performance X X X
General Recruitment Support
MANAGEMENT
X X X
General infrastructure improvements X X X
Training of health managers X X X
Support for continuous professional
development, career paths
X X X
40 Language from Malawi, Rwanda HSS Round 5 Applications
41 The Global Fund (2005). Health Systems Strengthening in Malawi. Information retrieved on January 25, 2009 from:
http://www.theglobalfund.org/grantdocuments/5MLWH_1142_0_full.pdf . The Global Fund (2005). Assuring access to quality
care: The missing link to combat AIDS, Tuberculosis and Malaria in Rwanda. Information retrieved on February 2, 2009 from:
http://www.theglobalfund.org/grantdocuments/5RWNH_1199_0_full.pdf
42 Idem
43 Idem
44 The World Bank (2008). The World Bank’s Commitment to HIV/AIDS in Africa: Our Agenda for Action, 2007-2011. http://wwwwds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2008/05/30/000333038_20080530023715/Rendered/PDF/43
9600PUB0Box310only109780821374481.pdf
45 Idem
46 Idem
232

HRH Intervention Categories
WORLD BANK: Country Level Documents
DIRECT PROVIDERS INDIRECT
Clinical Non-Clinical All
EDUCATION
Pre-service education X47 X48 X49 In-service education X X X
Producing different types of health workers X
General Workforce Capacity Strengthening
FINANCIAL INCENTIVES
Salary payment
Top-ups
Rural allowances
X X X
Pay for performance X X X
General Recruitment Support
MANAGEMENT
X X X
General infrastructure improvements X X X
Training of health managers X X X
Support for continuous professional
development, career paths
X
X
X
Source: GWU Analysis of GHI Country Applications and Documents, Spring 2009.
PEPFAR
HRH Intervention Categories
PEPFAR II: Reauthorization Legislation
and Supplemental Report
DIRECT PROVIDERS INDIRECT
Clinical Non-Clinical All
EDUCATION
Pre-service education X50 X
In-service education X X
Producing different types of health workers X
General Workforce Capacity Strengthening
FINANCIAL INCENTIVES
X X X
Salary payment
Top-ups
X X
Rural allowances (X) 51
Pay for performance
General Recruitment Support
MANAGEMENT
General infrastructure improvements X X X
Training of health managers
Support for continuous professional
development, career paths
X X X
Source: GWU Analysis of GHI Laws, Policies, Guidances, Spring 2009.
47 The World Bank (2007). The Africa Multi-Country AIDS Program 2000-2006. Results of the World Bank’s Response to a
Development Crisis. http://siteresources.worldbank.org/EXTAFRREGTOPHIVAIDS/Resources/717147-
1181768523896/complete.pdf
48 Idem
49 Idem
50 The Library of Congress (2008). Tom Lantos and Henry J. Hyde United States Global Leadership Against HIV/AIDS,
Tuberculosis, and Malaria Reauthorization Act of 2008. Information retrieved on January 15, 2009 from:
http://thomas.loc.gov/cgi-bin/query/D?c110:1:./temp/~c110mj9jUg
51 PEPFAR technical considerations, which are sent to the countries but are not included in the COP guidance, state that
support for retention strategies, including rural retention, can be supported by PEPFAR funds.
233

HRH Intervention Categories
PEPFAR I: Country Level Documents
DIRECT PROVIDERS INDIRECT
Clinical Non-Clinical All
EDUCATION
Pre-service education X52 X53 X54 In-service education X X X
Producing different types of health workers X X X
General Workforce Capacity Strengthening
FINANCIAL INCENTIVES
Salary payment
Top-ups
X X X
Rural allowances X
Pay for performance X X X
General Recruitment Support
MANAGEMENT
X X X
General infrastructure improvements X X X
Training of health managers X X
Support for continuous professional
development, career paths
X
Source: GWU Analysis of GHI Country Applications and Documents, Spring 2009.
52 The United States President’s Emergency Plan for AIDS Relief (2008). Ethiopia Fiscal Year 2008 Country Operational
Plan. Information retrieved on April 22, 2009 from: http://www.pepfar.gov/about/opplan08/102020.htm Mozambique
Fiscal Year 2008 Country Operational Plan. Information retrieved on April 22, 2009 from:
http://www.pepfar.gov/about/opplan08/102016.htm http://www.pepfar.gov/documents/organization/113012.pdf
Kenya Fiscal Year 2008 Country Operational Plan. Information retrieved on April 22, 2009 from:
http://www.pepfar.gov/documents/organization/113012.pdf
53 The United States President’s Emergency Plan for AIDS Relief (2008). Cambodia Fiscal Year 2008 Country Operational
Plan. Information retrieved on April 22, 2009 from: http://www.pepfar.gov/documents/organization/112995.pdf
54 The United States President’s Emergency Plan for AIDS Relief (2008). Ethiopia Fiscal Year 2008 Country Operational
Plan. Information retrieved on April 22, 2009 from: http://www.pepfar.gov/about/opplan08/102020.htm
234

