This material is the copyright of the original publisher. Unauthorised ...

This material is the copyright of the original publisher. Unauthorised ...

This material is the copyright of the original publisher.

Unauthorised copying and distribution is prohibited.

2008, Vol. 22, No. 1 (pp. 73-81)

ISSN: 1172-7047

Original Research Article

Modelling the Budgetary Impact and Cost Effectiveness of Alteplase via Telemedicine

Terms and Conditions for Use of PDF

The provision of PDFs for authors' personal use is subject to the following Terms & Conditions:

The PDF provided is protected by copyright. All rights not specifically granted in these Terms & Conditions are expressly

reserved. Printing and storage is for scholarly research and educational and personal use. Any copyright or other notices

or disclaimers must not be removed, obscured or modified. The PDF may not be posted on an open-access website

(including personal and university sites).

The PDF may be used as follows:

• to make copies of the article for your own personal use, including for your own classroom teaching use (this includes

posting on a closed website for exclusive use by course students);

• to make copies and distribute copies (including through e-mail) of the article to research colleagues, for the personal use

by such colleagues (but not commercially or systematically, e.g. via an e-mail list or list serve);

• to present the article at a meeting or conference and to distribute copies of such paper or article to the delegates

attending the meeting;

• to include the article in full or in part in a thesis or dissertation (provided that this is not to be published commercially).

CNS Drugs 2008; 22 (1): 73-81

ORIGINAL RESEARCH ARTICLE 1172-7047/08/0001-0073/$48.00/0

© 2008 Adis Data Information BV. All rights reserved.

National Use of Thrombolysis with

Alteplase for Acute Ischaemic Stroke

via Telemedicine in Denmark

A Model of Budgetary Impact and Cost Effectiveness

Lars Ehlers, 1 Wilhelmina Maria Müskens, 2 Lotte Groth Jensen, 1 Mette Kjølby 1 and

Grethe Andersen 3

This material is

1 HTA Unit, Aarhus University Hospital, Aarhus, Denmark

2 Institute of Economics, Aarhus University, Aarhus, Denmark

3 Department of Neurology, Aarhus University Hospital, Aarhus, Denmark

the copyright of the

Aim: The purpose of this analysis was to assess the budgetary impact and cost

effectiveness of the national use of thrombolysis with alteplase (recombinant

original publisher.

tissue plasminogen activator; rt-PA) for acute ischaemic stroke via telemedicine


in Denmark.

Methods: Computations were based on a Danish health economic model of

thrombolysis treatment of acute ischaemic stroke via telemedicine. Cost data for

stroke units and satellite clinics were taken from the first practical experiences in

Denmark with implementing thrombolysis via telemedical linkage to the Stroke

Unauthorised copying

Department at Aarhus University Hospital. Effectiveness data were taken from a

published pooled analysis of results from randomized controlled trials of


Results: The calculations showed that the additional total costs to the hospitals of

and distribution

implementing thrombolysis with alteplase for acute ischaemic stroke via

telemedicine were approximately $US3.0 (range 2.0–5.8) million per year in the

case of five centres and five satellite clinics, or $US3.6 (range 2.4–7.0) million per

year based on seven centres and seven satellite clinics. The incremental cost-

is prohibited.

effectiveness ratio was calculated to be approximately $US50 000 when taking a

short time perspective (1 year), but thrombolysis was dominant (both cheaper and

more effective) after as little as 2 years and cost effectiveness improved over

longer time scales.

Conclusion: The budgetary impact of using thrombolysis with alteplase for acute

ischaemic stroke via telemedicine depends on the existing capacity and organizational

conditions at the local hospitals. The health economic model computations

suggest that the macroeconomic costs may balance with savings in care and

rehabilitation after as little as 2 years, and that potentially large long-term savings

74 Ehlers et al.

are associated with thrombolysis with alteplase delivered by telemedicine,

although the long-term calculations are uncertain.


