Impact of a peer-review network on the quality of inpatient low secure mental health services: cluster randomised control trial

Abstract

Abstract Background Peer-review networks aim to help services to improve the quality of care they provide, however, there is very little evidence about their impact. We conducted a cluster randomized controlled trial of a peer-review quality network for low-secure mental health services to examine the impact of network membership on the process and outcomes of care over a 12 month period. Methods Thirty-eight low secure units were randomly allocated to either the active intervention (participation in the network n = 18) or the control arm (delayed participation in the network n = 20). A total of 75 wards were assessed at baseline and 8 wards dropped out the study before the data collection at 12 month follow up. The primary outcome measure was the quality of the physical environment and facilities of the services. The secondary outcomes included: safety of the ward, patient mental wellbeing and satisfaction with care, staff burnout, training and supervision. We hypothesised that, relative to control wards, the quality of the physical environment and facilities would be higher on wards in the active arm of the trial 12 months after randomization. Results The difference in the primary outcome between the groups was not statistically significant (4.1; 95% CI [− 0.2, 8.3] p = 0.06). The median number of untoward incidents rose in control services and remained the same at the member of the network (Difference between members and non-members = 0.55; 95% IC [0.29, 1.07] p = 0.08). At follow up, a higher proportion of staff in the active arm of the trial indicated that they felt safe on the ward relative to those in the control services (p = 0.04), despite reporting more physical assaults (p = 0.04). Staff working in services in the active arm of the trial reported higher levels of burnout relative to those in the control group. No difference was seen in patient outcomes. Conclusions We did not find evidence that participation in a peer-review network led to marked changes in the quality of the physical environment of low secure mental health services at 12 months. Future research should explore the impact of accreditation schemes and examine longer term outcomes of participation in such networks. Trial registration ISRCTN79614916. Retrospectively registered 28 March 2014.

Keywords

• Peer-review networks
• Quality improvement
• Randomised trial
• Low secure services
• Forensic mental health