The Early Youth Engagement (EYE-2) intervention in first episode psychosis services: a pragmatic cluster RCT and cost-effectiveness evaluation

Background Early Intervention in Psychosis (EIP) services improve outcomes for young people but approximately 30% disengage.

Aims We aimed to test whether a new motivational engagement intervention would prolong engagement and whether it was cost-effective.

Methods We conducted a multi-centre, single-blind, parallel-group, cluster randomized controlled trial, involving 20 EIP teams in 5 UK National Health Service sites. Teams were randomised using permuted blocks stratified by NHS Trust. Participants were all young people (14-35 years) presenting with a first episode of psychosis between May 2019 and July 2020 (N=1,027). We compared the novel Early Youth Engagement (EYE-2) intervention plus standardised EIP (sEIP) to sEIP alone. The primary outcome was time to disengagement over 12-26 months. Economic outcomes were mental health costs, societal costs, and socio-occupational outcomes over 12-months. Assessors were masked to treatment allocation for primary disengagement and cost-effectiveness outcomes. Analysis followed intention to treat principles. The trial was registered with ISRCTN 51629746

Results Disengagement was low at 15.9% overall in standardised stand-alone services. The adjusted hazard ratio for EYE-2+sEIP (n=652) versus sEIP alone (n=375) was 1.07 (95% CI: 0.76-1.49; p=0.713). The health economic evaluation indicated lower mental health care costs linked to reductions in unplanned mental health care with no compromise to clinical outcomes, some evidence for lower societal costs and more days in education, training, employment, and stable accommodation in EYE-2.

Conclusions We found no evidence that EYE-2 increased time to disengagement, but some evidence for cost-effectiveness. This is the largest study to date reporting positive engagement, health and cost outcomes in a total EIP population sample. Limitations included high loss to follow-up for secondary outcomes, and low completion of societal and socio-occupational data. COVID-19 impacted fidelity and implementation. Future engagement research should target engagement to those with greatest need, including inpatients and those with socio-occupational goals.