References
* GWU Departments of Health Policy and Epidemiology & Biostatistics
[1] World Bank, IDA at work: HIV/AIDS – Supporting Effective Prevention, Treatment and Care,
June 2008.
[2] GAVI (http://www.gavialliance.org/support/who/index.php; accessed 19 February 2009).
[3] Center for Global Development, Following the Funding for HIV/AIDS – A Comparative Analysis
of the Funding Practices of PEPFAR, the Global Fund and World Bank MAP in Mozambique,
Uganda and Zambia October 2007
[4] The Global Fund (http://www.theglobalfund.org/en/how/?lang=en; accessed 16 March 2009).
[5] The Global Fund to Fight AIDS, Tuberculosis and Malaria Terms of Reference of the Technical
Review Panel. As amended at the Seventeenth Board Meeting, 28-29 April 2008 (GF/B17/DP5)
[6] The Global Fund (http://theglobalfund.org/en/eligibility/?lang=en; accessed 19 February
2009).
[7] The World Bank (www.worldbank.org; accessed 19 February 2009).
[8] World Bank, IDA (International Development Association) at work: HIV/AIDS – Supporting
Effective Prevention, Treatment and Care, June 2008.
[9] The U.S. President’s Emergency Plan for AIDS Relief (http://www.pepfar.gov/press/80064.htm;
accessed 19 February 2009)
[10] HR 5501, Pub L 110-293
[11] Office of the US Global AIDS Coordinator (OGAC). The President’s Emergency Plan for AIDS
Relief General Policy Guidance for All Bilateral Programs [FINAL] October 2005
[12] Center for Global Development. (2008). Seizing the Opportunity on AIDS and Health Systems
(http://www.cgdev.org/content/publications/detail/16459; accessed 19 February 2009).
235

Conclusion: Directions for future research
on positive synergies
Research on positive synergies to date is part of a broader effort that is expanding the tools
available for health systems analysis [1-5]. Today, a growing number of investigators are placing
practical health systems policy and implementation problems at the centre of their research, then
selecting and combining research methods and data sources to produce evidence that can guide
real-world action [6,7]. Their work is feeding strategically into change processes by which new
policies are proposed, approved, implemented and evaluated [8-10]. Key to this focus on
informing practical action is the understanding that both quantitative and qualitative data are
required to build policy-relevant knowledge on the factors that influence health systems
performance and health outcomes [4,11-13]. As they look to provide evidence that policymakers
and implementers can apply to concrete problems, researchers also increasingly recognize the
need to situate health systems challenges and solutions within countries’ specific environmental,
epidemiological, economic and political contexts—acknowledging how context shapes options
for change [5,14].
As these approaches come together, a new multidisciplinary field of global health systems
research is emerging. Though it builds on intellectual traditions reaching back to the dawn of
modern public health [15], this multidisciplinary field is still in its early stages. But it is acquiring
new rigour today, as well as new methodological breadth. It marshals learning strategies from
many domains—from epidemiology and clinical medicine to the social sciences, law, political
economy, systems engineering, and management sciences—to analyse health delivery systems as
complex social systems [3]. As it develops, this emergent field will produce distinctive forms of
evidence that will equip policymakers and implementers to make better management decisions;
direct health resources where they can do the most good; accelerate delivery of new technologies
and clinical innovations in resource-constrained settings; improve health outcomes; and
strengthen equity [16]. Maximizing Positive Synergies (MPS) research to date is situated within this
long-term agenda. MPS has contributed to mapping opportunities and challenges for
multidisciplinary health systems research; honing methodological questions; testing the field’s
limits; and confirming its strengths.
The initial phase of country-level research on positive synergies has produced a rich body of data,
reflected in the case studies that make up this report. MPS case studies have brought
understanding of GHI-health systems interactions to a new level of breadth, detail and contextual
specificity. As with any research, however, there are limitations to the work that has been done in
this first phase. In particular, the case studies focus mainly on the one-directional effect of GHIs on
health systems, and provide little information on how specific health system attributes have
affected GHIs’ ability to achieve their objectives in improving health outcomes. In addition, the
quantitative component of the mixed-methods case research was limited by the lack of available
data on health care processes and outputs at the facility level. Finally, as the research is
observational and retrospective, researchers are appropriately cautious in attributing observed
effects to GHI action per se. In the epidemiological, political and social contexts under study,
complex causal interactions and “feedback” patterns exist, which make precise attribution difficult
[3].
236

