Background: The aim of this study was to examine feasibility of trial processes and group-based, structured exercise training in patients with first-episode psychosis.

Methods: Twenty-five patients with first-episode psychosis took part in a two-arm randomised feasibility trial. They were individually randomised (1:1) via a computer-generated randomisation sequence and allocated to either an exercise intervention group (INT) or a control group (CON). Patients allocated to INT completed a physical exercise training programme at moderate-to-vigorous intensity, 1 h three times weekly for 8 weeks. CON patients were encouraged to continue their usual level of activity and were offered the training programme after 8 weeks. Primary outcomes included screening rate, recruitment rate, retention rate, attendance and adverse events. Secondary outcomes included heart rate response during training, cardiovascular health (VO2max, resting heart rate, blood pressure), body composition (muscle mass, fat percentage), muscle strength (sit-to-stand, grip strength, jump height) and balance.

Results: Recruitment lasted 6 weeks and 86 out of 324 patients (27%) were screened, 71 of whom (83%) were deemed eligible. Twenty-five (35%) accepted inclusion (mean age 25.5; mean body mass index 25.1) and were subsequently randomised (INT = 13, CON = 12). Retention of patients was 76% and 52% at the 8-week and 16-week follow-up, respectively. Attendance was 43% (min. 9%, max. 96%). No significant changes were observed between groups in secondary physiological outcome measures.

Conclusions: Feasibility was challenged by limited recruitment and retention rates, suggesting that modifications are required if a large-scale randomised controlled trial is to be conducted. Recommendations for modifications are presented and discussed.

TRIAL REGISTRATION: Clinicaltrials.gov, NCT03409393 . Retrospectively registered.