Identifying Primary Care Practice Components Leading to Optimal Diabetes Care in Patient-Centered Medical Homes

Although national measures of the quality of diabetes care delivery demonstrate improvement, progress has been slow. Throughout the United States, primary care presents a critical leverage point for improving diabetes outcomes, but improving support for diabetes care in primary care practices requires a new model of care delivery that increases coordination, emphasizes prevention, and enhances collaboration between multidisciplinary teams. Legislative initiatives focusing on health care reform have accelerated adoption of the Patient Centered Medical Home (PCMH) as the preferred model for primary care redesign, with improvement in diabetes care being the most common initial focus. The organizing principles of the PCMH are well established, however operational definitions vary and important gaps exist in our understanding about which domains of change drive practice improvement, which services or resources are most likely to improve clinical outcomes, and the circumstances under which new services might succeed or fail when introduced by a practice. A better understanding of the impact of component services across organizational domains is essential for widespread adoption of incremental change in practice. Such an evaluation is critical for informing the adoption of new models of primary care and for determining whether new PCMH services provide consistent benefits across multiple settings and across diverse diabetes populations. In 2008 the Minnesota legislature adopted pioneering legislative health care reform that established standardized criteria for PCMH certification including a comprehensive set of specific services and resources. Since that time, 384 (66%) of the 581 eligible practices in Minnesota supporting care for 3.64 million people have become PCMH certified. This `natural experiment' provides a unique opportunity to leverage the data collected as a result of this reform and prior policy collaboratives to intensively evaluate the impact of primary care practice redesign on diabetes care and outcomes. The overall goal of this proposal is to identify the specific services and resources associated with primary care PCMH practice redesign that result in the greatest improvement in diabetes care. The study evaluates change across multiple organizational domains and its impact on diabetes care and health care utilization, adjusting for differences in practice and patient characteristics. Our analyses use difference-in-differences, propensity scoring in a Bayesian framework, and instrumental variable methods to draw causal conclusions under increasingly flexible assumptions about the non-random nature of PCMH certification. Overall, the proposal provides a unique opportunity to discover better ways of providing care for individuals with diabetes, and has the potential to stimulate national change in the delivery of diabetes care in primary care settings.