| High health care spending, inadequate quality, and fragmented delivery systems characterize the current U.S. health care system. Payers, providers, and policymakers are pursuing payment and delivery system reforms to address these challenges. The Patient-Centered Medical Home (PCMH) targets primary care as a means to improve quality and reduce costs. This study evaluates the PCMH in the NH Citizens Health Initiative Multi-Stakeholder Medical Home Pilot operating in nine family practices between 2009 and 2011.;Site visits were conducted to examine the various PCMH models and implementation processes, with a total sample of eighty-three interview respondents. Participants at each site completed a Medical Home Index and a survey of relational coordination. The major source of secondary data was the New Hampshire Comprehensive Health Care Information System multi-payer claims database. Generalized Estimating Equations, Ordinary Least Squares, and Logistic Regression with a random effect and a propensity-score matched sample were used to conduct difference-in-differences analyses for the utilization, cost, and quality data, respectively. In addition, pre-post sub-analyses with the nine sites were conducted to assess the impact of relational coordination and medical homeness on utilization, cost, and quality measures.;The nature of the PCMH model implemented across the nine sites varied significantly. Specifically, practices started with different capacities, and implemented team-based care, care coordinators, registries, and open access to varying degrees. Overall, status as an officially recognized PCMH did not yield significant utilization, cost, or quality findings at all, or in the expected direction. In the sub-analyses of the nine pilot sites, higher levels of medical homeness were associated with lower likelihood of readmission, lower ED utilization, and better performance on several diabetes quality measures. There were limited statistically significant findings for relational coordination.;The PCMH model exhibited a spectrum of characteristics, suggesting that classifying practices simply as a PCMH or not is not useful in understanding the true nature of care delivery at the practice, and is not predictive of better outcomes. Instead, attention should be paid to the specific features at the practice (e.g., presence of a care coordinator) and what practices can do to achieve higher levels of medical homeness. |
