Exercise may help shift workers sleep, evidence thin

Shift work disrupts circadian rhythms and is linked with poor sleep, cardiometabolic risk and cognitive impairment, a combination that matters most in safety-critical jobs where a lapse in vigilance has consequences. Exercise is already known to improve sleep and cognition in the general population, but shift workers are a different case: their sleep opportunity is fragmented, their body clock is repeatedly shunted, and their spare energy is limited. That leaves an open question about whether structured training helps people whose problem is scheduling rather than inactivity alone. This review set out to gather the randomised evidence and see how far it actually goes.

Following PRISMA and Cochrane guidance, the authors searched six databases from inception to January 2025 for randomised controlled trials of adult shift workers doing structured exercise, including aerobic, resistance, combined, HIIT and in-shift activity breaks. Primary outcomes were sleep quality, quantity and continuity measured with the Pittsburgh Sleep Quality Index, actigraphy or polysomnography and the Karolinska Sleepiness Scale, alongside cognition and alertness measured with the Psychomotor Vigilance Task. Risk of bias was rated with RoB 2.0 and methodological quality with PEDro. Ten RCTs totalling 420 participants qualified, 60% of them set in healthcare, and because interventions and outcome measures varied so much the results were synthesised narratively rather than pooled.

Eight of the ten trials reported significant improvements in at least one sleep outcome — PSQI score, total sleep time, sleep efficiency or wake after sleep onset — although the size of the effects and their clinical relevance were inconsistent across studies. Three trials showed improvements in alertness, reaction time and short-term memory, and these tended to be the ones where exercise was timed after a shift or delivered as supervised sessions in the workplace. Six studies contributed mechanistic data pointing to circadian phase shifting, improved autonomic balance measured by heart rate variability, and reduced inflammatory markers. Against that, 80% of the trials carried some concerns or a high risk of bias, and fatigue plus irregular rosters made adherence a recurring problem.

The review extends findings from the general population into a group whose sleep problem is structural, and the direction of travel is consistent rather than conclusive. Because no meta-analysis was possible, there is no pooled effect size to compare against sleep hygiene or light-based approaches, and the trials were mostly small and mostly in healthcare settings. Open questions include which modality works best, whether timing relative to the shift is decisive, and whether benefits persist beyond the intervention period. The authors are explicit that adequately powered trials with standardised outcomes and more varied industries are needed before guideline-level recommendations are possible.

For anyone working rotating or night shifts, this is a reason to think structured exercise is worth considering as part of managing sleep and daytime alertness, not a demonstration that it fixes shift-work sleep loss. The pattern in the trials suggests that when and where the training happens may matter as much as the training itself, with post-shift and supervised workplace sessions showing up most often among the positive cognitive results. The evidence base is small, inconsistent in effect size and largely drawn from healthcare workers, so individual results will vary. Treat it as an early and promising signal rather than a settled protocol.