The Patient-centered Medical Home

BACKGROUND: The patient-centered medical home (PCMH) has increasingly been looked to by policy makers, health care providers, and private insurers as a potential solution to the fragmented and inefficient US health care system. Whether the PCMH achieves these goals is not known. OBJECTIVES: To evaluate a PCMH demonstration project implemented in 2011 in 8 New Jersey primary care practices covering over 10,000 plan members. RESEARCH DESIGN: We conduct difference-in-differences analysis, comparing changes in outcomes at 8 medical home practices to a group of 24 comparison practices before (2010) and after (2011) the medical home implementation occurred. We use Mahalanobis distance matching to select the 24 comparison practices, matching on practice characteristics. We focus on the effect of the PCMH pilot on 3 groups of outcomes: health care utilization, costs, and quality. RESULTS: The study cohort included 35,059 members during the study period 2010-2011-10,004 in the 8 PCMH practices and 25,055 in the 24 comparison practices. Health care utilization and costs did not significantly change with adoption of the PCMH model. In testing for changes in Healthcare Effectiveness and Data Information Set (HEDIS) quality measures, rates of mammography increased in PCMH practices after PCMH implementation compared to non-PCMH practices, by 2.2 percentage points on a base of 69.5% (P<0.001). Rates of nephropathy screening also increased (by 6.6 percentage points on a base of 51.8%; P=0.05). Changes in 7 other HEDIS quality measures following PCMH implementation were not statistically significant. CONCLUSIONS: We find little evidence of reductions in health care utilization or cost and minimal evidence of improvements in quality of care. Ongoing work is needed to understand why this model of care seems to work in some cases and not others and to evaluate how to improve the medical home.