Junior doctor rotas fail a fatigue test the rail industry has used for years
Everyone has a tired doctor story. Mine is unremarkable, which is rather the point: a friend, foundation year two, describing the drive home after her fourth consecutive night shift. She said she did not remember most of it. She said it the way you would mention the weather, because in her world it was the weather. Everyone drove home like that. It was simply what the rota produced.
What stayed with me was not the danger, though the danger is real. It was the resignation. The rota was treated as a natural phenomenon, like tides, rather than what it actually is: a designed artefact, drawn up by a person, checkable before anyone works it.
In 2019, a research team at a large UK teaching hospital demonstrated exactly that. They took 95 anonymised foundation doctor rotas across general medicine, general surgery and emergency medicine, and ran every shift through the Health and Safety Executive's Fatigue and Risk Index, the same tool the rail and nuclear industries use to assess shift patterns.
Nearly half of all shifts showed increased risk of fatigue-related error and an increased probability of high sleepiness. There was wide variation between specialties. And the authors noted something worse: because they scored the rotas as published rather than as actually worked, the true picture was almost certainly an underestimate. No rota survives contact with a winter on-call.
The study's conclusion was not that doctors should work less, or that any particular rota was illegal. Every rota assessed was, presumably, contract-compliant. The conclusion was that fatigue can and should be considered at the point of rota design, and that a tool already exists to do it.
Compliant is not the same as safe
I spent over six years building compliance software before this, and if that career taught me one sentence, it is this one: passing the check is not the same as being safe. I watched organisations treat a green dashboard as the end of the enquiry rather than the beginning. The dashboard answers the question it was built to answer, and not one question more.
NHS rotas are checked against contractual rules: maximum hours, rest requirements, limits on consecutive shifts. Those checks matter, but they answer a legal question, not a physiological one. Two rotas can both pass contract checks while one leaves a doctor driving home at 9am after a fourth consecutive night and the other does not. The contract cannot tell them apart. A fatigue model can.
The FRI asks the physiological question. It models cumulative sleep opportunity, time of day, shift length and rest between duties, and returns two numbers: the probability a person on that pattern will experience high sleepiness, and the relative risk of a fatigue-related error. It was built for safety-critical industries where the cost of a tired human is measured in lives. Medicine qualifies.
Why almost nobody runs it
The HSE tool was a free Excel spreadsheet, and I say this with affection for what it contains: a 2006 spreadsheet with macros. It got worse: the HSE withdrew it from its own website in June 2021, partly because the version of Excel it ran on could no longer be supported and partly because its design needed improvement to help people understand its outputs and limitations. The methodology remains sound and published. The delivery vehicle was retired, and nothing official replaced it. Anyone who has tried to feed a full medical rota into a surviving copy understands why the 2019 study needed a research team. Shift-by-shift manual entry, per person, security warnings about macros, and output landing in cells that then need interpreting against thresholds published in a rail industry appendix.
A rota coordinator managing dozens of doctors across multiple templates does not have a spare research team. So the assessment does not happen, and fatigue stays where it has always lived in the NHS: in exception reports filed after the fact, and in the guardian of safe working's quarterly summary of things that already went wrong. The information arrives, but it arrives late, the way a smoke alarm would be less useful if it went off the day after the fire.
What good would look like
Imagine the check ran the other way around. A draft rota goes in before publication. Every shift comes back scored. The handful that breach sensible thresholds are flagged, with the specific pattern feature causing the problem: the short gap after nights, the run of long days, the early start after a late finish. The coordinator adjusts, re-scores, publishes. Ten minutes, not a research project.
That is the entire premise of RotaPulse. It runs the HSE Fatigue and Risk Index on a whole rota in the browser, shows which shifts carry the risk and why, and lets you compare a redesign against the original before anyone works it. The methodology is the published, peer-evaluated one. The difference is that it fits inside the job of the person who actually builds the rota.
My friend's drive home was not a mystery and it was not inevitable. It was visible in the pattern, weeks in advance, to anyone with the means to look. The means should not be the hard part.
Score one of your rota templates with the free calculator and see which shifts light up.
Sources
- Cumber E, McCulloch C, Sy I, et al. Can a tool developed for industry be used to assess fatigue risk in medical rotas? A pilot study of foundation doctors' rotas in a tertiary centre. BMJ Open 2019;9:e023470
- HSE Research Report RR446, The development of a fatigue / risk index for shiftworkers (2006), the methodology and calculator the study used.
- HSE, Managing shift work (HSG256) via the HSE fatigue topic page
- ORR, Managing Rail Staff Fatigue, Appendix D, source of the fatigue index and risk index interpretation and the indicative thresholds.
- CIEHF, Fatigue Risk Management resources, including the white paper on fatigue risk management for health and social care.