File(s) under permanent embargo

Reason: This item is currently closed access.

An explanatory randomised controlled trial testing the effects of improving sleep on mental health: an interventionist-causal model approach with mediation analysis

journal contribution
posted on 11.01.2018, 14:28 by Traolach S. Brugha, Daniel Freeman, Colin Espie, et al
Background: Sleep difficulties may be a contributory causal factor in the occurrence of mental health problems. If this is true, improving sleep should benefit psychological health. Our primary objective was to determine whether treating insomnia leads to a reduction in paranoia and hallucinations. The effects on other mental health outcomes were also tested. Methods: For our parallel group, single blinded, explanatory randomised controlled trial (called OASIS), we randomised university students with insomnia to digital cognitive behaviour therapy for insomnia or usual practice. Online assessments were at 0, 3, 10 (post-treatment), and 22 weeks. The primary outcomes were insomnia, paranoia, and hallucinatory experiences. The secondary outcomes included anxiety, depression, nightmares, and psychological well-being. The trial was registered (ISRCTN61272251). Findings: 3,755 people were randomised between 5th March 2015 to 17th February 2016. Compared to usual practice, the sleep intervention at 10 weeks reduced insomnia, adjusted difference=4.78 (95% C.I. 4.29;5.26), p<.0001, d=1.11, paranoia, adjusted difference=-2.22 (95% C.I. -2.98;-1.45), p<.0001, d=0.19, and hallucinations, adjusted difference=-1.58 (95% C.I. -1.98;-1.18), p<.0001, d=0.24. Insomnia was a mediator of change in paranoia and hallucinations. There were improvements in other mental health outcomes. No adverse events were reported. Interpretation: This is the largest randomised controlled trial of psychological intervention for a mental health problem. It provides strong evidence that insomnia is a causal factor in the occurrence of psychotic experiences and other mental health problems. Whether the results generalise beyond a student population requires testing. The treatment of disrupted sleep may require a higher priority in mental health provision.



Lancet Psychiatry, 2018, TBA

Author affiliation

/Organisation/COLLEGE OF MEDICINE, BIOLOGICAL SCIENCES AND PSYCHOLOGY/School of Medicine/Department of Health Sciences


AM (Accepted Manuscript)

Published in

Lancet Psychiatry







Acceptance date


Publisher DOI

Publisher version



Usage metrics