Background Current recommendations advise against exercising in the evening because of potential adverse effects on sleep.
Objectives The aim of this systematic review was to investigate the extent to which evening exercise affects sleep and whether variables such as exercise intensity or duration modify the response.
Methods A systematic search was performed in PubMed, Cochrane, EMBASE, PsycINFO, and CINAHL databases. Studies evaluating sleep after a single session of evening physical exercise compared to a no-exercise control in healthy adults were included. All analyses are based on random effect models.
Results The search yielded 11,717 references, of which 23 were included. Compared to control, evening exercise significantly increased rapid eye movement latency (+ 7.7 min; p = 0.032) and slow-wave sleep (+ 1.3 percentage points [pp]; p = 0.041), while it decreased stage 1 sleep (− 0.9 pp; p = 0.001). Moderator analyses revealed that a higher temperature at bedtime was associated with lower sleep efficiency (SE) (b = − 11.6 pp; p = 0.020) and more wake after sleep onset (WASO; b = + 37.6 min; p = 0.0495). A higher level of physical stress (exercise intensity relative to baseline physical activity) was associated with lower SE (− 3.2 pp; p = 0.036) and more WASO (+ 21.9 min; p = 0.044). Compared to cycling, running was associated with less WASO (− 12.7 min; p = 0.037). All significant moderating effects disappeared after removal of one study.
Conclusion Overall, the studies reviewed here do not support the hypothesis that evening exercise negatively affects sleep, in fact rather the opposite. However, sleep-onset latency, total sleep time, and SE might be impaired after vigorous exercise ending ≤ 1 h before bedtime.