Support from people with lived experience reduces readmission to mental health crisis units
With expertise rooted in personal experience, people who have had mental health problems could offer support, encouragement and a model for recovery, helping reduce readmission rates.
Care from peer support workers with lived experience of mental health conditions may help reduce the likelihood of readmission for people who have recently left acute mental health care, according to a randomised controlled trial of more than 400 people in England published in The Lancet.
The study found that fewer people who received this type of support were readmitted to acute care a year after the study began, compared to people who only received a workbook.
In the UK, more than half of people admitted to acute care are readmitted within a year, and there is no robust evidence on how these readmissions can be reduced.
Support from people with lived experience of mental health problems is used in the UK and USA in programmes such as the NHS' Implementing Recovery through Organisational Change and the USA's Wellness Recovery Action Plan. This study is the first randomised trial to evaluate the effectiveness and shows positive results. However, more research, including to understand what causes the effect found, is now needed before the strategy is implemented nationally in the UK.
Self-management interventions may help people manage their mental health better, and in this study, the authors combined a self-management workbook with help from a support worker who had also experienced a mental health problem.
"People discharged from community crisis services are often readmitted to acute care. Not only does this impede recovery, but also consumes resources that might otherwise be dedicated to longer term improvements in functioning and quality of life," says lead author Professor Sonia Johnson, UCL, UK. "Peer support workers could provide support and encouragement that is particularly warm and empathetic because it is rooted in personal experience, as well as providing service users with a role model for their recovery."
The new study took place across six crisis resolution teams in England, and people were recruited after they had been discharged by a crisis resolution team. Participants had a variety of diagnoses including schizophrenia, bipolar affective disorder, psychosis, depression anxiety disorder, post-traumatic stress disorder, and personality disorder. All people within the study continued treatment and usual care throughout the study.
Participants either received a personal recovery workbook (220 people) or peer support and the workbook (221 people). The workbook included sections on setting personal recovery goals, re-establishing their place in the community and support networks, identifying early warning signs and creating an action plan to avoid or delay relapse, and planning strategies to maintain wellbeing. Participants were asked to record observations and plans in each of these areas.
Participants who received support from a person who had also experienced mental health problems were offered ten one-hour sessions, which took place each week. The support worker listened to their problems and aimed to instil hope by sharing the skills and coping strategies they learnt during their own recovery. Support workers received training beforehand in listening skills, cultural awareness, self-disclosure, and confidentiality, and how to use the workbook.
The authors monitored participants' health records to determine whether they were readmitted to acute care (such as acute inpatient wards, crisis resolution teams, crisis houses, and acute day care services) within one year, and conducted interviews with participants at four and 18 months to obtain their views on the intervention.
After one year, readmission to acute care was lower in the intervention group than in the control group—with 29% (64/218) of participants readmitted in intervention group versus 38% (83/216) of participants in the control group.
Uptake of the intervention was good—72% (160/221) of people offered the support and workbook attended at least three meetings with their peer support worker, and a third (65/198) attended all ten meetings.
Similar numbers of participants in the intervention and control groups read the booklet (84% [133/158] of people in the control group vs 88% [142/162] of people in the intervention group), but more people in the intervention group used it to make written plans (58-64% of people in the intervention group vs 28-44% of people in the control group).
During the study there were 71 serious adverse events (29 in the intervention group and 42 in the control group), but none were deemed related to the study. These included 55 readmissions to acute care, 11 attempted suicides, one attempted murder, two suicides and two deaths with unclear circumstances.
"Our study provides the most robust evidence for the effectiveness of any peer-provided support in a UK secondary mental health setting," continues Professor Johnson. "Our novel findings are potentially important as the intervention is acceptable to patients and feasible for service managers and users who would like to avoid relapse and readmission to acute care."
The authors note some limitations, including that the secondary outcomes of the study gave unclear results, so it is not possible to identify which part of the intervention caused the improvement in patient outcomes. In addition, a high proportion of people in the control group used the booklet, and readmission rates in this group were below the national average, which may suggest that the booklet is effective on its own too. This could mean the effectiveness of peer support was under estimated in the study.
Writing in a linked Comment, Drs Marcia Valenstein and Paul Pfeiffer, University of Michigan, USA, says: "There is much to celebrate but also much left to do following the publication of these findings. The completion of a substantial, adequately powered randomised controlled trial with high ascertainment of the primary outcome and blinded raters is welcome. The finding of reduced readmissions is also welcome, given that implementation of peer support has proceeded apace despite the scarcity of substantial evidence regarding its efficacy. With the publication of this trial, peer-delivered self-management programmes might be considered one of the configurations of peer support with the strongest evidence. That said, the lack of a significant effect on almost all secondary outcomes and potential mediators leaves the mechanism of reduced readmissions unclear. We are left with a complex intervention that appears to reduce readmissions but with few insights into what the active or necessary parts of the intervention might be—eg, support and sharing of life experiences with the peer, engaging actively with the recovery workbook, changes in self-management behaviours that were not apparent through the self-reported measure."