State health policy making determinants, theory, and methods: A synthesis

The American states exhibit considerable differences in health policy and market characteristics. Not only do they display substantial variation in spending, but they also display substantial variation in the strategies chosen to control costs, improve access, and ensure quality care. This article synthesizes studies that use 50-state statistical techniques to model policy adoption in the health sector. The purpose is to assess the strengths and weaknesses of this literature, to place it in the context of comparative state policy research generally, and to identify factors that best predict 17 health policy outcomes at the state level. A database was assembled containing 245 equations abstracted from 63 studies published between 1975 and 2002. Some predictors (such as income, aged population, public opinion, and nursing home beds) were studied much more frequently than others (e.g., education, divided government, federal Medicaid mandates, other states' adoptions). Results show that 43 of the 87 policy making determinants examined consistently predict two or more state-level outcomes, including four that predict five outcomes (non-white, urban, income, unemployment), two that predict six (tax capacity/effort, hospital beds), and two that predict seven (nursing home beds, liberal public opinion). Gaps are shown to exist in our understanding of the policy making effects of political system and intergovernmental characteristics.