Do Common Insurance Mechanisms Support Receipt of Evidence-Based Care? A National Analysis of Chronically-Ill Persons

• In Event: Shifting Insurance Landscapes and Access to Care: Insights from Public and Private Insurance Markets

Abstract

Research Objective: High-deductible health plans (HDHPs) are a highly-common insurance mechanism in the United States, enrolling 57% of privately-insured persons. We evaluate whether HDHPs meet the medical needs of chronically-ill people, who require regular access to health care.  More than 50% of Americans have one or more chronic illness.

Study Design: Our identification strategy leveraged multiple quasi-experimental approaches: difference-in-differences models combined with instrumental variables and entropy balancing weights.  The treatment group consisted of persons forced to enroll in a HDHP; the control group consisted of persons in nonHDHPs. Forced new enrollment in a HDHP was instrumented by employment in a full-replacement firm (an employer that only offered HDHPs). Entropy balancing was used to address any residual confounding.

Our outcomes were use of annual recommended medical care, which was abstracted by physician-investigators from disease-specific evidence-based clinical practice guidelines and converted into programming algorithsm suitable for use in claims data. Recommended medical care consisted of: 1) clinic visits; 2) lab tests; 3) prescription drugs; and 4) a composite measure evaluating all three. Results were pooled across disease type to improve the policy relevance of findings; disease-specific analyses were also conducted.

Population Studied: We study a longitudinal cohort of chronically-ill adults aged 18-64 in the 2016-2018 MarketScan data. MarketScan contains administrative and claims data from approximately 350 payers nationwide. Cohort members had asthma, diabetes, hypertension, coronary artery disease, heart failure, or major depressive disorder, chosen as they are the most common chrocni condiitons in the U.S. 

Principal Findings: Our cohort consisted of 343,137 adults. Groups exhibited covariate balance after entropy balancing.  In this multiply-morbid cohort, restricted-choice enrollment into a HDHP reduced use of recommended medical care, with persons in HDHPs reducing their use of recommended clinic visits by 3.1 percentage points (95% CI -4.9 to -1.2), p < 0.001), their use of recommended prescription drugs by 9.0 percentage points (95% CI -11.8 to -6.2, p < 0.001), and their use of recommended annual labs by 5.7 percentage points (95% CI -8.2 to -3.2, p < 0.001). Overall, HDHP enrollees were 4.7 percentage points less likely to receive recommended medical care compared to nonHDHP enrollees (95% CI -6.2 to -3.3 percentage points, p < 0.001).

Conclusions: HDHPs do not support receipt of recommended medical care, across a variety of chronic conditions.

Implications for Policy or Practice: Our results have important implications for recently-proposed U.S. federal legislation that proposes to exempt chronic illness management from HDHP deductibles.

Author

• Risha Gidwani, University of Colorado Denver
  Presenting Author