to assess the cost effectiveness of thrombolysis for

Intravenous thrombolysis with alteplase within

3 hours of symptom onset is more effective [1] and

cost effective [2-7] than conservative treatment for

acute ischaemic stroke. No countries have, however,

acute ischaemic stroke in Denmark. [7] In the current

analysis, the model was extended to include thrombolysis

at satellite clinics linked to a larger throm-

bolysis centre. All unit costs and quantities were

updated to mirror the expected price level for the

implemented the widespread use of thrombolysis

This material is

The model was set up as a choice between two

the copyright of the

because such treatment places large demands on the

national healthcare system. [8,9]

Pilot studies using telemedicine to assist with the

use of thrombolysis for stroke (‘telestroke’) in Germany,

among other countries, have shown promis-

ing results, and telemedicine could be an opportuni-

ty to overcome barriers related to the limited time

window for thrombolysis and the lack of medical

specialists familiar with its use at local hospitals.

[10-12] In Denmark, intravenous thrombolytic

treatment with alteplase for acute ischaemic stroke

is currently available at five regional stroke centres.

year 2007 and converted into US dollars (exchange

rate $US100 = DKK 569.64). All quantities (includ-

ing mortality tables) were updated accordingly.

alternatives (see figure 1). The patient could receive

either thrombolysis with alteplase or conservative

treatment. Depending on the treatment instituted,

the patient could be exposed to either a larger or

original publisher.

smaller risk of intracranial haemorrhage in connec-

tion with the treatment. After the initial treatment at

the hospital, the patient could be classified accord-

At present,

Modelling the Budgetary Impact and Cost Effectiveness of Alteplase via Telemedicine 75

Intracranial haemorrhage


Intravenous thrombolysis with alteplase



No bleeding


Acute ischaemic stroke



This material is

Conservative treatment

Intracranial haemorrhage

the copyright

No bleeding

of the

Fig. 1. Markow decision analytic model for intravenous thrombolysis with alteplase for acute ischaemic stroke (reproduced from Ehlers et

al., [7] with permission).

avoided, as thrombolysis does not result in changed

with alteplase was associated with approximately 13 mortality. [1]

of 100 stroke patients avoiding disability, measured The risk of major intracranial haemorrhage is

5.9% with thrombolysis and 1.1% with conservative

otherwise have sustained. Since pilot studies with

treatment. [1] The risk of minor haemorrhage was not

telemedicine indicate that the same good effect for

included in the model, which was not the case in the

thrombolysis with alteplase is obtainable through

other international health economic studies, [2-7] pri-

marily because it was assumed that this type of

both thrombolysis centres and satellite clinics linked

to a centre via telecommunication.

haemorrhage is not associated with any significant

costs or any loss of QALY.

The effect measure in the model was the number

All patients were assumed to be 68 years of age at

patient received thrombolysis in place of conservaafter

the time of the index stroke. The overall death rate

tive treatment. Each of the seven different functional

the first year was assumed to exceed that of the

states (R0–R5 and death) were thus translated into average population by approximately 2.5-fold. The

QALYs. In the absence of Danish preference data risk of stroke recurrence is 5.2% per year, [2-7] which

within this field, we used an American study [13] in the model was assumed to be independent of the

previously used in other health economic studies. [2,3] result of the initial treatment with thrombolysis or

The model did not calculate the number of deaths the conservative treatment.

tilyse ® ) 1 . [1] original

The analysis showed that thrombolysis


by an MRS score of R2–R4, which they would

Unauthorised copying

telemedicine, [10-12] the same effect was assumed for

and distribution

of quality-adjusted life-years (QALY) gained if the

is prohibited.


1 The use of trade names is for product identification purposes only and does not imply endorsement.

© 2008 Adis Data Information BV. All rights reserved. CNS Drugs 2008; 22 (1)

76 Ehlers et al.

Organization Model

Calculation of the total extra costs incurred by the

hospital services in Denmark if nationwide throm-

Denmark has a population of 5 million, covered bolysis was implemented was based on two organiby

five regions of equal size. There is a thrombolysis zational models – model A with five centres and five

centre in each region and, in each region, a satellite satellite clinics, and model B with seven centres and

clinic will be connected to the regional centre in seven satellite clinics. Model A was chosen since

order to reduce patient transport time to

Modelling the Budgetary Impact and Cost Effectiveness of Alteplase via Telemedicine 77

telemedical treatment. The implementation costs of for each patient treated with thrombolysis instead of

telecommunication equipment cover the installasumed

conservative treatment. [5,9] In the model, it was as-

tion, software and hardware at the centre and the

that all centres (but not satellite clinics)

satellite clinics, and were depreciated on a straightservice

would be upgraded to provide a full 24-hour medical

line basis over 6 years. The annual operating costs

at the Department of Neurology. Thus, the

cover a service agreement, IT staff and leased lines. existing neurology coverage would be expanded to

