Annual Meeting Proceedings Part 1 - American Society of Clinical ...

6028 Poster Discussion Session (Board #16), Tue, 8:00 AM-12:00 PM and
11:30 AM-12:30 PM
How do elderly cancer survivors fare? A comparison of characteristics,
health status, health behaviors, and spending between Medicare beneficiaries
with and without a cancer history. Presenting Author: Xuehua Ke,
University of Maryland School of Pharmacy, Batlimore, MD
Background: Prior research on cancer (Ca) survivors has been limited to
small cohorts or claims data with limited measures. In this study we used
population-based survey data to characterize multiple dimensions of the
survivor experience including socioceonomic status, health status and
behaviors, and healthcare spending, comparing survivors to beneficiaries
w/o a Ca history. Methods: Data were pooled from the 1997-2007 Medicare
Current Beneficiary Survey (MCBS) including linked claims (3 yrs prior to
MCBS enrollment to concurrent). Ca survivors were identified from ICD-
9-CM codes on claims or self report of a prior Ca dx, survived 2� yrs from
observed Ca dx, and had no active treatment, hospice enrollment or death
in the selected survivorship yr. Beneficiary characteristics and healthcare
use were based on self report augmented by claims-based measures.
Bivariate and multivariate analyses were used to compare survivors’
characteristics and spending to controls. Results: The study included
3,921 survivors and 7,056 subjects w/o Ca. Among survivors, breast
(21%), prostate (21%), and colorectal (16%) were most common Ca sites.
Compared to controls without (w/o) Ca, survivors had higher income and
assets, were more likely to have supplemental medical and prescription
drug coverage (74% vs. 70%), had more comorbid conditions as measured
by Charlson Comorbidity Index score ��2 (18% vs. 15%), and a smoking
history (60% vs. 53%); survivors were less likely to avoid visiting doctors
(22% vs. 30%), and more satisfied with the quality of medical care
received (96% vs. 92%). Survivors had higher annual healthcare spending
(mean $12,095 vs. $8,928) and outpatient drug spending ($2,205 vs.
$1,898). All differences in healthcare use and spending remained significant
after adjusting for socioeconomic status, health conditions, and
behaviors. Conclusions: Elderly Ca survivorshad more supplemental insurance
coverage, stronger preferences for medical care, used more preventive
services, and had higher spending compared to beneficiaries w/o Ca. These
patterns may reflect the effects of survivorship on need for healthcare and
care seeking behaviors.
6030 Poster Discussion Session (Board #18), Tue, 8:00 AM-12:00 PM and
11:30 AM-12:30 PM
Longitudinal changes in health care utilization by adult survivors of
childhood cancer in the Childhood Cancer Survivor Study (CCSS). Presenting
Author: Jacqueline N. Casillas, UCLA, Los Angeles, CA
Background: The incidence of late effects increases as childhood cancer
survivors age. Survivors require lifelong care focused on the risks arising
from prior cancer therapy (survivor-focused care). Methods: We assessed
longitudinal changes in health care utilization in adult survivors of
childhood cancer participating in the CCSS. Utilization at baseline and
most recent follow-up was classified into one of three mutually exclusive
hierarchical categories: no health care, general medical care, or survivorfocused
care. Relative risk (RR) and 95% confidence intervals (CI) were
calculated for predictors of reduction in care over time from survivorfocused
to general or no care. Multivariable models, adjusted for key
treatment exposures, were created to assess the risk factors for reductions
in level of care over time. Results: Among 8591 eligible survivors, mean age
at last follow-up was 35.1 years (SD�7.8) with a mean of 11.6 years
(SD�2.2) since baseline. Of 3993 (46%) survivors who reported survivorfocused
care at baseline, 2383 (59.7%) reported a lower level of care at
follow-up. Among 4598 (54%) not receiving survivor-focused care at
baseline, 915 (20%) reported survivor-focused care at follow-up. Baseline
predictors of a decreased level of care were no health insurance (RR�1.5,
95% CI 1.2-1.9), male sex (RR�1.4, 95% CI 1.2-1.6), being 10-19 years
from diagnosis compared with 20� years (RR�1.4, 95% CI 1.1-1.7). In
contrast, factors associated with a maintenance in survivor-focused care
were Canadian residency compared to U.S. residency with insurance
(RR�0.7, 95% CI 0.6-0.9), unemployment (RR�0.8, 95% CI 0.7-0.9),
physical limitations (RR�0.7, 95% CI 0.6-0.9), cancer-related pain
(RR�0.7, 95% CI 0.5-0.8), poor emotional health (RR�0.7, 95% CI
0.5-0.9), having mild-moderate (RR�0.5, 95% CI 0.4-0.6) or severedisabling
chronic health condition (RR�0.6, 95% CI 0.5-0.7). Conclusions:
Less than a third of adult survivors of childhood cancer report survivorfocused
care. Rates decrease over time. Targeted interventions to maximize
survivor-focused care in at-risk survivors should be tested so preventive and
risk-reducing opportunities are not lost.