Bearing these limitations in mind, the studies carried out to date chart promising directions for
future research to uncover the causal mechanisms that lie behind both positive and negative GHIhealth
systems interactions. Looking across the case studies compiled here, five key directions
emerge for the next phase of research on positive synergies:
• Build consensus on metrics and methods for evaluating GHI-health systems
interactions: The emergent field of health systems research needs to progressively foster
consensus around a set of metrics for systems performance, as well as common metrics
and methods for assessing GHI-health systems interactions. MPS research partners have
advanced these discussions and the consensus-building process through their joint work.
By definition, the task reaches beyond the confines of any single consortium or project.
Moving forward, however, MPS research partners can continue to play a catalytic role.
Subsequent phases of MPS work could provide an appropriate platform for shaping the
needed consensus by convening an expanding network of multidisciplinary researchers,
GHI leaders, programme implementers, system managers and monitoring and evaluation
(M&E) experts.
• Move from description to explanation: MPS country case studies have achieved a new
level of descriptive detail in illustrating how GHIs and health systems are interacting in
country contexts. This wealth of data represents an advance in knowledge. In some
instances, MPS studies have already provided evidence that can enable the framing of
initial explanatory hypotheses about the positive or negative interactions that have been
observed. The next phase of research will involve waves of more tightly focused, multimethods
studies in countries that will successively: (1) “drill down” on specific areas (e.g.,
supply chain, health workforce) to generate additional data that may be needed to
generate explanatory hypotheses; (2) formulate and test such hypotheses on causal
relationships—tackling the critical “why?” questions. A new generation of
analytic/explanatory studies on positive synergies can assess evolving national
government policies, along with the internal structures, policies and management styles of
the different GHIs. The evidence obtained will inform the ongoing learning process that
has characterized major GHIs.
• Extend cross-country comparisons and document effective practices: In addition to
drilling down on causal questions within particular country contexts, the next phase of
research on positive synergies will capitalize on the large body of data already generated,
along with results from additional studies, to conduct cross-country analyses that can
compare policy and implementation strategies and results across countries. Comparative
analysis will enable researchers to build an inventory of the policy approaches countries
have adopted to manage GHI-health systems interactions, and then to begin the task of
systematically correlating policy and implementation options with outcomes measures.
Comparison of country experiences and results poses methodological challenges,
particularly for qualitative research. MPS researchers are beginning to address these
challenges as they analyse and compare the data collected during the first phase of work.
Comparative studies, even among relatively small groups of countries, will help identify
effective strategies with wider applicability and also point to key contextual factors in
countries that may facilitate or constrain synergies between GHIs and health systems.
Outputs of this phase of research might include an evolving inventory of best practices, as
well as a more systematic account of the contextual factors that influence positive or
negative GHI interactions with health systems.
• Strengthen civil society and community participation in research: The MPS process has
confirmed the importance of civil society participation in all aspects of research related to
237

GHIs and health systems. Civil society and community members have been engaged in the
majority of MPS country case studies: as key respondents in studies conducted by
academic partners, and as researchers. The role of civil society organizations (CSOs) as
investigators and producers of new knowledge is especially critical. Future research on
positive synergies must continue and extend this pattern. Multidisciplinary health systems
research and the monitoring and evaluation of GHI-supported programmes provide
opportunities to strengthen the knowledge-generation and analytic capacities of
community systems, and thus their ability to positively influence policy and its
implementation.
• Activate knowledge through communities of practice: Future efforts to build the
evidence base on GHI-health systems interactions can incorporate new ways of mobilizing
and disseminating knowledge through communities of practice. Emerging
communications technologies support virtual communities that can share experiences
and engage in joint problem-solving in real time. These capacities are only beginning to be
harnessed for strengthening health delivery in resource-constrained environments, but
early results are promising. Better health systems research requires innovative tools and
strategies for virtual community-building and knowledge-sharing. Knowledge networks of
practitioners will be the key relay to operationalize lessons about positive synergies in
service delivery on the frontlines. Similarly, building communities of practice connecting
researchers, programme managers and local implementers will be key in meeting another
challenge highlighted in MPS studies: the need to engage frontline health workers more
effectively in the collection and use of data, and to ensure that data routinely collected at
local health facilities are applied to improve performance and outcomes in those local
settings.
238

