# Building an Emergency Department Schedule That Actually Holds

> Build an ED schedule around your arrival curve, honor circadian limits, distribute nights fairly, and stop post-publication swaps from breaking all three.

- Canonical URL: https://youshift.com/blog/emergency-department-scheduling
- Author: Cristian Villegas, Growth & Operations
- Published: 2026-07-27
- Updated: 2026-08-17
- Language: en
- Publisher: YouShift (https://youshift.com/)

Markdown edition of the page at the canonical URL above.

---

An emergency department schedule must match local demand and provider capacity, limit harmful rotation patterns, and distribute undesirable work under rules the group accepts. Arrivals alone are not demand: acuity, patients already in the department, treatment spaces, boarding, and productivity by role all affect the coverage required each hour.

## Key takeaways

- ED arrivals show repeatable local patterns. In one Danish ED, hour of day explained half the variance in arrivals, but the authors noted that weekly patterns differ between sites ([Hertzum, *Ergonomics Open Journal*, 2016](https://benthamopen.com/FULLTEXT/TOERGJ-9-1)). Use your own data rather than importing another department's peak.
- Emergency medicine tops the AMA's 2025 specialty ranking for burnout at 49.8 percent ([AMA, 2026](https://www.ama-assn.org/press-center/ama-press-releases/ama-physician-burnout-rates-are-falling-specialty-gaps-remain)), against a physician average of 41.9 percent that has [fallen three years running](https://www.ama-assn.org/practice-management/physician-health/physician-burnout-rate-continues-decline-falling-nearly-42).
- In a French study of 40 emergency physicians, three of four tested cognitive abilities declined significantly after a 24-hour shift; no statistically significant decline was detected after a 14-hour night shift ([Persico et al., *Annals of Emergency Medicine*, 2018](https://www.annemergmed.com/article/S0196-0644%2817%2931769-9/fulltext)).
- There is no reliable published benchmark for how often an attending ED schedule changes after publication. Measure local swaps, absences, open shifts, and rule exceptions before treating post-publication churn as a major cost or retention driver.

## What makes ED staffing different?

ED coverage is continuous and unscheduled: US emergency departments recorded 155 million visits in 2022 ([CDC/NCHS National Hospital Ambulatory Medical Care Survey, Cairns et al., 2024](https://www.cdc.gov/nchs/data/nhamcs/web_tables/2022-nhamcs-ed-web-tables.pdf)). When provider capacity falls short, the department may absorb the gap through longer waits, higher left-without-being-seen rates, extra load on the remaining team, or premium coverage that belongs in the [full cost of clinical coverage](https://youshift.com/blog/cost-of-clinical-coverage).

Emergency medicine is predominantly on-site shift work across days, evenings, nights, weekends, and holidays. That makes rotation direction, shift length, recovery, and distribution of undesirable hours recurring design questions.

Calendar patterns are useful but local. The Danish ED above was busiest on Mondays, while the same paper cites different weekly patterns elsewhere. Arrival peaks also vary by population and setting. A department should estimate its own hourly distributions and refresh them for seasonal and service changes.

## How should coverage follow demand?

Start with at least twelve months of local data. Arrivals by hour are one input. Add patients already in the department, acuity, treatment-space availability, boarding, admission patterns, service time, and the productivity of attendings, residents, and APPs over the course of a shift. Choose a service target explicitly; there is no universal 70th- or 80th-percentile staffing rule.

Provider productivity is not flat across a shift, partly because clinicians accumulate active patients, documentation, disposition work, and handoffs. In a retrospective model from one academic ED, staggered resident start times reduced projected mean demand-capacity mismatch from 2.69 to 2.14 patients ([Stenson et al., *Journal of Emergency Medicine*, 2021](https://www.jem-journal.com/article/S0736-4679%2821%2900314-0/fulltext)). The schedule was modeled, not prospectively implemented, and used patients placed in treatment spaces plus local resident-capacity assumptions.

**Figure.** An illustrative hourly ED demand curve with flat and staggered coverage overlaid. Illustrative only. Replace the shape with local demand and capacity data before changing coverage.

## What does ACEP recommend for shift rotation?

[ACEP's policy on emergency physician shift work](https://doi.org/10.1016/j.annemergmed.2017.08.028) recommends circadian-compatible schedules, shifts of 12 hours or less in most settings, regularly scheduled periods of at least 24 hours off, and clockwise rotation from day to evening to night. Its companion [education paper](https://www.acep.org/siteassets/new-pdfs/preps/circadian-rhythms-and-shift-work---prep.pdf), revised in 2026, describes eight to ten hours as ideal for alertness.

