Relationship between attributes of primary care medical homes, patient engagement, and the processes and outcomes of care among individuals living with diabetes

Objective: How primary care sites implement key attributes of the medical home may be directly associated with how patients experience care, engage in their care, and ultimately their health outcomes. This study examines how experiences with care and patient engagement (as measured by the Patient Activation Measure (PAM)) varies across four (4) patient centered medical homes (PCMHs). Methods: 1,300 randomly selected adult patients with diabetes seen at 4 National Committee for Quality Assurance Level 3 accredited academic primary care practices participated in a telephone survey. Survey items included the PAM and questions on experiences with their medical home. Ordered logistic regression analysis (controlling for race/ethnicity, marital status, education, health status, and age) was used to determine the relationship between a medical home practice and the extent to which patients were more engaged in their care. In-depth interviews with administrative and clinical staff provided information on how each clinic operated. Findings: The qualitative interviews showed that despite all 4 clinics achieving NCQA accreditation, there are stark differences in the manner and degree to which key components of the PCMH model are implemented. For example, clinicians at site A had undergone intensive specialized training in diabetes care and provided a comprehensive disease management program to its patients. Survey findings showed that a higher proportion of patients at location A reported positive experiences with care. 88 percent of patients at Location A thought that their doctor “always listened carefully to them” compared to 79, 69 and 70 percent at the other locations. Similarly, 84 percent of patients at Location A reported that their doctor “always explained things to them in a way they could understand” compared to 74, 64, and 76 percent of patients in the other three clinics. Patients at location A had higher overall PAM scores compared to patients seen at the other three clinics (mean scores of 68 compared to 63, 60, and 63). Adjusted ordered logistic regression analysis confirmed that patients at Location A were significantly more likely to have higher PAM score compared to patients at the other three clinics. Patients who reported that their doctor provided clear instructions also had higher PAM scores. Conclusion: Despite receiving similar NCQA ratings, how medical homes offer and provide services can vary. This variation may lead to differing patient experiences with care and ultimately a difference in how patients can perceive their ability to manage and engage in their care.