Assessing change in primary care sensitive measures before and after implementation of a Patient-Centered Medical Home model in Alaska

Background: The patient-centered medical home (PCMH) model of primary health care can be defined as a system in which the responsibility for patient care, and the coordination of that care, resides with a personal medical provider working as part of a health care team. There is a need for systematic evaluations of PCHM models to determine if they provide improved patient access to quality health care, including health promotion and disease prevention services. This paper describes preliminary results of an evaluation of a PCMH model of care at the largest Alaska Native-owned and managed primary care system in the United States. Methods: Process and outcome measures for key primary care sensitive health conditions were aggregated by month over a minimum of three years both prior to and after the implementation of a PCMH model. An interrupted time series analysis of monthly process and outcome measures was calculated for this period. Semi-structured interviews were conducted with patients, health system data managers and leadership, clinicians, administrators, and tribal leaders to provide contextual data on these quantitative findings and to collect first-person experiences with the PCMH model of care. Findings: Our findings combine quantitative and qualitative data regarding primary care sensitive conditions to triangulate and contextualize the metrics employed. For example, the time series analyses suggest some improvements in inpatient stays for diabetic complications following implementation of the PCMH model: the rate of inpatient stays for this primary care sensitive condition increased by an average of 0.02 (p<0.01) per month in the three years prior to implementation of the PCMH model, and decreased by 0.01 per month (p<0.01) following that implementation. Qualitative descriptions of changes in organizational structure and processes reveal a commitment to improved diabetes management during the PCMH intervention. Conclusions: Preliminary findings from a mixed-methods evaluation of primary care sensitive health conditions indicate some improvements in health care processes and outcomes among a population of Alaska Native/American Indian patients. Time-series evaluations of PCMH transitions in medical systems can be constrained by the availability of process and outcome data related to primary care sensitive conditions. These findings may prove useful for the development, implementation, and evaluation of PCMH models in other health care systems.