The current landscape of emergency medicine resident scheduling

Abstract

Background

The Accreditation Council for Graduate Medical Education (ACGME) and Residency Review Committee oversee resident physician work hours with additional specifics for U.S. emergency medicine (EM) residency programs. While there are maximum work hours, the regulatory bodies do not describe minimum work hours to achieve competency, leading to variable scheduling practices. This study aimed to understand the current landscape of U.S. EM residency scheduling given the expansion of programs, evolution of policies, and increased emphasis on wellness.

Methods

We conducted a cross‐sectional study to assess current strategies of U.S. EM residency scheduling. The RedCap survey was sent to all ACGME‐accredited EM residency programs across the United States via individualized emails between January 10, 2023, and March 15, 2023. Data were combined using Microsoft Excel.

Results

A total of 138 of 278 (50%) programs responded to the survey. A total of 73.2% of programs were using thirteen 28‐day blocks with the remainder using twelve 1‐month blocks or reported “other” block scheduling. The number of blocks in the ED increases with each postgraduate year (PGY). For PGY‐1 through PGY‐3, the most commonly used shift duration was 9 h. The mean total shifts per ED block and hours worked per ED block are as follows: 19 shifts and 185.1 h (PGY‐1), 18.2 shifts and 173.9 h (PGY‐2), 17.3 shifts and 163.6 h (PGY‐3), and 14.8 shifts and 157.2 h (PGY‐4). Programs provide a median for 4 weeks of vacation per year of residency.

Conclusions

Given the expansion of U.S. EM residency programs, we reevaluated the landscape of resident scheduling. We described scheduling patterns related to night shifts, vacations, requested time off, conference coverage, charting time, and circadian rhythms. Programs should utilize these data as a starting point for setting a clinical experience for their residents.

INTRODUCTION

The Accreditation Council for Graduate Medical Education (ACGME) and Residency Review Committee (RRC) creates and enforces policies surrounding maximum resident physician work hours (duty hours) with specific additional limits for emergency medicine (EM) residents. While the American Board of Emergency Medicine (ABEM) sets a minimum number of training weeks for residents per year, ¹ neither the ACGME nor the ABEM specifically state the minimum clinical hours residents need to work to achieve clinical competency. ² As a result, there can be significant variability in scheduling practices for individual programs.

In 2017, the American College of Emergency Physicians (ACEP) also put forth a policy statement for scheduling patterns that promote patient safety and physician well‐being, specifically recommending circadian rhythm scheduling, shifts less than 12 h in duration, and regularly scheduled periods of 24 h off. ³ These recommendations may be challenging to follow when scheduling resident physicians given the limited timeline of the training period. One recent study linked schedule design to overall feelings of wellness in resident physicians, noting that resident preferences did not necessarily fully align with ACEP recommendations. ⁴ Prior research from 1995 and 2015 sought to evaluate EM residency scheduling practices ⁵ , ⁶ ; however, given the expansion of programs, evolution of policies, and increased emphasis on wellness, there is a critical need to better understand the current landscape of residency scheduling to inform current and future EM programs. This study sought to provide an updated description of the current clinical training experiences available based on clinical hours, shift length, and rotation numbers. In addition, we aimed to further investigate scheduling patterns related to night shifts, vacations, requested time off, charting time, and circadian rhythms.

METHODS

Study design

We performed a cross‐sectional survey study regarding scheduling practices for EM residents in the United States. Following survey best practices, we began by assembling a group of experts in residency leadership and scheduling to develop content validity. ⁷ , ⁸ We conducted a literature search to identify existing survey tools. Finding no ideal tool for this, we developed a tool informed by these existing studies and current needs identified by the group. We then engaged in a series of meetings to refine the questions. To determine response process validity, we piloted the survey on five EM residency program directors (PDs), assistant or associate program directors (APDs), and program coordinators (PCs). This study was reviewed and granted an exemption by the institutional review board of the University of Florida College of Medicine‐Jacksonville.