References
[1] Bennett S, Adam T, Zarowsky C, Tangcharoensathien V, Ranson K, Evans T, et al. From Mexico to Mali:
progress in health policy and systems research. Lancet, 2008, 372:1571-8.
[2] Evans T, Nishtar S, Atun R, Etienne C. Scaling up research and learning for health systems: time to act.
Lancet, 2008, 372:1529-30.
[3] Atun R, Menabde N. Health systems and systems thinking. In: Coker R, Atun R, McKee M, editors. Health
systems and the challenge of communicable diseases: experiences from Europe and Latin America.
Maidenhead, Open University Press, 2008, p. 121-40.
[4] Atun RA, Lennox-Chhugani N, Drobniewski F, Samyshkin YA, Coker RJ. A framework and toolkit for
capturing the communicable disease programmes within health systems: tuberculosis control as an
illustrative example. European Journal of Public Health, 2004, Sep;14(3):267-73.
[5] Atun RA, McKee M, Drobniewski F, Coker R. Analysis of how the health systems context shapes responses
to the control of human immunodeficiency virus: case-studies from the Russian Federation. Bulletin of the
World Health Organization, 2005, Oct;83(10):730-8.
[6] Banteyerga H, Kidanu A, Stillman K. The Systemwide Effects of the Global Fund in Ethiopia: Final Study
Report. Bethesda, MD, The Partners for Health Reformplus Project, Abt Associates Inc., 2006.
[7] Bennett S, Fairbank A. A Conceptual Framework for Analyzing the System-wide Effects of the Global
Fund. Bethesda, MD, Partners for Health Reformplus, Abt Associates Inc., 2003.
[8] Bennett S, Boerma JT, Brugha R. Scaling up HIV/AIDS evaluation. Lancet, 2006, Jan 7;367(9504):79-82.
[9] Brugha R. Evaluation of HIV programmes. British Medical Journal, 2007, Jun 2;334(7604):1123-4.
[10] World Health Organization. Leadership and effective government. In: World Health Report 2008: Primary
health care: Now more than ever. Geneva, WHO, 2008, p. 81-96.
[11] O'Cathain A, Murphy E, Nicholl J. Multidisciplinary, interdisciplinary, or dysfunctional? Team working in
mixed-methods research. Qualitative Health Research, 2008, Nov;18(11):1574-85.
[12] O'Cathain A, Murphy E, Nicholl J. Why, and how, mixed methods research is undertaken in health
services research in England: a mixed methods study. BMC Health Services Research, 2007, 7:85.
[13] O'Cathain A, Thomas K. Combining qualitative and quantitative methods. In: Pope C, Mays N, editors.
Qualitative Research in Health Care. Oxford, Blackwell, 2006. p. 102-11.
[14] Atun R, Ohiri K, Adeyi O. Integration of Health Systems and Priority Health, Nutrition and Population
Interventions. Washington, DC, World Bank, 2008.
[15] Virchow R. Collected Essays on Public Health and Epidemiology. Canton, MA, Science Hill Publishers,
1985.
[16] McIntyre D, Mooney G, editors. The economics of health equity. Cambridge, Cambridge University Press,
2007.
239

Annex 01: Data sources for tables
in country case studies
In most instances, the country case studies included in this volume feature standardized
background data tables whose purpose is to provide readers with an overview of key
socioeconomic, demographic and health indicators for the countries where Maximizing Positive
Synergies (MPS) research has been undertaken, along with summary information on the
magnitude and evolution of major GHIs’ investments in these countries. Except where otherwise
noted, the data included in the standardized country tables are drawn from the following sources:
Basic socioeconomic, demographic and health indicators:
Population; geographic size; GDP per capita; Gini index:
World Bank. World Development Indicators Online. Proprietary online database. For conditions of access,
see: http://go.worldbank.org/6HAYAHG8H0.
Government expenditure on health as percentage of general government expenditure; per capita
government expenditure on health:
WHO. National Health Accounts (NHA). Available at: http://www.who.int/nha/country/en/.
Physician density; nursing and midwifery density; maternal mortality ratio; DTP3 coverage:
WHO. WHO Statistical Information System (WHOSIS). Available at: http://www.who.int/whosis/en/.
Estimated adult HIV prevalence:
UNAIDS. Report on the global AIDS epidemic 2008. Geneva, UNAIDS, 2008.
Estimated antiretroviral therapy coverage:
WHO, UNAIDS, UNICEF. Towards universal access: Scaling up priority HIV/AIDS interventions in the health
sector: Progress report 2008. Geneva, WHO, 2008.
Note: Estimated antiretroviral therapy coverage data for a small number of countries were not included in
the 2008 Towards universal access report. In such instances, coverage estimates have been derived from
HIV/AIDS progress reports submitted by countries to the United Nations General Assembly. Periodic national
progress reports are part of countries’ obligations under the UN Declaration of Commitment on HIV/AIDS,
signed at the 2001 UN General Assembly Special Session on HIV/AIDS (UNGASS). Countries’ 2008 UNGASS
progress reports are available at:
http://www.unaids.org/en/KnowledgeCentre/HIVData/CountryProgress/2007CountryProgressAllCountries.a
sp.
Estimated tuberculosis prevalence:
WHO. Global tuberculosis database. Available at:
http://www.who.int/tb/country/global_tb_database/en/index2.html.
Estimated malaria deaths:
WHO. World malaria report 2008. Geneva, WHO, 2008.
240

