The evidence-based night shift rulebook: three rules to enforce
A 15-researcher expert consensus distilled night-shift safety into three schedulable limits, built on admittedly thin evidence.
Reviewed against primary sources on July 19, 2026 by the Soon operations research team
The evidence in one line
A 2020 expert consensus of 15 shift-work researchers (Garde et al., 2020) distilled night-work safety into three scheduling limits: no more than 3 consecutive night shifts, at least 11 hours between shifts, and no more than 9 hours per shift. The panel was explicit that these numbers rest on limited existing literature with major knowledge gaps, so treat them as a defensible default rather than a proven formula. They are also constraints a scheduling tool can enforce automatically, which is what makes them practical.
The three rules the panel agreed on
In 2020, a group of 15 shift-work researchers published an expert-consensus paper (Garde et al., 2020) that translated the field's messy evidence into a short operating standard for night work. Their recommendation has three parts: no more than 3 consecutive night shifts, at least 11 hours between shifts, and no more than 9 hours per shift. Each of the other articles in this library circles one lever or one health outcome. This capstone puts the levers together in the form the panel actually proposed.
What makes the recommendation useful for schedulers is its shape. It is three ceilings, not a philosophy. A ceiling of 3 nights, a floor of 11 hours, and a ceiling of 9 hours per shift are the kind of limits you can write into a policy and check against a published roster before anyone works it.
Consensus is a starting point, not proof
The most important sentence in the paper is the one that undercuts its own confidence. The authors state plainly that the three limits rest on limited existing literature with major knowledge gaps. This is a consensus recommendation, not a law, and not the output of a randomized trial. It is what 15 experts judged reasonable given incomplete evidence.
Read that caveat the way the panel intended. Limited evidence does not mean the limits are wrong, and it does not mean night work below these thresholds is proven safe. It means the numbers are the best available default, open to revision as better studies arrive. Treating them as hard science would overstate the case; treating them as arbitrary would ignore the closest thing the field has to agreement.
Writing the rules as constraints a scheduler can enforce
The three limits map cleanly onto checks that scheduling software can run automatically. Consecutive-night count, rest interval between consecutive shifts, and single-shift length are all computable from a roster before it is published. A tool can flag a run longer than 3 consecutive nights, a turnaround shorter than 11 hours, or a shift longer than 9 hours, and hold the schedule until someone resolves it.
The advantage of constraints over guidelines is that constraints do not depend on a tired planner remembering them late at night. Encode the ceilings once and every future roster inherits them. That is the argument for treating the Garde 2020 standard as configuration rather than as a poster on the break-room wall.
The exceptions the panel named
The consensus did not claim the three rules fit every workplace. The authors noted exceptions, including settings such as oil rigs and isolated workplaces, where the logistics of getting people to and from the site change the calculation. In those environments a compressed run of shifts may be the lesser risk, and a blanket 3-night ceiling may not apply.
The panel also pointed to stricter guidance for pregnant workers, whose risk profile differs enough to warrant tighter limits than the general standard. Build the three rules as your default, but leave room in the policy for the named exceptions rather than pretending one ceiling covers everyone.
What this means for your schedule
- Cap consecutive night shifts at 3 and let the schedule refuse a longer run.
- Require at least 11 hours between the end of one shift and the start of the next.
- Keep night shifts at or under 9 hours rather than stretching them to 12 for coverage.
- Encode all three limits as automatic checks so no roster publishes in violation.
- Carve out documented exceptions for isolated sites and stricter limits for pregnant workers instead of applying one ceiling everywhere.
The business case
The three limits from Garde 2020 give you a defensible, written safety standard that predates any incident, which matters when a regulator or insurer asks how your rosters were designed.
Because the limits are constraints a scheduling tool can enforce, compliance costs almost nothing per roster once configured, unlike manual review that scales with headcount.
Adopting a published expert standard also signals to shift workers that fatigue risk is managed by policy, not left to chance, which supports retention on hard-to-staff night teams.
Frequently asked questions
- Where do the three night-shift rules come from?
- They come from an expert-consensus paper by 15 shift-work researchers (Garde et al., 2020). The panel recommended no more than 3 consecutive night shifts, at least 11 hours between shifts, and no more than 9 hours per shift. It is a consensus recommendation, not a legal requirement.
- Are the three rules scientifically proven?
- No. The authors were explicit that the limits rest on limited existing literature with major knowledge gaps. They represent the best available expert judgment, not the result of a randomized trial, and staying under them is not proven to make night work safe. Treat them as a sound default that may change as evidence improves.
- Do the rules apply to every workplace?
- Not entirely. The panel named exceptions such as oil rigs and isolated workplaces, where travel logistics can justify a different pattern. It also pointed to stricter guidance for pregnant workers. Use the three limits as your default and document where a named exception applies.
- How can scheduling software enforce these limits?
- All three are computable from a roster before it is published: the count of consecutive nights, the gap between shifts, and the length of each shift. A tool can block a run beyond 3 consecutive nights, a turnaround under 11 hours, or a shift over 9 hours, and hold the schedule until it is fixed. That turns the standard into a constraint rather than a guideline someone has to remember.
Sources
Every figure on this page is drawn from a cited primary source and checked against the original publication.
Garde, A. H., Harma, M., Bjorvatn, B., Bonde, J. P., Hansen, J., Hansen, A. M., et al. (2020). How to schedule night shift work in order to reduce health and safety risks. Scandinavian Journal of Work, Environment & Health, 46(6), 557โ569. https://pmc.ncbi.nlm.nih.gov/articles/PMC7737811/
Expert-consensus paper (workshop of 15 shift-work researchers)
None of the studies cited here evaluated Soon.They examine scheduling practices, shift patterns, and working hours as studied by independent researchers, so their findings describe what those practices are associated with, not what any particular software produces.
This article summarizes published research for scheduling and operations decisions. It is not medical advice. Individual health questions belong with a qualified clinician.
Keep reading
Does the 11-hour rest rule between shifts work?
The one shift-design change with randomized-trial evidence behind it: keep at least 11 hours between the end of one shift and the start of the next.
Read →How many night shifts in a row is safe?
Injury risk rises with each night worked in a row and peaks on the fourth, which is why the consensus caps night blocks at three.
Read →8 vs 10 vs 12-Hour Shifts: What Is the Safety Tradeoff?
Accident risk climbs the longer a shift runs, so the choice is a real tradeoff to manage, not a length to ban.
Read →Shift pattern library
Related resource
Open →Staffing coverage math
Related resource
Open →