Study population

All U.S. EM residency programs that were ACGME accredited across the United States were included. Nonaccredited EM programs, programs that had not yet had their first class of residents at the time of the survey as well as programs outside of the United States were excluded.

Data collection

We surveyed all ACGME‐accredited EM residency programs across the United States using a REDCap survey link. The survey was sent via individualized emails to each person, between January 10, 2023, and March 15, 2023. Following the Dillman method, ⁹ we sent weekly emails to each respondent for a minimum of 3 weeks. Emails were sent first to PDs and PCs and then to APDs. If no response was obtained after three emails to each person, investigators reached out to other available program contacts (e.g., other faculty members). If there was still no response or no further contacts were available, the program was excluded from the study. If duplicate responses were received, the most recent complete response was utilized.

Data analysis

We report data descriptively, including mean with standard deviation and median with interquartile range (IQR). Data were analyzed using Microsoft Excel (2018). All data were stored in REDCap, a secure data management program. ¹⁰ , ¹¹

RESULTS

A total of 138 of 278 (50%) programs responded to the survey. Of these programs, 78 (58%) programs self‐identified as academic, 40 (29%) as community, 13 (9%) as county, and seven (5%) as other. The majority of the programs were 3‐year (n = 114; 83%), with 24 programs (17%) being 4‐year. Schedules were created by faculty (PD/APD/faculty) in 58 programs, PCs in eight programs, or residents in 127 programs. These data are presented in Supplemental Table S1.

We found that 73.2% (101/138) of programs were using thirteen 28‐day blocks. The remainder of the programs reported the following: 17.4% (24/138) reported twelve 1‐month blocks and 9.4% (13/138) reported other block scheduling. Number of blocks in the emergency department (ED), number of shifts per block, total hours per block, and overnight shifts by PGY are presented in Table 1. The duration of shifts by PGY are presented in Table 2. The most commonly used shift duration for PGY‐1, PGY‐2, and PGY‐3 was 9‐hour shifts. For PGY‐4s, the most common durations were 8‐and 12‐h shifts. Over half of programs 59.9% (82/137) reported following circadian rhythm scheduling. The number of blocks in the ED by PGY‐1 to ‐3 versus PGY‐1 to ‐4 EM programs is presented in Table 3. With respect to overnight shifts, 42.0% (58/138) reported grouping all of their night shifts together, 8.7% (12/138) split their night shifts into multiple groups during an ED block, 31.9% (44/138) used a mix of either grouping all night shifts together or splitting up the night shift blocks, and 10.1% (14/138) left it up to resident preference.

TABLE 1.

28‐day blocks in ED annually, number of shifts per ED block per PGY, total hours per ED block, and number of overnight shifts per ED block per PGY.

	Total 28‐day blocks in ED	Total shifts per ED block	Total hours per ED block	Overnight shifts per ED block
PGY‐1	6 (5.5–7.0)	19 (18–20)	183 (172–200)	6 (3–6)
PGY‐2	8 (7.3–9)	18 (17–19)	175 (162–181)	5 (3–6)
PGY‐3	9 (8.5–10)	17 (16–18)	163.5 (153–171)	5 (2–5)
PGY‐4	9 (8–9.3)	16 (15–17)	160 (145–167)	5 (2–4)

Note: Data are reported as median (IQR).

TABLE 2.

Percentage of EM programs with various shift durations per PGY (n = 137).

	6‐h	7‐h	8‐h	9‐h	10‐h	11‐h	12‐h
PGY‐1	0.7% (1/138)	0.7% (1/138)	18.8% (26/138)	30.4% (42/138)	27.5% (38/138)	1.5% (2/138)	24.6% (34/138)
PGY‐2	1.5% (2/138)	0% (0/138)	23.2% (32/138)	37.0% (51/138)	28.3% (39/138)	0.7% (1/138)	24.6% (34/138)
PGY‐3	0.7% (1/138)	0.7% (1/138)	24.6% (34/138)	41.3% (57/138)	27.5% (38/138)	0.7% (1/138)	24.6% (34/138)
PGY‐4	0% (0/24)	0% (0/24)	41.7% (10/24)	33.4% (8/24)	20.8% (5/24)	0% (0/24)	41.7% (10/24)

Note: Some programs use more than one shift duration.

