Virtual reality-based cognitive remediation versus virtual reality control in people with mood or psychosis spectrum disorders in Denmark: a single-centre, double-blind, randomised controlled trial

Background: There is a paucity of real-life-like cognitive training programmes to bolster the transfer of treatment-related cognitive gains to daily-life functioning. We aimed to investigate the effects of 4-week, intensive virtual reality-based cognitive remediation therapy (VR-CRT) involving daily-life challenges versus an active virtual reality control treatment on cognition and functioning in patients with mood or psychosis spectrum disorders. Methods: This was a single-centre, double-blind, parallel-group, randomised controlled trial at the Psychiatric Centre Copenhagen in Denmark. Clinically stable outpatients aged 18–55 years with an ICD-10 diagnosis of unipolar disorder, bipolar disorder, or a psychosis spectrum disorder, and with clinically relevant objective and subjective cognitive impairment were included. Participants were randomly assigned (1:1) to 4 weeks of VR-CRT or virtual reality control treatment and assessed at baseline, treatment completion (5 weeks) and follow-up (17 weeks). Randomisation used the REDCap module with block randomisation (block sizes four to eight), stratified by age (<35 vs ≥35 years) and diagnosis (mood disorder vs psychosis spectrum disorder), with allocation concealed from the enrolling author. The primary outcome was a global functional cognitive capacity score on the Cognition Assessment in Virtual Reality test at week 5. This trial was registered with ClinicalTrials.gov, NCT06038955, and is completed. Findings: Between Oct 10, 2022, and Aug 9, 2024, 103 candidates were assessed for eligibility, of whom 62 participants were enrolled and randomly assigned (VR-CRT group n=31 and control group n=31). Of the 30 participants commencing VR-CRT, 28 (93%) completed treatment per protocol. At week 5, the mean Z score improvement in the global functional cognitive capacity score was 1·5 (SD 0·6) in the VR-CRT group and 0·4 (0·8) in the virtual reality control group (treatment effect 0·98 [95% CI 0·65–1·32]; p<0·0001; d=1·55). This effect was maintained at follow-up at week 17 (mean Z score improvement 1·6 [0·5] vs 0·6 [0·7]; treatment effect 0·91 [0·61–1·21]; p<0·0001; d=1·53). The intervention was well tolerated with no treatment-related serious adverse events. Interpretation: Improvement in functional cognitive capacity was significantly greater in the VR-CRT group than the control group, and the intervention was well tolerated. Our findings suggest that embedding cognitive strategy training in immersive virtual reality can enhance transfer to real-world functioning, offering a feasible, engaging solution for cognitive rehabilitation in psychiatry. Funding: TrygFonden, Axel Muusfeldts Foundation, Jascha Foundation, Ivan Nielsen Foundation, and Familien Hede Nielsen Foundation.