296 - Tapping into Informatics: Using Automated Survey-generation to Understand Pediatric Trainee Interest in Feedback on Clinical Encounters

Poster Number: 296

Margaret Robinson, University of California, San Francisco, School of Medicine, San Francisco, CA, United States; Christy Boscardin, UCSF, San Francisco, CA, United States; Gillian E. Earnest, UCSF, San Francisco, CA, United States; Jennifer M. Creasman, UCSF Benioff Children's Hospital San Francisco, San Francisco, CA, United States; Glenn Rosenbluth, Univeristy of California San Francsico, San Francisco, CA, United States

Background: Governing and accrediting organizations have called for improved and increased use of ‘big data’ from electronic health records (EHR) for trainees to track clinical outcomes and quality benchmarks. The ACGME’s Clinical Learning Environment Review recommends that trainees receive data specific to their own patients rather than the entire service or clinical group. Prior work has shown that educators often do not utilize EHR for these purposes. One known barrier is lack of accurate EHR-based attribution of patients to trainees, particularly in inpatient settings. This problem may be related to an incomplete understanding of trainee interest in feedback – specifically which patients, what kind of feedback, and why.  Exploring these topics may improve and refine the use of EHR-based patient attribution for trainee feedback.

Objective: To characterize: 1) Prevalence of pediatric trainees’ interest in feedback on practice outcomes based on EHR access logs 2) Reasons for interest in patient care feedback and types of feedback desired

Design/Methods: We developed an online survey tool that uses EHR access logs to auto-populate a list of 10 patients whose charts had been opened by users in the past 24 hours. Surveys contained questions linked to each patient regarding the trainee's role and interest in feedback, including why and what type (e.g. specific management, quality metrics). Pediatric residents who participated received surveys every two days for two weeks during an inpatient rotation.

Results: From December 2020 to April 2021, 14 pediatric residents (6 PGY-1, 4 PGY-2, 4 PGY-3) completed 677 patient-provider surveys. Based on analysis of surveys linked to an initial, unique patient encounter (n=405), trainees indicated interest in feedback for only 29% (n=117) of encounters. Trainees were most interested in feedback “if something unexpected occurred” (76%) and much less so regarding their general clinical management (44%) or aggregated quality metrics (39%) (Table 1). Most common reasons for wanting feedback included feeling responsibility/ownership, something new learned or significant decision made; least common reasons included concern about vulnerability of patient/family and personal attachment (Table 1).Conclusion(s): Pediatric trainees were interested in feedback on less than one-third of encounters during an inpatient rotation. This is lower than we expected, and the finding may be a barometer on the state of the current post-graduate medical training learning environment. Future studies examining trainee reasons for not wanting feedback may shed more light on this surprising finding.
CVRobinson_CV.pdf