BackgroundThe transition from prison to community is difficult for prisoners with mental illness. Critical time intervention (CTI) is designed to provide intensive support to meet health, social care and resettlement needs through close working between client and key worker pre, and up to 6 weeks post, release.ObjectivesTo establish whether or not CTI is effective in (1) improving engagement of discharged male prisoners who have mental illness with community mental health teams (CMHTs) and (2) providing practical support with housing, finance and re-establishing social networks.Trial designA multicentre, parallel-group randomised controlled trial, with follow-up at 6 weeks and at 6 and 12 months. A subset of prisoners and case managers participated in a complementary qualitative study.SettingEight English prisons.ParticipantsOne hundred and fifty adult male prisoners, convicted or remanded, cared for by mental health in-reach teams and diagnosed with severe mental illness, with a discharge date within 6 months of the point of recruitment.InterventionParticipants were randomised to either the intervention or the control (treatment as usual). The intervention group was assigned a case manager who assessed mental and physical health before and following release, made appropriate links to health, housing and financial services and supported the re-establishment of family/peer contact.OutcomeThe primary outcome measure was engagement with a CMHT 6 weeks post discharge. Secondary outcomes included contact with mental health services at 6 and 12 months. A health economic evaluation was undertaken using service contact at the follow-up time points. We were unable to assess the intervention’s effect on reoffending and longer-term health-care use because of study delays.ResultsOne hundred and fifty prisoners were recruited: 72 were randomised to the intervention and 78 were randomised to the control. Engagement with teams at 6 weeks was 53% for the intervention group compared with 27% for the control group [95% confidence interval (CI) 0.13% to 0.78%;p = 0.012]. At 6 months’ follow-up, intervention participants showed continued increase in engagement with teams compared with control participants (95% CI 0.12% to 0.89%;p = 0.029); there were no significant differences at 12 months. Increased engagement resulted in higher levels of service use and costs for the intervention than for the control. Qualitative data showed the intervention group reporting better continuity of care and improved access to services.ConclusionThe intervention significantly improved contact with services at 6 weeks, although at a higher cost than the control. This is important as, in the days and weeks following release, recently released individuals are at a particularly high risk of suicide and drug overdose. Further research is required to establish how teams can better maintain contact with clients when the intervention ends.Future workFurther studies are indicated for groups with different needs, for example women, young prisoners and those in police custody, and at other transition points, for example following arrest and short-term custody, and at points of transition between different mental health services.Trial registrationCurrent Controlled Trials ISRCTN98067793.FundingThis project was funded by the National Institute for Health Research (NIHR) Health Services and Delivery Research programme and will be published in full inHealth Services and Delivery Research; Vol. 5, No. 8. See the NIHR Journals Library website for further project information.