No abstract

No abstract

No abstract

Objective To determine whether Medicare managed care penetration impacted the diffusion of endoscopy services (sigmoidoscopy, colonoscopy) among the fee-for-service (FFS) Medicare population during 2001–2006. Methods We model utilization rates for colonoscopy or sigmoidoscopy as impacted by both market supply and demand factors. We use spatial regression to perform ecological analysis of county-area utilization rates over two time intervals (2001–2003–2004–2006) following Medicare benefits expansion in 2001 to cover colonoscopy for persons of average risk. We examine each technology in separate cross-sectional regressions estimated over early and later periods to assess differential effects on diffusion over time. We discuss selection factors in managed care markets and how failure to control perfectly for market selection might impact our managed care spillover estimates. Results Areas with worse socioeconomic conditions have lower utilization rates, especially for colonoscopy. Holding constant statistically the socioeconomic factors, we find that managed care spillover effects onto FFS Medicare utilization rates are negative for colonoscopy and positive for sigmoidoscopy. The spatial lag estimates are conservative and interpreted as a lower bound on true effects. Our findings suggest that managed care presence fostered persistence of the older technology during a time when it was rapidly being replaced by the newer technology.

The hospital competition literature demonstrates that estimates of the effect of local market structure on competition are sensitive to geographic market definition. Our spatial lag approach effects smoothing of the explanatory variables across the discrete market boundaries. This approach results in robust estimates of the impact of market structure on hospital pricing, which can be used to estimate the full effect of changes in prices inclusive of spillovers that cascade through the neighboring hospital markets. In markets where concentration is relatively high before a proposed merger, we demonstrate that OLS estimates can lead to the wrong antitrust policy conclusion while the more conservative lag estimates do not.1