TABLE 3.

3‐year versus 4‐year EM programs EM block per PGY and total ED blocks (n = 138).

	PGY‐1	PGY‐2	PGY‐3	PGY‐4	Total all PGYs
3‐year	6.5 (6–8)	8 (7.6–9)	9.5 (9–10)		24 (23–26)
4 year	6.1 (4.8–7)	8 (7–9)	8.9 (8–9.5)	9 (8.0–10.0)	22 (20.5–24.5)

Note: Data are reported as median (IQR).

There was heterogeneity in the definition of weekend shifts, with the majority 54.7% (75/137) of programs defining weekend shifts as Saturday and Sunday only. The remainder of programs describe weekends as follows: 13.9% (19/137) Friday afternoon/evening/overnight shifts plus Saturday and Sunday, 24.1% (33/137) Friday overnight shift plus Saturday and Sunday, and 7.3% (10/137) used other models. The median number of weekend days off per block was 3 (IQR 2–4 days). Programs were divided on weekend scheduling, with 47.5% (65/137) of programs grouping weekend shifts together routinely (i.e., working both Saturday and Sunday of the same weekend) and 43.8% (60/137) of programs leaving this variable.

We collected information about how EM programs organize multiple facets of their schedule, including call scheduling, post–sign‐out charting duration, conference scheduling, schedule requests, vacation, and moonlighting. These data may be helpful to programs in revising their scheduling processes and thus are described in Table 4.

TABLE 4.

Call schedules, sign‐out duration, conference scheduling, schedule requests, vacation, and moonlighting.

Call schedule
Presence of a call schedule	92 (66)
Call schedule stratified by PGY level	60 (43.4)
Residents on call per day	1 (0–2)
Shift Payback when activated	50 (36.2)
Post–sign‐out charting in shift duration
Post–sign‐out charting time included in shift duration	48 (34.8)
Post–sign‐out charting duration included in shift	21.1 (±33.7)
Conference shifts
Shifts scheduled during conference	4 (2.9)
Overnight shifts scheduled before conference	78 (56.5)
Shift coverage during conference
No upstaffing	17 (12.3)
Additional attendings	74 (53.6)
Off‐service	59 (42.8)
Advanced practice providerss	86 (62.3)
Other	9 (5.6)
Schedule requests
Schedule request accepted during ED blocks	128 (93.4) a
Requests per block	3 (2–4)
Vacation
Vacation weeks per academic year	4 (3–4)
Moonlighting
Moonlighting offered	128 (92.8)
External moonlighting	45 (32.6)
Internal moonlighting	23 (16.7)
Both	61 (44.2)
Year allowed to moonlight
PGY‐1	1 (1.8) b
PGY‐2	5 (8.8) b
PGY‐3	37 (64.9) b
PGY‐4	14 (24.6) b

Note: Data are reported as n (%), median (IQR), or mean (±SD).

^a 137 respondents.

^b 57 Respondents.

DISCUSSION

This study represents the largest and most recent study to date of EM residency scheduling practices. We described the number of ED blocks and ED shifts worked by residents during each year of their clinical training. We also explored more nuanced components of resident scheduling including overnight shifts, shift duration, and schedule practices. Description of scheduling practices can help inform new programs initially developing their EM residency schedule as well as established programs that are trying to reevaluate their scheduling practices or benchmark their scheduling practices with other EM programs nationally.

Scheduling practices of residency programs was reviewed a decade ago ⁵ , ⁶ ; however, since then the number of U.S. EM residency programs has more than doubled. ¹² For the 2011–2012 academic year, Stowell et al. ⁵ reported a wide range of clinical hours residents worked in the ED, with a focus on blocks in the ED, shifts and hours worked during that block, and how that extrapolated to a 3‐ or 4‐year residency. We also examined the block rotation schedule and shifts worked during an ED block; however we also investigated scheduling patterns related to night shifts, vacations, requested time off, charting time, and circadian rhythms.

