REF-974 · Stutz et al., 2018 · Sports Medicine, Vol. · Paper · PubMed

Effects of Evening Exercise on Sleep in Healthy Participants: A Systematic Review and Meta-Analysis

A-fairly strong evidenceMeasured on a mixed groupWritten before October 2026, being re-checked

Guidelines currently tell people not to exercise in the evening, on the grounds that it may disturb sleep. This systematic review measured how much evening exercise changes sleep and whether intensity or duration alter the response.

Of 11,717 references identified, 23 studies were pooled with random-effects models; each compared healthy adults' sleep after one evening exercise session with a no-exercise control. Evening exercise slightly lengthened rapid eye movement latency, slightly raised slow-wave sleep and reduced stage 1 sleep.

The authors conclude that the reviewed studies give no support to the hypothesis that evening exercise harms sleep — if anything they point the other way — though vigorous exercise that ends ≤ 1 h before bedtime might impair sleep-onset latency, total sleep time and sleep efficiency.

Participants

Healthy adults in 23 included studies; the pooled number of participants is not given in the abstract.

Protocol

Systematic search of PubMed, Cochrane, EMBASE, PsycINFO and CINAHL; 11,717 references, 23 included.
Inclusion: healthy adults, with sleep measured after one session of evening physical exercise and compared with a no-exercise control.
Random-effects models; moderator analyses (temperature at bedtime, physical stress, exercise mode).

Key Results

Versus control: rapid eye movement latency rose by 7.7 min (p = 0.032), slow-wave sleep by 1.3 percentage points (p = 0.041), and stage 1 sleep fell by 0.9 pp (p = 0.001).
Higher temperature at bedtime: linked to lower sleep efficiency (b = - 11.6 pp; p = 0.020) and to more time awake after sleep onset (b = + 37.6 min; p = 0.0495; WASO).
Higher physical stress (how intense the exercise was relative to the person's baseline physical activity) went with 3.2 pp lower sleep efficiency (p = 0.036) and 21.9 min more WASO (p = 0.044).
Running versus cycling: less WASO with running (- 12.7 min; p = 0.037).
Removing a single study made every significant moderator effect vanish.

Practical Application

A meta-analysis of 23 studies in healthy adults found that a single evening exercise session did not harm that night's sleep compared with no exercise — slow-wave sleep rose by 1.3 percentage points and stage 1 sleep fell by 0.9 — while the authors caution that vigorous exercise that ends ≤ 1 h before bedtime might impair sleep-onset latency, total sleep time and sleep efficiency. To "is it bad to train in the evening?": no, as a rule; the case to avoid is a hard session that finishes within the last hour before bed.
The moderators point the same way but are fragile: going to bed still hot and a session much harder than the person's usual activity were linked to lower sleep efficiency and more time awake, and all of these effects disappeared after removal of one study.
Inference, not tested here: when an evening session is the only slot, keep it easy to moderate, or finish hard work early enough to cool down before bed.
Not from this paper; the base's own practice: where hard sessions sit in the athlete's day is decided in SYNTH-07.

Limitations

Single sessions only: no evidence here on habitual evening training.
Healthy adults, not athletes specifically; the pooled sample size is not in the abstract.
The main effects are small, and the moderator findings rested on one study.
The caution about vigorous exercise ending ≤ 1 h before bedtime is stated as "might be impaired"; no pooled number is given for it in the abstract.
Written from the abstract and record metadata only (full text not read).
How this entry was checked and changed
Why this level

systematic review and meta-analysis of 23 controlled studies; small pooled effects, and every moderator effect disappeared after removal of one study