Abstract
RESEARCH OBJECTIVE: Employers are increasingly purchasing “value‐based health insurance” designs that reduce out‐of‐pocket costs for select beneficial health services. For example, “preventive drug list” (PDL) plans make cardioprotective medications $0 to encourage adherence. Research has demonstrated that PDL designs substantially improved cardioprotective medication adherence among low‐income diabetes patients enrolled in health savings account‐eligible high‐deductible health plans (HSA‐HDHP). However, effects on health outcomes are unknown. STUDY DESIGN: Interrupted time series with matched control group. POPULATION STUDIED: Using data from 2004–2017, we analyzed a large commercial health insurance claims dataset with enrollees in all US states. The intervention group included 1498 low‐income commercially‐insured HSA‐HDHP patients with diabetes age 12–64 years who experienced employer‐mandated switches to PDL coverage; the control group included 11,984 matched HSA‐HDHP counterparts whose employers offered no PDL. We defined the index date as the day that the intervention group received the PDL benefit that reduced cardioprotective drugs to $0 out‐of‐pocket, and matched controls had a corresponding contemporaneous index date. We measured monthly rates of aggregated acute preventable diabetes complications (such as hyperglycemic coma, hypoglycemia, angina, cellulitis, and pneumonia) presenting to the outpatient setting or the emergency department. We constructed cumulative interrupted time series plots of these two measures and conducted segmented regression analysis and marginal effects estimation. PRINCIPAL FINDINGS: In both the intervention and control groups, 46% of subjects were female, 49% lived in low‐income neighborhoods, and 2.2% lived in predominantly black neighborhoods. The study groups had similar baseline trends in outcomes. From before to after the index date, the PDL intervention group experienced a 16.2% decrease (absolute change, −8.3/1000) in emergency department visits for acute, preventable diabetes complications relative to controls. Corresponding relative and absolute changes for outpatient complication visits were 0.6% and 1.0 visit/1000. CONCLUSIONS: Low‐income HSA‐HDHP members with diabetes who were switched to a generous value‐based medication benefit experienced a 16% reduction in emergency department visits for acute, preventable diabetes complications. IMPLICATIONS FOR POLICY OR PRACTICE: Value‐based health insurance designs might improve health outcomes of vulnerable populations. Policymakers, health insurers, and employers should consider encouraging uptake among low‐income, chronically ill patients. PRIMARY FUNDING SOURCE: Centers for Disease Control and Prevention.