Additional costs for ambulance transportation

cover evenings and nights. Cost figures from Aarhus

occur for three different reasons. First, the patient

University Hospital were used as a proxy for all

must be transported to a stroke department where

centres in the model.

thrombolysis treatment is performed instead of beby

CT scanning on arrival and by control scanning

It was assumed that the patients are examined

patients are thus transported by ambulance for a

24 hours after treatment was initiated. CT is used for

diagnosing and selecting patients eligible for thromlonger

period of time. Second, all patients with a

bolysis at the vast majority of thrombolysis cenpossible

indication of thrombolysis must be transtres.

[14] It was assumed that the necessary scanning

capacity would be available so that no further inthe

local hospital. In most cases, however, contrainvestments

in scanning capacity were required. The

dications to the treatment will appear, and only

costs of CT scans were estimated from average unit

about one-quarter of patients can expect to undergo

prices, using the Danish DRG-casemix system for

eligible for thrombolysis will be transported back to

ambulatory care.

their local hospital where treatment will be completed.

Cost-Effectiveness Analysis

ing transported to the nearest hospital. On average,

This material is

ported to the relevant stroke department instead of

the copyright of the

thrombolysis. [7] Thirdly, some of the patients not

original publisher.

The national use of thrombolysis will also require

Unauthorised copying

consultant time previously calculated to be required

and distribution

In the calculations of the cost effectiveness of

telemedical treatment with thrombolysis, the per-

spective was expanded from the hospital services to

embrace all the macroeconomic costs and savings

associated with introducing the national use of

24-hour neurology coverage at the thrombolysis

centres. This will be associated with physician costs

that are beyond the approximately 4 extra hours of

Table I. Sensitivity analysis of the variable extra costs ($US) per patient receiving thrombolysis compared with conservative treatment

Worst case Base case Best case References

Alteplase (Actilyse ® ) 1288 902 644 1,7

Other interventions (blood tests, ECG, etc.) 391 313 235 7,15

is prohibited.

Consultant time 330 264 132 5,7,9

Nursing time 299 266 133 5,7,9

Transportation costs (per alteplase-treated patient) 263 211 158


More patients transported for diagnosis 737 590 442


More patients transported back to own hospital 494 393 296


CT scanning 189 172 155 1,7

Costs associated with intracranial haemorrhage 189 151 113 7

Fewer bed days 0 –675 –898 1,2,7

Total 4180 2587 1410

a Data from the Department of Neurology, Aarhus University Hospital, Aarhus, Denmark.

© 2008 Adis Data Information BV. All rights reserved. CNS Drugs 2008; 22 (1)

78 Ehlers et al.

Table II. Sensitivity analysis of the extra fixed costs ($US) per year per thrombolysis centre with an affiliated satellite clinic

Worst case Base case Best case References

Duty at the Department of Neurology 391 623 313 298 234 973


Extra nurses on duty b 147 105 NA NA


Telemedical linkage 14 263 11 411 8 558 10

Training of health professionals 8 993 4 497 3 373 9,10

Total 561 984 329 206 246 904



Data from the Department of Neurology, Aarhus University Hospital, Aarhus, Denmark.

In the worst case, employment of the extra nurses on duty (in addition to the 8 hours of extra nursing time per patient) are included.

The cost is calculated as one extra emergency bed per thrombolysis centre/satellite clinic with 2.3 nurses per emergency bed and

part-time costs for a secretary and a social worker. Extra nurses on duty have been included in the worst case because reallocation

of capacity and resources in the short-term may be difficult in practice.

thrombolysis in Denmark. This means that the costs

and savings to both the hospital system and the

social services were included.

The total costs after initial hospitalization were

taken from a 1-year follow-up study of 588 stroke

tation, outpatient visits, general practitioner visits,

medicine, etc., as well as social services such as

NA = not applicable.

