The impact of physical activity on short-run healthcare utilization and costs in U.S. adults

As healthcare costs have grown well in excess of inflation, policy debate has frequently focused on whether and how prevention behaviors could reduce healthcare utilization and costs. I attempt to identify the extent of substitution between physical activity, a form of health investment, and medical utilization in the short-run. The short-run is an important duration for employers and health insurers with high employee and member turnover rates. My model posits two benefits of physical activity--a reduction in the probability or extent of disease and a signal which reduces uncertainty regarding one's health. The 'signaling' effect likely substitutes for diagnostic care while the 'physiological effect' more likely substitutes for relatively low-intensity forms of treatment in the short run. Unlike prior statistical efforts examining physical activity; my empirical approach uses various novel instrumental variables, including spouse's physical activity, which are correlated with one's own activity pattern, but plausibly is not related to one's medical utilization in particular scenarios. Evidence from the National Health Interview Survey/Medical Expenditure Panel Survey (2002-4) suggests that individuals respond to their physical activity by reducing more discretionary forms of care. Specifically, several specifications indicate that physical activity reduces "check-up" and physician office visits by approximately 15-20 percent in the short run. However, there were no significant changes in the probability of less discretionary forms of care such as hospitalization or ER visits. This result is consistent with a model in which the signaling or near-term physiological benefit from physical activity substitutes for more discretionary forms of short-run utilization, but not the decision to seek intensive treatment-focused care. Overall, these shifts do not appreciably affect overall utilization or costs.