Abstract
Purpose :
Post-clinic emergency room and inpatient ophthalmic consultations ("on call") allow unique insights due to increased resident autonomy. We use these encounters to evaluate resident performance in relation to multiple systemic variables to help optimize training and patient care.
Methods :
A retrospective study used electronic medical records to collect logistics, clinical data, and resident variables for resident on-call encounters between 7/2019 – 7/2020. Resident data was anonymized and initial encounters were compared with serial follow-up visits for accuracy. Performance was scored by a modified ACR RADPEER system of review. Diagnostic accuracy = 0/1. Diagnostic difficulty = 1/2/3. Management accuracy = 0/1. Management difficulty = 1/2/3, where 2 should involve a senior resident and 3 should involve faculty. Disposition appropriateness = 0/1. Adverse outcome = 0/1 if results were secondary to management and disposition appropriateness.
Results :
209 of 501 encounters had serial follow-up data to use for grading. 23/501 patients were triaged to a later clinic appointment without direct encounter. 9/209 patients followed up 7.1±6.8 days after indirect encounters. Of the 200 direct encounters (difficulty 1.6±0.7), 191 diagnoses were accurate (difficulty 1.6±0.6). 9 were inaccurate (difficulty 2.2±0.8) and encountered a late PGY-2 on average. 2/11 inaccurate diagnoses led to suboptimal management, although both had appropriate follow-up triage and no adverse outcomes. 194/200 had appropriate initial management (difficulty 1.7±0.7). 6/200 were suboptimal (difficulty 1.8±0.4) and encountered a late PGY-2 on average but had no adverse outcomes. 199/200 had appropriate disposition (location and duration). 3/209 were marked as adverse outcomes, one from each level of training: 2 were from delayed IOP management and 1 was a mistriage based on ED staff evaluation.
Conclusions :
Sub-optimal performances occurred infrequently in our current redundancy-designed safety escalation system. Performance scores did not statistically vary with different residents, clinical difficulty, or call logistics. Due to appropriate follow-up, suboptimal performances were not associated with worse visual outcomes, but rare cases show the need for prudent high IOP management and reliable ED physician sign out of ocular vitals.
This is a 2021 ARVO Annual Meeting abstract.