This material is

The results of the computations are shown in

the copyright of the

patients at Hvidovre Hospital [15] and covered health- for acute ischaemic stroke can be seen in table I.

original publisher.

servative treatment, the cost of the Danish DRG-

Unauthorised copying

nursing homes, residential homes, day centres/day

nurseries, aids and appliances, meal deliveries,

transport services, etc. In the model, it was assumed

that the annual social service costs remained at a


tables I, II, III and IV.

The extra variable costs per patient receiving

thrombolysis as opposed to conservative treatment

care services such as further admission for rehabili-

Table II shows the extra fixed costs per year per

thrombolysis centre with an affiliated satellite clinic.

As an estimate of the total average costs of con-

constant level for


each of the subsequent


years (at a

approximately one-third of the level for the first

is prohibited.

level that equalled the costs of the first year), while

the annual healthcare costs dropped to a level of

year. [7,9]

Incremental (marginal) cost-effectiveness ratios

(ICERs) were calculated as the additional costs upon

thrombolysis per QALY gained with a time horizon

of 1, 2 and 30 years. All costs and health consequences

in the long-term calculations were discounted

at a rate of 5%, to reflect the present value

to society.

Table III. Sensitivity analysis of the total costs ($US) per year for

model A (five centres and five satellite clinics) and model B (seven

centres and seven satellite clinics)

Model A

Worst case Base case Best case

Variable costs 2 089 910 1 293 690 705 065

Fixed costs 3 661 732 1 725 566 1 294 174

Total a 5 751 642 3 019 256 1 999 239

Model B

Variable costs 1 903 389 1 144 473 593 153

Fixed costs 5 126 424 2 415 792 1 811 844

Total b 7 029 813 3 560 265 2 404 997

a The base case has been calculated on an assumption of 500

patients per year (the total amounts to $US2 501 780 for 300

patients, $US3 536 731 for 700 patients and $US4 312 945

for 1000 patients).

b The base case has been calculated on an assumption of 500

patients per year (the total amounts to $US3 102 476 for 300

patients, $US4 018 054 for 700 patients and $US4 704 738

for 1000 patients).

© 2008 Adis Data Information BV. All rights reserved. CNS Drugs 2008; 22 (1)

Modelling the Budgetary Impact and Cost Effectiveness of Alteplase via Telemedicine 79

Table IV. Incremental cost-effectiveness ratio (ICER) a

Time horizon Costs ($US) QALY Change ICER

alteplase conservative alteplase conservative QALY $US $US per QALY



Expected value (1 year) 30 021 27 014 0.44 0.38 0.06 3006 50 100

Expected value (2 years) 39 854 40 175 0.83 0.71 0.12 –321 Dominance

Expected value (30 years) 109 131 129 019 3.07 2.64 0.43 –19 888 Dominance


The costs and QALY are calculated cumulatively on the assumption of 500 patients annually.

QALY = quality-adjusted life-years.

Our calculations indicate that increasing the na-

tional use of thrombolysis with alteplase for acute

ischaemic stroke via telemedicine is cost effec-

tive. [2-7] Also, the budgetary costs of establishing

new satellites seem fair. However, there are a number

of uncertainties that must be addressed. We do

not have the empirical data needed to give a good

estimate of the variation in costs and cost effectiveness

of telestroke. A mathematical analysis such as a

tive treatment; however, the cost will decrease to

$US109 131 if he/she is treated with thrombolysis.

This material is

equal share of the extra fixed costs per year

the copyright of the

It should be noted that this figure assumes that

original publisher.

Monte Carlo simulation to evaluate the robustness

Unauthorised copying

the ‘true’ ICER.

and distribution

case of $US7244 per patient can be applied. The

extra total costs for treating an acute stroke patient

with intravenous thrombolysis instead of conservative

treatment is $US6039 (calculated as the sum of

the variable costs per patient [see table I] and an

[see table II]). Hence, the total average costs per

telestroke treatment in our model amounted to

$US(7244 + 6039) = $US13 283 per patient.

telestroke costs are shared equally between stroke

centres and satellite clinics.

