Variation in Triple Aim Measures: Implications of Clinical Signatures in Family Medicine Residency Programs

Abstract

Evidence from several specialties suggests that practice patterns developed in residency influence the quality and cost of care long after completion of training. Improving the quality, cost, and patient experience of care (the "Triple Aim") is foundational to future health systems change. We measured variation in Triple Aim measures among family medicine residency programs in a regional quality improvement collaborative (I3 Population Health Collaborative). We calculated medians and interquartile ranges for each of 11 Triple Aim measures and compared them with median splits of population and practice characteristics, including payer mix, patient race and age, electronic health record used, registry use, and National Committee for Quality Assurance patient-centered medical home recognition. All 22 participating family medicine residency programs provided baseline data. The number of practices reporting data on individual measures ranged from 9 to 17 (41%-77%). We found variation averaging 51% across all measures, from a low of 12% for readmission rates to 94% for emergency department visit rates. Variations were stable over time. We found no significant relationships between practice or population characteristics and measures, nor between practice characteristics and outcomes variation. The 22 family medicine residency programs in our study showed substantial variation in quality, cost, and patient experience of care. These variations did not appear to result from differences in practice characteristics, payer mix, or patient demographics.