Health Services Research
6029 Poster Discussion Session (Board #17), Tue, 8:00 AM-12:00 PM and
11:30 AM-12:30 PM
Patient-centered medical homes may improve breast cancer surveillance
among survivors. Presenting Author: Stephanie B. Wheeler, University of
North Carolina at Chapel Hill, Chapel Hill, NC
Background: Community Care of North Carolina (CCNC) initiated a medical
home (MH) program in the early 1990s focused on improving care in
Medicaid-insured populations. CCNC has been successful in improving
asthma, diabetes, and cardiovascular disease outcomes, but has not been
examined in the context of cancer care. We sought to determine whether
CCNC enrollment was associated with improved cancer surveillance among
breast cancer survivors. Methods: Using state cancer registry records linked
to Medicaid claims, we identified women ages 18-64 diagnosed with stage
0, I, or II breast cancer from 2003-2007. We included only cases insured
by Medicaid for at least 12 of 15 months following the index cancer
diagnosis. Reflecting ASCO guidelines for breast cancer surveillance for
survivors (2006), we defined outcomes as time to first surveillance
mammogram post-diagnosis and overall receipt of mammogram by 15months
post-diagnosis. Our primary independent variable was enrollment
in CCNC, categorized as never enrolled, enrolled up to 6 months, and
enrolled 7 months or more. We used multivariate Cox proportional hazards
stratified by receipt of radiation therapy (RT) and logistic regressions.
Results: 840 women were included in our sample. Approximately half were
enrolled in CCNC for at least some time during the study period, 38% for
more than 7 months post-diagnosis. Among women who received RT, being
in a MH for at least 7 months corresponded to earlier follow-up mammogram
(Hazard Ratio: 1.34; p�0.028), controlling for all other factors.
Enrollment in a MH for at least 7 months post-diagnosis also was
associated with overall receipt of mammogram by 15 months (p�0.01).
Interaction terms indicated that women enrolled in MHs and living in a
rural area had a statistically significant higher likelihood of receiving
mammography. Conclusions: Results suggest that MH enrollment is associated
with improved cancer surveillance among breast cancer survivors
insured by Medicaid. Given the growing population of cancer survivors and
increased emphasis on MHs in the Affordable Care Act, more research is
needed to explore how patient-centered medical homes can be enhanced to
improve the transition from cancer patient to cancer survivor.
6031 Poster Discussion Session (Board #19), Tue, 8:00 AM-12:00 PM and
11:30 AM-12:30 PM
How radiation oncologist accessibility influences treatment choice and
quality in early-stage breast cancer: A SEER database analysis. Presenting
Author: Thomas M. Churilla, The Commonwealth Medical College, Scranton,
PA
Background: Mastectomy and breast conserving therapy (BCT, partial
mastectomy and adjuvant radiotherapy) are equivalent in survival for
treatment of early stage breast cancer. This study evaluated the impact of
radiation oncologist accessibility on choice of mastectomy versus BCT, and
the receipt of radiotherapy after BCT. Methods: In the NCI SEER database,
breast cancer cases from 2004-2008 were selected with the following
criteria: T2N1M0 or less, lobular or ductal histology, and treatment with
simple mastectomy or partial mastectomy (�/-) adjuvant radiation. The
HRSA Area Resource File was combined to define average radiation
oncologist density (ROD, number of radiation oncologists/100K people) by
county over the same time period. Tumor characteristics, demographic
information, and ROD were evaluated with respect to mastectomy rates and
receipt of radiation therapy after BCT in univariate and multivariate
analyses. Results: In the 118,961 cases analyzed, mastectomy was
performed 33.3% of the time relative to BCT. After adjustment for
demographic and tumor variables, the odds of having mastectomy versus
BCT were inversely associated with ROD (OR [95% CI] � 0.94 [0.93-
0.96]; p�0.001). Adjuvant radiation therapy was not administered in
23.4% of BCT cases. Likewise, the odds of having BCT without adjuvant
radiation were inversely associated with ROD (0.96 [0.95-0.98]; p�0.001,
table). Conclusions: There was a significant, inverse and linear relationship
between ROD and mastectomy rates independent of demographic and
tumor variables. An inverse trend was also observed for the omission of
radiotherapy after BCT. Access to radiation oncologists was a factor in
surgical choice and receiving appropriate BCT in early stage breast cancer.
Radiation
oncologist
density a
(percentile)
Mastectomy
rate b (%)
Odds of
omission of
radiotherapy
after BCT c [95% CI] p value
0-10 th d 45.4 1.30 1.22-1.39 �0.001
10-25 th 40.6 0.92 0.85-0.99 0.020
25-75 th 37.9 1.00 Referent --
75-90 th 40.0 0.85 0.79-0.92 �0.001
90-100 th 36.4 0.89 0.83-0.95 �0.001
a Number of radiation oncologists/100K.
b Mastectomy/[Mastectomy � BCT w/ radiation].
c Multivariate.
d 0-10 th represents 0 radiation oncologists/100K.
Visit abstract.asco.org and search by abstract for the full list of abstract authors and their disclosure information.
389s