Does Health Insurance Encourage or Crowd Out Beneficial Nonmedical Care? A Dynamic Analysis of Insurance, Health Inputs, and Health Production

Abstract

Determination of evidence-based best practice guidelines and evaluation of medical and nonmedical determinants of health requires modeling of consumer demand as well as health production. Toward this end, we examine the impact of endogenous health insurance coverage, which creates variation in the prices of medical care across individuals as well as a change in the relative price of nonmedical health inputs. Specifically, we analyze the decisions of individuals with diabetes to monitor, treat, and manage their condition and their subsequent health outcomes. Diabetics can experience serious or fatal complications without regular monitoring (of blood glucose and other indicators of disease severity) and, in some cases, prescription medication. These activities can present a significant financial burden that could be substantially attenuated by health insurance. Through cross-price effects, insurance may also influence important nonmedical input choices such as exercise and diet. Using a sample of nonelderly diabetics from the Medical Expenditure Panel Survey, we jointly estimate demand equations for health insurance, medical treatment, lifestyle decisions, and health, controlling for their common unobserved determinants. We find that insurance with drug coverage leads to better adherence to diabetic care guidelines. There is, however, some evidence of ex ante moral hazard: those covered by insurance with a drug plan show slightly lower probabilities of exercising regularly.