ACEP describes two main strategies when a group lacks physicians who consistently work nights: isolated nights that preserve a daytime orientation, or long night rotations, generally a month or more, supported by anchor sleep and minimal daytime obligations. Permanent nights can reduce rotation only when the physician can maintain a stable sleep pattern and is appropriately compensated. A run of four to seven nights followed by an immediate return to days is the pattern ACEP's education paper most clearly advises against.

Programs scheduling residents must also satisfy the separate [ACGME duty-hour rules](https://youshift.com/blog/acgme-duty-hour-scheduling), including limits that do not apply to independent attending schedules.

Much of the comparative evidence on rotation direction comes from nurses rather than emergency physicians, including a cohort of 144 nurses where forward rotation produced less fatigue than backward ([Di Muzio et al., *JAMA Network Open*, 2021](https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2785077)). ACEP's policy is the more relevant basis for an emergency physician schedule.

## Is the 24-hour shift defensible?

In a prospective cross-sectional study of 40 physicians at one French ED, three of four tested cognitive abilities declined significantly after a 24-hour shift compared with a rested condition. The study detected no significant decline after a 14-hour night shift ([Persico et al., *Annals of Emergency Medicine*, 2018](https://www.annemergmed.com/article/S0196-0644%2817%2931769-9/fulltext)). Its size and setting limit generalization, and absence of a significant finding after 14 hours does not establish that every 14-hour shift is safe.

Sleep obtained during the studied shift was not associated with the test result, but the study was not designed to rule out strategic naps as a fatigue countermeasure. The defensible operational conclusion is narrower: avoid routine 24-hour ED physician shifts and do not rely on self-assessed alertness as the only safeguard.

## How much does the schedule contribute to attrition?

Emergency medicine had the highest burnout rate, 49.8 percent, in the AMA's 2025 Organizational Biopsy dataset, compared with 41.9 percent overall ([AMA, 2026](https://www.ama-assn.org/press-center/ama-press-releases/ama-physician-burnout-rates-are-falling-specialty-gaps-remain)). The nearly 19,000 responses came from 106 participating health systems, so this is a large benchmark rather than a census of US emergency physicians.

In a survey of more than 2,000 US physicians at large practices, poor control over patient load, workload, team composition, and clinical schedule was independently associated with burnout ([Sinsky et al., *Annals of Internal Medicine*, 2025](https://www.acpjournals.org/doi/10.7326/ANNALS-24-00884)). The study was not specific to emergency medicine and cannot isolate schedule design from the rest of the work environment.

A 2026 qualitative study of 46 emergency physicians who had left or seriously considered leaving emphasized moral injury, stagnant resources, rigid career paths, and inadequate support for caregiving ([Hua et al., *JAMA Network Open*, 2026](https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2849832)). Flexible scheduling was one proposed response, not a complete explanation. Boarding, staffing levels, organizational culture, compensation, and career structure remain separate retention problems.

## How can a group make distribution auditable?

Start by defining undesirable work and the denominator used to compare it. Raw counts can mislead when clinicians differ in FTE, contract, role, site, shift intensity, compensation, or volunteered assignments. Publish both the rules and the adjusted distribution after each cycle. The same normalization problem appears when groups compare [anesthesia call and late-shift burden](https://youshift.com/blog/anesthesia-coverage-cost).

A manual schedule can be auditable if its inputs and results are visible. Automation can make consistent application easier, but the evidence is limited. In one internal medicine program, first-choice rotation assignment improved after automation, while only 22 of 82 residents completed the fairness survey ([Howard et al., *PLOS ONE*, 2020](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0236952)). A small academic EM group reported high satisfaction with asynchronous self-scheduling in a shared spreadsheet, without a comparison group ([London et al., *JACEP Open*, 2023](https://pmc.ncbi.nlm.nih.gov/articles/PMC9867877/)).

These studies support structured input and transparent results. They do not establish that one scheduling method is superior for every group.

## What should happen after publication?

A posted schedule will change through swaps, illness, leave, and open shifts. Each change can alter rest intervals, rotation direction, consecutive nights, or the distribution of undesirable work. The effect depends on the group's rules and approval process; many departments already use formal workflows, while others rely on messages and manual review.

No peer-reviewed benchmark establishes how often attending ED schedules change or what proportion of changes break a local rule. Measure the number, reason, premium cost, approval time, and rule exceptions in your own department. Apply the same explicit checks to a proposed change that the group uses at publication, whether the review is manual or automated.

