High deductible health plans (HDHPs) remain controversial in the ongoing health care debate, yet HDHPs have attracted considerable attention from employers, individuals, and policy makers, largely because of their perceived potential to reduce costs. The percent of large employers offering HDHPs has increased from 13% in 2009 to 28% in 2013.1 The popularity of HDHPs is rooted in the idea that greater cost sharing will lead to reductions in health care spending. However, there are tradeoffs. Exposing patients to higher spending reduces the risk protection associated with insurance. Moreover, foregone care in the short-run may lead to a less healthy patient in the long-run, ultimately increasing future spending. Despite the importance of the long- run implications of reduced health care usage, there is little evidence as to how HDHPs affect health over a period of greater than two-years. This research will contribute to the literature on HDHPs by looking at multiple large firms that self-insure through a single large national health insurance carrier with members enrolled in health plans across the United States over a four-year (2009-2012) period. Instrumental variables (IV) methods will be used to obtain plausibly causal estimates of the effect of enrollment in a HDHP on a number of outcomes of interest to policymakers and stakeholders ? namely health care utilization, spending, and health stock deterioration. All employers in the data set can offer a HDHP to their employees; however, not all do. Past work has found evidence of favorable selection into HDHPs. This selection would bias the estimates if a simple ordinary least squares methodology was used. To address the selection issue, the employer's offer of a HDHP will be used as an instrument for being enrolled in a HDHP. The goal of this proposed research is to identify the effects of HDHP enrollment in the short- and long-run. Health care affordability and efficiency is a key priority area for AHRQ, and this research will address this topic not only for the working-age population as a whole but also for specific AHRQ Priority Populations for which there are identified evidence gaps. Specifically, this research will illustrate how reductions in health investments in the current period affect an individual's future health state.
The proposed study will address health insurance benefit design and how it can influence health care spending and service use. Using an instrumental variables design, the causal effects will be estimated for how raising the initial out-of-pocket cost of care impacts service utilization and health. The study is expected to generate new evidence germane to ongoing policy questions surrounding higher cost sharing and its impact on health outcomes.