GHI investments:
GAVI: GAVI Alliance. Country information database. Available at:
http://www.gavialliance.org/performance/country_results/.
Global Fund: Global Fund. Grant portfolio. Data by country available at:
http://www.theglobalfund.org/en/portfolio/?lang=en.
PEPFAR: (1) PEPFAR. Fiscal Year 2006 Operational Plan. Washington, DC, PEPFAR, 2005. Available at:
http://www.pepfar.gov/about/81929.htm. (2) PEPFAR. Fiscal Year 2008 PEPFAR Operational Plan.
Washington, DC, PEPFAR, 2007, available at: http://www.pepfar.gov/documents/organization/107838.pdf
World Bank MAP: World Bank. HIV/AIDS in Africa/ACT Africa, projects and programs. Online database.
Available at: http://go.worldbank.org/ZC8O2E5ZR0
241

Annex 02: Contributing Institutions
ANRS, French Agency for AIDS Research
Beijing Normal University, School of Social Development and Public Policy, China
Brigham and Women's Hospital, Division of Global Health Equity, Boston, USA
Cayetano Heredia University School of Public Health, Health, Sexuality and Human Development Unit,
Peru
ORS-PACA , Centre for Disease Control of South-Eastern France
Center for Health System Development, Kyrgyzstan
Centre for Economics (CEPN), University of Paris, France
Curatio International Foundation, Georgia
Fann's Teaching Hospital, Infectious Diseases Department, Dakar, Senegal
George Washington University, Department of Health Policy and Department of Epidemiology &
Biostatistics, USA
Global Fund to Fight AIDS, TB, and Malaria, Geneva, Switzerland
Global Health Delivery Project, Boston, USA
Harvard Medical School, Department of Global Health and Social Medicine, Boston, USA
Harvard School of Public Health, François-Xavier Bagnoud Center for Health and Human Rights, Boston,
USA
Heartfile, Islamabad, Pakistan
Imperial College London, UK
INSERM, UMR 707, Pierre & Marie Curie University, Paris, France
INSERM/IRD/Aix-Marseille Universities Research Unit 912, Economic & Social Sciences, Health Systems &
Societies, Marseille, France
Institute of Economic and Social Research, University of Zambia
Institute of Tropical Medicine, Antwerp, Belgium
IRSA, Catholic University of Central African States, Yaoundé, Cameroon
Kenyatta University, Nairobi, Kenya
London School of Hygiene and Tropical Medicine, UK
Ministry of Health, Dakar, Senegal
242

Ministry of Public Health, Division of Health Operations Research, Yaoundé, Cameroon
National Center for HIV Reference, Bujumbura, Burundi
Partners In Health, Boston, USA
Paul Ango Ela Fondation for Promotion of Geopolitics in Central Africa (FPAE) & University of
Yaoundé, Cameroon
Public Health Foundation of India, New Delhi, India
Royal College of Surgeons, Ireland
School of Public Health and School of Social Work, Kyiv-Mohyla Academy, Kyiv, Ukraine
The AIDS Support Organization (TASO), Uganda
University of Buea, Faculty of Health Sciences, Cameroon
University of Burundi, Bujumbura, Burundi
University of Malawi, College of Medicine and Centre for Social Research
University of Pretoria, South Africa
University of Western Cape, School of Public Health, South Africa
University of Yaounde, Centre for the Study and Control of Communicable Diseases and Faculty of
Medicine & Biomedical Sciences, Cameroon
Zanmi Lasante, Cange, Haiti
243