Methods: This was a retrospective cohort study of interview data from a three-year emergency medicine residency at a tertiary-care academic medical center. Using archived data from Interview Broker, we examined scheduling patterns between one in-person (2019–2020) and two virtual interview cohorts (2020–2021 and 2021–2022). Our outcomes were the overall cancellation rates relative to interview slots as well as the proportion of cancellations that occurred within 7 or 14 days of the interview date.

Results: There were 453 interview slots and 568 applicants invited. Overall, applicants canceled 17.1%of scheduled interviews. Compared with in-person interviews, applicants canceled significantly fewer virtual interviews (in person: 40/128 (31.3%), virtual year 1: 22/178 (12.4%), virtual year 2: 15/143 (10.5%), P = 0.001). Conversely, applicants canceled significantly more virtual interviews within both the 14-day threshold (in person: 8/40 (20%), virtual year 1: 12/22 (55.5%), virtual year 2: 12/15 (80%), P < 0.001) and the 7-day threshold (in person: 0/40 (0%), virtual year 1: 3/22 (13.6%), virtual year 2: 4/15 (26.7%), P = 0.004).

Conclusion: While limited, at our site, changing to a virtual interview format correlated with fewer cancellations overall. The proportion of cancellations within 14 days was much higher during virtual interview seasons, with most cancellations occurring during that time frame. Additional studies are needed to determine the effects of cancellation patterns on emergency medicine recruitment.

Learning Objectives: To determine average RVUs per hour for emergency medicine residents at a tertiary-care, university-based academic medical center and to characterize change in mean RVUs per hour as residents advance in training.

Background: Physician productivity is often reported in relative value units (RVU). However, RVUs are infrequently reported during residency. Studying RVUs in varied training settings may better define benchmarks for progression of resident productivity.

Methods: This was a retrospective, observational study of PGY 1-3 residents at an academic, tertiary-care center. PGY2s and PGY3s were considered equivalent in shift scheduling and responsibilities. From 07/2019 to 09/2021, RVUs were extracted from the electronic health record (Epic) using E&M billing codes (excluding procedures, which were not tied to specific residents). In the PGY1 year and continuing longitudinally, residents received individual productivity reports. Individual metrics were de-identified, coded, and analyzed.

Results: The primary outcome was the measure of mean RVUs/hr per resident overall and by class. Secondary outcomes were the change in RVUs/hr between classes at the end of each year, as well as the change in RVUs/hr for the same class year to year. Descriptive statistics were reported in mean with standard deviations. One-way ANOVA was used to determine if PGY-level had a significant effect on RVUs generated; the remainder of comparisons were made with student’s t-test. 60 RVU data points were obtained, representing 40 residents. Two classes were followed longitudinally (Table 2). Overall mean RVU/hr per resident was 2.89 RVU/hr (SD 0.89). Mean RVU/hr per resident for PGY1s, PGY2s, and PGY3s were 1.97 RVU/hr (SD 0.26), 2.67 (SD 0.77), and 3.35 (SD 0.36) respectively. Class year was predictive of RVUs generated (p<0.001). There was no significant difference in RVUs within a single class from PGY2 to PGY3 (p =0.528), but there was a significant increase from PGY1 to PGY2 (p<0.001).

Conclusion: Resident RVUs in our academic ED were associated with training year, but longitudinally, the only statistically significant increase was from PGY1 to PGY2.

Introduction: Ensuring high-quality scholarly output by graduate medical trainees can be a challenge. Within many specialties, including emergency medicine (EM), it is unclear what constitutes appropriate resident scholarly activity. We hypothesized that the quantity and quality of scholarly activity would improve with a clearer guideline, including a point system for eligible scholarly activities.

Methods: A resident Scholarly Activity Guideline was implemented for EM residents in a university setting. The guideline consists of a point system in which point values, ranging from 1–10, are assigned to various types of scholarly activities. Residents must earn at least 10 points and present their work to meet their scholarly graduation requirement. We tracked scholarly activities for graduates from the classes of 2014–2020, with the guideline being implemented for the class of 2016. In a blind analysis, we compared median total points per resident, mean counts of the Boyer model of scholarship components per resident, and mean counts of significant scholarly output per resident before vs after the guideline was implemented. Significant scholarly output was defined as an implemented protocol,a research project with data collection and analysis, a research abstract presentation, or an oral abstract presentation.

Results: Among 64 residents analyzed, 48 residents used the guideline. We found that median points per resident increased after the guideline was implemented (median, interquartile range: before 7 [7], after 11 [10, 13], P = 0.002). Post-guideline scholarly activities were found to represent more of Boyer’s components of scholarship [mean before 0.81 [SD 0.40], mean after 1.52 [SD 0.71], mean difference 0.71, 95% confidence interval [CI] 0.332 ± 1.09, P < 0.001. There was no difference in the mean significant scholarly output per resident (mean before 1.38 [SD 1.02], mean after 1.02 [SD 1.00], mean difference 0.35, 95% CI 0.93 ± 0.23, P = 0.23).

Conclusion: Implementation of a Scholarly Activity Guideline point system significantly increased the quantity and, by one of two measures, increased the quality of scholarly output in our program. Our point-based guideline successfully incorporated traditional and modern forms of scholarship that can be tailored to resident interests.