papers

The patient-centered medical home and associations with health care quality and utilization: a 5-year cohort study

Ann Intern Med 164:395-405.

Weill Cornell Medical College, New York, NY, USA

BACKGROUND: Effects of the patient-centered medical home (PCMH) are unclear. Previous studies had relatively short follow-up and may not have distinguished effects of the PCMH (which involves electronic health records [EHRs] plus organizational changes) from those of EHRs alone.

OBJECTIVE: To determine effects of the PCMH on health care quality and utilization compared with paper records alone and EHRs alone, with extended follow-up.

DESIGN: Prospective cohort study (2008 to 2012), including 3 years after PCMH implementation. (ClinicalTrials.gov: NCT00793065).

SETTING: The Hudson Valley, a multipayer, multiprovider region in New York.

PARTICIPANTS: 438 primary care physicians in 226 practices, with 136 480 patients across 5 health plans.

INTERVENTION: Level III PCMH, as defined by the National Committee for Quality Assurance.

MEASUREMENTS: Claims-based outcomes included 8 quality and 7 utilization measures. Generalized estimating equations were used to compare adjusted differences in rates of change across study groups.

RESULTS: Patterns of quality were fairly similar across groups. Utilization patterns were similar across groups from 2008 to 2011 but showed modest differences between the PCMH and control groups on most measures in 2012. For example, hospitalizations were relatively stable from 2008 to 2011 (approximately 3.9 to 5.2 per 100 patients per year) but decreased in the PCMH group in 2012 (incidence rate ratio, 0.79 [95% CI, 0.69 to 0.90] compared with paper records). Emergency department visits were highest for the PCMH group (16.7 per 100 patients at baseline and 15.4 per 100 patients at the end of the study period) and lowest for the paper group (14.3 per 100 patients at baseline and 12.2 per 100 patients at the end of the study period), but the rate of change did not differ across groups.

LIMITATION: Possible unmeasured confounding.

CONCLUSION: The PCMH was associated with modest changes in most utilization measures and provided similar quality compared with EHRs and paper records.

PMID: 26882534