Factors Associated with Health Care Costs in Older Adults with Type 2 Diabetes: Insights for Value-Based Payment Models.

Abstract

Diabetes accounts for 1 in 4 health care dollars spent. Succeeding in value-based payment models depends on identifying those at risk for high costs and providing them with appropriate treatment. The objective was to determine factors associated with type 2 diabetes mellitus costs. In this cohort study, this study used longitudinal data from a national insurer between 2016 and 2020. The authors included individuals aged 65 and older with type 2 diabetes mellitus with at least 12 months of continuous enrollment in Medicare Advantage. Exclusions included those who died during the study period or had incomplete data. Factors included study year, demographics (age, sex, race/ethnicity, language, dual eligibility, rurality), and diabetes complications (Diabetes Complications Severity Index). The outcomes of interest were medical and prescription costs. The study included 49,843 individuals. Diabetes complications (coefficient = $3582.11, P < 0.001), year (coefficient = $1003.22, P < 0.001, 2020 vs. 2016), sex (coefficient = $238.35, P < 0.001, female vs. male), dual eligibility (coefficient = $618.61, P < 0.001, yes vs. no), and rurality (coefficient = $1242.38, P < 0.001, yes vs. no) were associated with higher medical costs. Age (coefficient = $-2851.67, P < 0.001), race/ethnicity (coefficient = $-1458.03, P < 0.001, Black vs. White; coefficient = $-1679.81, P < 0.001, Hispanic vs. White), and language (coefficient = $-2523.29, P < 0.001, Spanish vs. English) were associated with lower medical costs. Individuals who had complications, were female, were dually eligible, and lived in rural communities had higher medical costs. Black, Hispanic, and Spanish-speaking individuals had lower medical costs, mirroring well-known disparities. Policy makers and health care organizations can use these data to more efficiently deliver care to some while ensuring adequate access for others.