Market Structure and Public Reporting: Evidence from Nursing Homes

Public reporting of quality is intended to induce quality improvement on the part of providers.  But theoretically, the conditions required for widespread quality improvements in response to public reporting may not apply in many health care markets.  Depending on the quality of a provider relative to the quality of the competing providers within the market, no increase or even decreased quality may be optimal for some, particularly when the cost of improving quality is high relative to the expected increase in demand from improving quality.  Thus, a type of sorting or separating equilibrium may result.  The empirical literature on public reporting has largely ignored these crucial aspects of market structure and strategic behavior.  The objective of this study is to fill that gap by examining the effect of key aspects of market structure on quality improvement in response to public reporting, drawing on evidence from Nursing Home Compare.

We analyze a 1999-2010 panel of Minimum Data Set (MDS) assessments for quality information and payer mix, merged with Online Survey and Certification data for time-varying facility-level characteristics.  We classify nursing homes by payer mix and by the competitiveness of the market (monopoly, duopoly, and more competitive counties), and further define the clinical quality of the nursing home relative to the other nursing homes in the market. We use descriptive analyses of coefficients of variation and fixed-effect (at the market or facility level) regressions to assess the improvement in reported quality measures of each nursing home type under Nursing Home Compare, controlling for the baseline level of quality, the range of quality in the market, and the relative position of each nursing home. Our sample includes all nursing homes nationally from 1999-2010 that are certified for Medicare and Medicaid (n=15,514 at baseline). 

We find that approximately 24 percent of nursing home markets are monopolies; 21 percent duopolies; and 56 percent more competitive.  While significant variation in performance on reported quality measures by type of market existed at the launch of Nursing Home Compare in 2002, performance across markets converged dramatically for most measures by 2010. Performance within markets also exhibited some degree of convergence, with low-quality nursing homes improving more and high-quality homes sometimes declining.  Convergence was higher for markets with a larger range of quality at baseline.  Thus, contrary to fears that Nursing Home Compare would result in a growing gap in quality, performance on the clinical performance measures has converged over time. This may imply that the cost of improvement on these measures is relatively low. Although convergence in performance on clinical quality measures (generally in a positive direction) can be seen as a success of Nursing Home Compare, it may render the scores on these measures less meaningful, as little residual variation exists to differentiate providers.