Exercise tied to improved sleep quality in patients with mental illness

  • Lederman O & al.
  • J Psychiatr Res
  • 3 Nov 2018

  • curated by Antara Ghosh
  • UK Clinical Digest
Access to the full content of this site is available only to registered healthcare professionals. Access to the full content of this site is available only to registered healthcare professionals.


  • Exercise is associated with a large sleep-improving effect in people living with severe mental illness.
  • Findings support inclusion of exercise regimen in addition to current treatment practices for improved management of poor sleep amongst individuals with mental illness.

Why this matters

  • Findings highlight importance of exercise in improving health outcomes in people living with mental illness.
  • Further research determining the efficacy of exercise on sleep in people with psychiatric illness and exploring the effects of exercise intervention elements such as modality, frequency, intensity and delivery settings is required.

Study design

  • This systematic review and meta-analysis was conducted after searching major electronic databases from inception until June 2018 to include 8 randomised controlled trials that evaluated sleep quality in patients with severe mental illness.
  • 463 participants with severe mental illness were allocated to an exercise intervention and 736 control participants received treatment as usual, waitlist or non-active interventions.
  • Funding: None disclosed.

Key results

  • Exercise had large beneficial effects on overall sleep quality in participants assigned to exercise vs control participants (Hedges’ g, 0.73; 95% CI, 0.18-1.28; I2, 91.15%).
  • Exercise significantly improved sleep quality in studies of individuals with depression (Hedges’ g, 1.83; 95% CI, 0.17–3.49; I2, 95.83%).
  • Significant improvement in sleep was observed in studies supervised by a qualified exercise professional (Hedges’ g, 1.003; 95% CI, 0.14–1.87; I2, 93.05%) but not in studies conducted by less qualified supervisors (Hedges’ g, 0.16; 95% CI, 0.37–0.69; I2, 53.78%).


  • Heterogeneity among studies.
  • Small number of studies represented each clinical group.