We found that programs increase the number of ED blocks for each year of training for a median of 6.5 blocks as a PGY‐1 to eight blocks as a PGY‐2 and nine blocks as a PGY‐3. According to ACGME requirements, 60% of each resident's clinical experience, including pediatric EM experiences, must take place in the ED. Our data show that programs are scheduling their interns with less than 60% of their clinical experiences in the ED but ramping up over the later years of training to meet that requirement. When comparing our data to prior data, the number of ED blocks for PGY‐1s has increased from a median of five to a median of 6.5 in our data, while PGY‐2 and PGY‐3 EM blocks have remained the same. ⁵

As the number of blocks in the ED increase over years of training, we see a concomitant decrease in the number of ED shifts or hours worked in the ED. Overall, programs typically decrease the number of shifts by approximately one shift or 10 h each PGY of training. PGY‐1s work a median of 183 h or 19 shifts per month. PGY‐2s work a median of 175 h or 18 shifts per month. PGY‐3s work a median of 163.5 h or 17 shifts per month. For 4‐year programs, PGY‐4s work a median of 160 h or 16 shifts per month. These data are similar to the data of Stowell et al. ⁵ for 28‐day blocks with 18.8, 18.0, and 17.3 shifts for PGY‐1, ‐2, and ‐3 years respectively.

Studies on EM resident scheduling have found that residents believe being able to request a single day off and having a full weekend off each block as the two most important aspects of their schedule. ⁴ Our study shows that many programs are providing the ability to request days off to their residents while on their ED rotations. Providing clinical support to allow for requests of days off and to be present at important life events may be more significant to resident wellness than number of hours worked.

While having data on how other programs are scheduling their residents can be insightful, program leaders must also be wary. Programs should utilize these data as a starting point for setting a clinical experience for their residents; however, numerical shift counts and hours should not be used in isolation. The number of hours in the ED does not necessarily correlate to education provided as the number of patients being seen during that time and the clinical experience needs to be taken into account. Program leaders need to consider the clinical experience, educational opportunities, and achievement of competency in addition to numerical shift counts or hours worked.

LIMITATIONS

This study is subject to several important limitations. As a survey study, this may not have captured some of the scheduling nuance of specific programs. Some programs with variable shift structures, which may not have been fully reflected here. We did not collect data on the specific number of hours per block and future work may consider adding this metric. There were also differences in how programs defined several metrics (e.g., block length, weekends, circadian rhythms), which created challenges in combining data. To address this, we separated out reporting where possible. However, this limits the ability to fully combine program characteristics in several areas. Additionally, while we had a response rate of 50%, this does not reflect all EM programs and it is possible that scheduling may have differed in the nonresponding programs. Finally, COVID‐19 has influenced multiple aspects of residency education and it is unclear how this may have affected scheduling. ¹³ , ¹⁴ Future research should evaluate residency scheduling post‐COVID.

CONCLUSIONS

Our study provides an updated description of the current clinical training experiences provided based on clinical hours, shift length, and rotation numbers. In addition, we described scheduling patterns related to night shifts, vacations, requested time off, conference coverage, charting time, and circadian rhythms. Programs should utilize these data as a starting point for setting a clinical experience for their residents.

AUTHOR CONTRIBUTIONS

Melissa Parsons, Alexandra Mannix, Katarzyna Gore, and Michael Gottlieb jointly conceived the study and composed the survey instrument. Jeanne Rabalais composed the list of accredited EM residency programs. Melissa Parsons supervised data collection. All authors drafted components of the manuscript, and all authors contributed substantially to its revision. Melissa Parsons takes responsibility for the paper as a whole.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Supporting information

Data S1:

Supplemental Table S1:

Parsons M, Mannix A, Gore K, Rabalais J, Gottlieb M. The current landscape of emergency medicine resident scheduling. AEM Educ Train. 2024;8:e10926. doi: 10.1002/aet2.10926