Association Between Electronic Health Record Use Metrics and Clinical Performance in an Outpatient Primary Care Resident Clinic.

Abstract

Purpose: Electronic health records (EHRs) can provide valuable insights into workflow, clinical reasoning, and personal attributes; however, the indicators for how an individual acts within the EHR (EHR use metrics) are not frequently analyzed. This study examines whether EHR use metrics are associated with internal medicine resident clinical performance.

Method: In this retrospective cohort study, data on EHR use metrics and achievement of 22 clinical performance measures (CPMs) were collected between November 2021-October 2022 from University of Cincinnati internal medicine residents during a year dedicated to ambulatory care. The CPMs were sorted on an attribution-contribution continuum for subgroup analysis. The EHR use metrics were used for agglomerative hierarchical clustering to group residents with similar EHR behaviors.

Results: Thirty residents (11 [37%] male and 19 [63%] female) were included. Clustering with a subset of 10 EHR use metrics resulted in 3 clusters with different clinical performance as indicated by achievement of CPMs. The clusters were characterized as lower-performing (n = 5; mean [SD] CPMs achieved, 11.4 [2.3]; 95% CI, 9.4-13.4), middle-performing (n = 23; mean [SD] CPMs achieved, 15.8 [2.1]; 95% CI, 14.9-16.6), and higher-performing (n = 2; mean [SD] CPMs achieved, 22 [0]; 95% CI, 22-22). After sorting the CPMs on an attribution-contribution continuum, the clusters performed differently in actions (F2,27 = 7.73, P = .002) and screenings (F2,27 = 9.60, P < .001) but not lab testing (F2,27 = 2.88, P = .07) or disease control (F2,27 = 1.01, P = .38). The lower-performing cluster had longer response times and incomplete work, whereas the higher-performing cluster was most responsive and communicative.

Conclusions: Hierarchical cluster analysis of EHR use metrics can identify EHR use patterns associated with resident clinical performance. Clustering provides a framework that will enable programs to apply EHR use metrics to augment resident assessment and feedback.