Comparison of health services utilization, medical costs, and activity limitations between cancer survivors with and without a disability.

Abstract

21 Background: Cancer survivors are one of the most vulnerable cohorts regarding health status and financial hardship. They are also more likely to have a disability than individuals without a cancer history. No prior research has systematically examined whether and how disability interacts with cancer survivorship. This study examines the association of disability with health services utilization, medical costs, and activity limitations among cancer survivors. Methods: We constructed a nationally representative cohort of cancer survivors using 2015-2019 Medical Expenditure Panel Survey Household Component File. Outcomes included health services utilization, measured by six service types (inpatient, institutional outpatient, office-based physician, office-based non-physician, ED, and prescription), total health care expenditure for all services combined and for each service type, and activity limitations (work at job, housework, and social relationship). Primary independent variables included six binary variables of functional disability conditions (hearing, vision, cognition, ambulation, self-care, and independent living) in the first model and a categorized total number of disabilities (0, 1, 2, 3, ≥4) in the second model. We employed multivariable generalized linear models and two-part models, adjusting for socio-demographics, general health status, and chronic conditions as well as accounting for complex survey design. Results: On average, cancer survivors (n = 9,359; weighted n = 21,046,285) had 0.57 days of hospital stays, 6.07 physician visits, and 16.3 prescriptions annually. The average annual health care expenditure for all services were $10,840, and 7.1% had one or more activity limitation. Notably, 38.8% of cancer survivors had at least one disability condition (ambulation 24.5%, hearing 13.8%, independent living 12.7%, cognition 11.5%). Compared to survivors without any disability, disability in self-care and ambulation was most strongly associated with the higher utilization of inpatient care (adjusted average marginal effect [AME] 0.85 days; 95% confidence interval [CI] 0.44-1.27) and physician visits (AME 1.03 visits; 95% CI 0.53-1.53), respectively. Disability in independent living (AME $5,046; 95% CI $2,638-7,455) and ambulation (AME $3,702; 95% CI $2,029-5,375) substantially increased total health care expenditure of all services. Survivors with four or more disabilities were 21.4 times more likely to have any activity limitations than those without a disability. Conclusions: Disability was associated with higher health services utilization and medical costs, and more activity limitations among cancer survivors, with the magnitude of increase varying by type of disability. Strategies to mitigate cancer disparities and financial hardship should take into account of the patients’ disability profiles.