Characterizing home care services use among US survivors of cancer.

Abstract

90 Background: Cancer survivors often require home care services, but accessibility barriers prevent many patients and care partners from harnessing their full benefit. We sought to characterize home care services use among cancer survivors in the United States. Methods: We identified patients with and without history of cancer from 2016 to 2019 in the nationally representative Medical Expenditure Panel Survey (MEPS). We extracted sociodemographics and clinical characteristics. We constructed a composite measure of patients being eligible for home care if they reported any one of three criteria: completely unable to do any activity, limitations doing housework, or dependent on activities of daily living. We identified home care use and associated details, including type of home care provided—by licensed professionals (e.g., home health nurse, physical therapy etc.) associated with agencies, hospitals, or nursing homes, by non-agency (self-employed persons), and by unpaid informal care providers not living with the patient, and the number of days such services were provided in a calendar year. We used multivariable logistic regression to identify factors associated with eligibility for and non-receipt of home care services. Results: We identified 9,585 cancer survivors (median age, 67 years, 53% female, 76% non-Hispanic White, 15% breast cancer) and 83,393 persons without cancer history. Cancer survivors were more often eligible for (22% vs 9%) and more often used home care (10% vs 3%), compared to those without cancer. Cancer survivors receiving home care annually remained stable from 2016-2019 (10%, 8%, 11%, 12%; p = 0.25). Among cancer survivors who received home care, the annual median/mean days of services were 28/87 days. Of these, the median/mean days of service were: agency (16/66), non-agency (0/4), and unpaid informal caregivers (0/17). In multivariable analysis, higher age (per year, aOR 1.02), female sex (aOR 1.29), income below poverty (aOR 1.82), unmarried status (aOR 1.65), public insurance (aOR 2.15), and number of comorbidities (per 1 comorbidity, aOR 1.70), had increased odds for being eligible for home care (p <0.05 for all). In multivariable analysis, higher age (per year, aOR 1.01), unmarried status (aOR 1.55), and number of comorbidities (per 1 comorbidity, aOR 1.58) had higher odds, while Hispanic ethnicity was associated with lower odds of receiving home care (p <0.05 for all). Conclusions: In this nationally representative study, we found that only 10% of US cancer survivors received home care services, with notable sociodemographic variability. Our data highlight an urgent need to understand underlying market factors, and whether these data reflect a capacity, incentive, or implementation problem. Policy level efforts from the recent American Rescue Plan Act and upcoming Cancer Moonshot initiatives should aim to fund home-based care programs.