## What to measure

Most departments measure whether the schedule got published. A more useful set includes what happened after:

| Metric | What it tells you |
|---|---|
| Acuity-adjusted patients in department against provider capacity, by hour | Whether coverage follows active workload rather than arrivals alone |
| Door-to-provider, LWBS, boarding, and treatment-space constraints | Whether the apparent staffing gap is an input, throughput, or output problem |
| Night, weekend, and holiday distribution normalized for FTE and contract | Whether undesirable work follows the group's stated rules |
| Premium and locum shifts used | The incremental cost of uncovered demand |
| Rest intervals, rotation direction, and consecutive nights | Whether worked assignments follow the adopted circadian rules |
| Post-publication changes and rule exceptions | The amount and consequence of local schedule churn |

## Frequently asked questions

**How many consecutive nights are too many?**
ACEP specifically advises against four to seven consecutive nights followed by an immediate return to days. Its alternatives are isolated nights or substantially longer night rotations supported by anchor sleep and minimal daytime obligations. Group size, physician preference, age, recovery time, and local workload still matter.

**Should we rotate forward or backward?**
ACEP recommends forward rotation, meaning day to evening to night, even with intervening days off. Comparative research is stronger in nurses than in emergency physicians.

**What is the most overlooked failure mode?**
It depends on the department. Common blind spots include staffing to arrivals instead of active workload, failing to distinguish boarding from provider capacity, and approving changes without checking the group's rest and distribution rules.

## Where to start

Build an hourly view that combines arrivals, active census, acuity, boarding, treatment spaces, and provider capacity by role. Compare it with door-to-provider and LWBS performance before changing shift starts. Adopt explicit night-rotation and recovery rules from ACEP, adjusted to local contracts and workload. Finally, audit post-publication changes for several cycles before deciding whether churn is a material clinical, retention, or cost problem.

---

YouShift builds emergency department schedules from a group's own rules, so circadian constraints, rest intervals, and equity targets are encoded before the schedule is published, and every swap, absence, and open shift afterward is re-checked against those same rules instead of eyeballed. It also shows night, weekend, and premium distribution by clinician, so fairness is something the group can see rather than take on faith. More than a thousand physicians schedule on it across the United States and Europe.

**See what your next month looks like generated.** Tell us your coverage rules and we will build it.

## Sources

### Clinical operations and workload

1. Hertzum, "Patterns in Emergency-Department Arrivals and Length of Stay," *Ergonomics Open Journal*, 2016. [Link](https://benthamopen.com/FULLTEXT/TOERGJ-9-1)
2. American Medical Association, "Physician Burnout Rate Continues Decline," 2026. [Link](https://www.ama-assn.org/practice-management/physician-health/physician-burnout-rate-continues-decline-falling-nearly-42)
3. American Medical Association, "Physician Burnout Rates Are Falling, Specialty Gaps Remain," 2026. [Link](https://www.ama-assn.org/press-center/ama-press-releases/ama-physician-burnout-rates-are-falling-specialty-gaps-remain)
4. Persico et al., "Influence of Shift Duration on Cognitive Performance of Emergency Physicians," *Annals of Emergency Medicine*, 2018. [Link](https://www.annemergmed.com/article/S0196-0644%2817%2931769-9/fulltext)
5. Cairns et al., "National Hospital Ambulatory Medical Care Survey: 2022 Emergency Department Summary Tables," CDC/NCHS, 2024. [Link](https://www.cdc.gov/nchs/data/nhamcs/web_tables/2022-nhamcs-ed-web-tables.pdf)
6. Stenson et al., "Understanding Demand and Capacity Mismatch," *Journal of Emergency Medicine*, 2021. [Link](https://www.jem-journal.com/article/S0736-4679%2821%2900314-0/fulltext)
7. American College of Emergency Physicians, "Emergency Physician Shift Work," *Annals of Emergency Medicine*, 2017. [Link](https://doi.org/10.1016/j.annemergmed.2017.08.028)

### Schedule design and work control

8. American College of Emergency Physicians, "Circadian Rhythms and Shift Work," PREP, 2026. [Link](https://www.acep.org/siteassets/new-pdfs/preps/circadian-rhythms-and-shift-work---prep.pdf)
9. Di Muzio et al., "Comparison of Sleep and Attention Metrics Among Nurses," *JAMA Network Open*, 2021. [Link](https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2785077)
10. Sinsky et al., "Association of Work Control With Burnout," *Annals of Internal Medicine*, 2025. [Link](https://www.acpjournals.org/doi/10.7326/ANNALS-24-00884)
11. Hua et al., "Perspectives on Attrition Among Emergency Physicians," *JAMA Network Open*, 2026. [Link](https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2849832)
12. Howard et al., "Implementation of an Automated Scheduling Tool," *PLOS ONE*, 2020. [Link](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0236952)
13. London et al., "Internet-Based Self-Scheduling," *JACEP Open*, 2023. [Link](https://pmc.ncbi.nlm.nih.gov/articles/PMC9867877/)