The result of the calculations of cost effective-

ness can be seen in table IV. The computations were

made using a time horizon of 1, 2 and 30 years, with

the assumption that by the latter timepoint approxi-

mately 98.5% of the treated patients would be dead.

The computations show the expected total average


Table III outlines the total budgetary impact per

year of implementing the national use of thromboly-

sis for acute ischaemic stroke in Denmark.

of the ICER could give a false sense of precision and

has not been performed. More empirical data on

costs and cost effectiveness are needed to evaluate

In our model, the budgetary impact was calculated

as $US3.0 (range 2.0–5.8) million per year based

on five centres and five satellite clinics, or $US3.6

(range 2.4–7.0) million per year with seven centres

and seven satellite clinics. The additional resources

needed at individual local centres/satellite clinics

may, however, exceed this amount considerably.

This is primarily due to the fact that for every

thrombolysis patient, two or three extra patients will

need to be transported to the centre/satellite clinic in

order to identify the patient eligible for thromboly-

sis. These extra patients will need conservative

treatment instead. [7,9] In this case, the resources

must, in principle, follow the patient and there will

cost per patient with acute ischaemic stroke, depen-

is prohibited.

ding on whether the patient received thrombolysis

with alteplase or the present (conservative) treatment.

Table IV also shows that, on average, a patient

with acute ischaemic stroke costs society approximately

$US27 014 in the first year after a stroke if

he/she is treated conservatively and $US30 021 if

he/she is treated with thrombolysis. Over the rest of

his/her life expectancy, the patient will cost approximately

$US129 019 (present value) with conserva-

© 2008 Adis Data Information BV. All rights reserved. CNS Drugs 2008; 22 (1)

80 Ehlers et al.

be a need for a reallocation of the capacities/resour- (range 2.0–5.8) million per year. The health ecoces

between regional/local hospitals. Our computa- nomic model computations suggest that the

tions only cover the value of the extra resources that macroeconomic costs may balance with the savings

are consumed at a national level as a result of the in care and rehabilitation after as little as 2 years,

introduction of thrombolysis – not the need for and that potentially large long-term savings are

adding resources at the individual hospital. associated with thrombolysis. However, the necessary

The costs to the individual stroke unit will deble

Danish data are not available to carry out relia-

pend on the existing competences and the capacity

computations of the possible long-term savings.

of the unit. Different types of extra costs may occur,

depending on the local circumstances, if the unit is Acknowledgements

not as readily prepared to handle the task as the most

well equipped departments and hospitals in the tion and the Aarhus University Hospital, Aarhus, Denmark.

country. Potentially large economic advantages can The authors have no conflicts of interest that are directly

be achieved through a national coordination of the relevant to the content of this study.

number of (and the geographical location of) stroke



1. Hacke W, Donnan G, Fieschi C, et al. Association of outcome

It is possible that the introduction of thrombolywith

early stroke treatment: pooled analysis of ATLANTIS,

ECASS, and NINDS rt-PA stroke trials. ATLANTIS Trials

sis may spur a further debate on the other initiatives Investigators, ECASS Trials Investigators, NINDS rt-PA

that may benefit individuals who have had a stroke.

2. Fagan SC, Morgenstern LB, Petitta A, et al. Cost-effectiveness

In this regard, the discussion concerning ambulance of tissue plasminogen activator for acute ischemic stroke.

helicopters/landing grounds and information cam-

NINDS rt-PA Stroke Study Group. Neurology 1998; 50: 883-


paigns towards the public might be relevant. These 3. Sinclair S, Frighetto L, Loewen P, et al. Cost-utility analysis of

conditions were not included in this model.

stroke: a Canadian healthcare perspective. Pharmacoeco-

The ICER with a time horizon of 1 year was

nomics 2001; 19 (9): 927-36

calculated to be $US50 100 per QALY. Since 4. Sandercock P, Berge E, Dennis M, et al. A systematic review of

thrombolysis with alteplase increases short-term

the effectiveness, cost-effectiveness and barriers to implemen-

This study has been funded by The Danish Heart Founda-

This material is

departments and satellite clinics that offer this treatthe

copyright of the

Study Group Investigators. Lancet 2004; 363: 768-74

original publisher.

