Using Statewide Data on Health Care Quality to Assess the Effect of a Patient-Centered Medical Home Initiative on Quality of Care.

Abstract

Patient-centered medical homes comprise a large portion of modern health care redesign. However, most efforts have reflected rigid, limited models of transformation. In addition, evaluations of their impact on quality of care have relied on data designed for other purposes. Minnesota's Health Care Home (HCH) initiative is a statewide medical home model relying on state-run, adaptive certification and supportive data infrastructure. This longitudinal study leverages a unique statewide system of clinic-reported, patient-level quality data (2010-2013) to assess the effect of being in a HCH clinic on health care quality. Measures included optimal quality (meeting all targets) and average quality (number of targets met) for asthma, vascular, and diabetes care; colorectal cancer screening; depression follow-up; and depression remission. Depending on measure and year, the analytic sample included 246,023 - 3,335,994 child and adult patients in 404-651 clinics. Using endogenous treatment effects models to address endogeneity, and including patient- and clinic-level covariates and clinic-level selection bias corrections, the authors produced potential outcomes means and average treatment effects (ATEs). HCH patients received better quality versus non-HCH patients for most outcomes. For example, the adjusted rate receiving optimal diabetes care was 453.7/1000 adult HCH patients versus 327.2/1000 non-HCH adult patients (ATE = 126.5; P < .001). By contrast, depression remission showed no HCH-related benefit. Findings on average care quality generally echoed optimal care findings. These findings indicate the usefulness of statewide quality data and support the effectiveness of adaptive, state-run medical home programs. Additional integration of services may be needed for mental health conditions.