The impact of health insurance type on the use of preventive care

This study investigates how health insurance type, namely FFS plans and HMOs, affects households\u27 utilization of preventive services. I study the 1993 and 1994 National Health Interview Survey and estimate both a bivariate probit model and a two-stage maximum likelihood model to control for the endogenous health insurance type. Enrollment in HMO leads to significantly more use of pap smears, blood pressure measurements, and general physical exams. For mammogram, the effect of HMO enrollment is insignificant among women 50–64 and women 30–39 but significantly positive among women 40–49. There is no significant difference between FFS plans and HMOs in counseling perimenopausal women about hormone replacement therapy. Results from the bivariate probit model suggest that HMOs are more effective in promoting smoking cessation. Overall, enrollees in HMO are more likely to use certain preventive services, though not necessarily more cost-effective ones. However, high school dropouts do not benefit more from enrollment in HMOs than high school graduates. Contradictory to the hypothesis of adverse selection, there is no evidence suggesting that HMOs attract healthier individuals with more demand for preventive care. The impact of health status on the utilization of preventive services is empirically mixed, and the correlation ratios from bivariate probit analyses indicate that individuals with more demand for preventive services are either indifferent or more likely to join FFS plans instead of HMOs