Exercise eases the broken sleep of perimenopause

Sleep often deteriorates through perimenopause, the transition years around the final menstrual period, and disturbed sleep in this window feeds into mood, energy and long-term health. Whether exercise genuinely improves sleep for perimenopausal women has been controversial, with individual studies pointing in different directions. Exercise is cheap, safe and widely accessible, so a clear answer matters more than for most interventions. This meta-analysis pooled the randomised evidence and, unusually, asked not just whether exercise works but which formats, frequencies and durations work best.

Two researchers independently searched seven databases, including PubMed, EMBASE, the Cochrane Library, Web of Science and three Chinese databases, using the PICOS framework, and assessed study quality with the Cochrane evaluation manual. Twelve randomised controlled trials involving 1493 perimenopausal women were included. The exercise interventions covered yoga, walking, fitness Qigong and aerobic exercise. Subgroup analyses then examined how the intervention type, the programme length, the weekly frequency and the duration of each session related to the size of the sleep improvement.

Pooled across the trials, exercise effectively improved sleep in perimenopausal women, with a standardised effect of 0.44 in favour of exercise, and significant benefits for both sleep quality and insomnia symptoms. The subgroup analysis identified a best-performing recipe: fitness Qigong, a gentle traditional Chinese practice, delivered over 10 to 12 weeks, more than 3 times a week, in sessions of 30 to 60 minutes. Other formats, including yoga, walking and aerobic exercise, also contributed to the overall positive result.

This pooling pushes a previously contested question towards a positive answer: exercise does improve sleep in the perimenopausal years. The moderate size of the effect is consistent with exercise research in other groups. The subgroup finding favouring Qigong is intriguing, especially since the winning format is gentle rather than vigorous, but subgroup comparisons in a 12-trial pool are the least certain part of any meta-analysis. Head-to-head trials comparing formats directly, and studies with objective sleep measurement, are the obvious next steps.

For women in the menopause transition whose sleep has slipped, this is a reason to see regular movement as a genuine, evidence-backed option rather than folk advice. The finding that a gentle practice ranked best is worth noting: the benefit did not depend on hard training, and sessions in the 30 to 60 minute range several times a week were the pattern behind the strongest results. As ever with pooled subgroups, the exact recipe is less certain than the headline. The reliable part is simple: consistent exercise, better sleep.