Abstract

Background Peer workers are people with personal experience of mental distress, employed within mental health services to support others with similar experiences. Research has identified a range of factors that might facilitate or hinder the introduction of new peer worker roles into mental health services. While there is mixed evidence for the effectiveness of peer worker delivered interventions, there are no studies exploring how implementation might be associated with effect. Methods This was a qualitative comparative case study using data from interviews with 20 peer workers and their five supervisors. Peer workers delivered peer support for discharge from inpatient to community mental health care as part of a randomised controlled trial. In the trial, level of participant engagement with peer support was associated with better outcome (hospital readmission). Study sites with higher levels of engagement also had higher scores on a measure of fidelity to peer support principles. We compared data from sites with contrasting levels of engagement and fidelity using an analytical framework derived from implementation theory. Results In high engagement-high fidelity sites, there was regular work with clinical teams preparing for working alongside peer workers, and a positive relationship between staff on inpatient wards and peer workers. The supervisor role was well resourced, and delivery of peer support was highly consistent with the intervention manual. In low engagement-low fidelity sites peer workers were employed in not-for-profit organisations to support people using public mental health services and in rural areas. Supervisors faced constrained resources and experienced barriers to joint working between organisations. In these sites, peer workers could experience challenging relationships with ward staff. Issues of geography and capacity limited opportunities for supervision and team-building, impacting consistency of delivery. Conclusions This study provides clear indication that implementation can impact delivery of peer support, with implications for engagement and, potentially, outcomes of peer worker interventions. Resourcing issues can have knock-on effects on consistency of delivery, alongside challenges of access, authority and relationship with clinical teams, especially where peer workers were employed in not-for-profit organisations. Attention needs to be paid to the impact of geography on implementation.