tissue plasminogen activator therapy for acute ischaemic

Unauthorised copying

costs but results in long-term savings, it is mislead- ischaemic stroke in the NHS. Health Technol Assess 2002; 6

and distribution

Danish follow-up studies exist in which thrombolyis


ing to only consider the ICER calculated for the first

year. The problem with adopting a time perspective

of several years is the consequent uncertainty. No

tation of thrombolytic and neuroprotective therapy for acute

(26): 1-112

5. Sandercock P, Berge E, Dennis M, et al. Cost-effectiveness of

thrombolysis with recombinant tissue plasminogen activator

for acute ischemic stroke assessed by a model based on UK

NHS costs. Stroke 2004; 35: 1490-8

6. Mar J, Begiristain JM, Arrazola A. Cost-effectiveness analysis

sis patients have been followed-up over several of thrombolytic treatment for stroke. Cerebrovasc Dis 2005;

years and costs related to the outcomes after the 20: 193-200

7. Ehlers L, Andersen G, Clausen LB, et al. Cost-effectiveness of

treatment measured. This makes long-term compuintravenous

thrombolysis with alteplase within a 3-hour window

after acute ischemic stroke. Stroke 2007; 38 (1): tations uncertain.


8. Demaerschalk BM, Yip TR. Economic benefit of increasing

utilization of intravenous tissue plasminogen activator for

Conclusion acute ischemic stroke in the United States. Stroke 2005; 36:


The extra costs to the hospital services for impletherapy

in patients with acute ischemic stroke [in Danish].

9. Bech M, Ehlers L, Andersen G, et al. HTA of thrombolytic

Aarhus: HTA Unit, Aarhus University Hospital, Department

menting telestroke at five new satellite clinics in

Denmark were calculated to amount to $US3.0 of Public Health, 2005

© 2008 Adis Data Information BV. All rights reserved. CNS Drugs 2008; 22 (1)

Modelling the Budgetary Impact and Cost Effectiveness of Alteplase via Telemedicine 81

10. Audebert HJ, Kukla C, von Claranau SC, et al. Telemedicine for window after stroke onset: a cohort study. Lancet Neurol 2006;

safe and extended use of thrombolysis in stroke: the telemedic 5: 661-7

pilot project for integrative stroke care (TEMPis) in Bavaria.

15. Porsdal V, Boysen G. Direct costs of transient ischemic attacks:

Stroke 2005; 36: 287-91

11. Dubinsky R, Lai SM. Mortality of stroke patients treated with

a hospital-based study of resource use during the first year

thrombolysis: analysis of nationwide inpatient sample. Neuro- after transient ischemic attacks in Denmark. Stroke 1998; 29:

logy 2006; 66: 1742-4


12. Davis S, Lees K, Donnan G. Treating the acute stroke patient as

an emergency: current practices and future opportunities. Int J

Clin Pract 2006; 60: 399-407

Correspondence: Dr Lars Ehlers, HTA Unit, Aarhus Univer-

13. Solomon NA, Glick HA, Russo CJ, et al. Patient preferences for

sity Hospital, Olof Palmes Allé 15, 8200 Aarhus N, Denstroke

outcomes. Stroke 1994; 25: 1721-5

14. Köhrmann M, Jüttler E, Fiebach JB, et al. MRI versus CT-based


thrombolysis treatment within and beyond the 3-hour time E-mail:

This material is

the copyright of the

original publisher.

Unauthorised copying

and distribution


is prohibited.

Vol. 20, No. 6, 2006, page 454: The second sentence of the second paragraph of section 5.2.5 states that idebenone has

been approved for use in Japan to treat the MELAS syndrome. This statement is incorrect, as idebenone is not approved

for use in any disorder, including mitochondrial disorders, in Japan.

[Scaglia F, Northrop JL. The Mitochondrial Myopathy Encephalopathy, Lactic Acidosis with Stroke-Like Episodes

(MELAS) Syndrome: A Review of Treatment Options. CNS Drugs 2006; 20 (6): 443-464]

© 2008 Adis Data Information BV. All rights reserved. CNS Drugs 2008; 22 (1)

More magazines by this user
Similar magazines