Evaluation of a peer-delivered, transitional and post-discharge support program following psychiatric hospitalisation

The time following discharge from a psychiatric hospital admission is recognised as one of the most vulnerable times for individuals with mental illness. In Australia, 14.4% of individuals are re-hospitalised within 28 days of discharge [1]. The risk of homelessness and suicide are also heightened during this period [2, 3]. Lack of connection with community-based treatment and support is considered to be a significant factor associated with all of these poor outcomes in the post-discharge period [4, 5].

In an effort to address these issues, a number of programs have been established to provide additional support to individuals during the post-discharge period. Meta-analyses [4, 6] of discharge planning and transitional support interventions have reported promising results in terms of averting re-admission and promoting engagement with community treatment. A number of different program models have been reported, however the key components of these programs have included: individualised, needs-based support; linkage with community services; and communication between inpatient and community services. A number of programs have been delivered by peer workers, that is, support workers with their own lived experience of mental illness and recovery.

One study [7] reported on a peer-delivered transitional support program in Australia. Peer workers delivered 8 to 12 h of support during the first weeks after hospital discharge. Support was focused on linking with community based treatment and support services, developing self-management strategies and other practical assistance. A total of 49 packages were delivered, with an estimated reduction of 300 hospital bed days.

Another study [5] reported results from another peer-delivered support program in the United States. Peer workers provided a “welcome basket” of daily supplies, oriented individuals to their local area and provided support to link with community agencies. Support commenced two weeks prior to discharge and continued for four weeks following discharge. Although readmission rates did not change, participants reported improvements in community functioning, life satisfaction and community integration.

Recognising the potential benefit of enhanced support around the time of discharge, a large non-government mental health service in New South Wales, Australia was funded to establish a pilot transition support program. The program, known as Hospital to Home (H2H) was rolled out in three geographical areas: two in metropolitan Sydney and one in a regional centre. The program was linked to inpatient services in each of the geographical areas. Staff from inpatient treating teams referred individuals to the program who were perceived to need additional support during the transition period. Workers in the H2H program all had their own lived experience of mental illness and recovery. Supports were tailored to the individual, but were primarily focused around providing practical and emotional support as well as linking participants with community-based supports. The program was designed to link with individuals prior to discharge and to provide approximately 6-weeks of post discharge support.

This study was led by an external, university-based research team and was established to evaluate the usefulness of the H2H program.

The University of Sydney Human Research Ethics Committee approved the study (protocol number 2015/868). Participants who provided written, informed consent were enrolled. Data collected during the study included a range of outcome, service efficiency and satisfaction measures as well as semi-structured interviews to enable a detailed evaluation of the program.

This study was developed to evaluate the usefulness of a peer-delivered transition and post-discharge support program. Data were collected on outcomes (recovery, personal wellness, hospital utilisation), service efficiency and participant satisfaction. From each of these data sources, the program appears to have been valuable for participants.

Consistent with other programs reported in the literature [5‐7, 12], the primary aim of H2H was to provide increased support in the post-discharge period. This increased support is designed to reduce hospital utilisation and promote participants’ recovery and wellbeing.

In this study, there was evidence of an estimated average of 11 days reduction in hospital utilisation per participant (or 19 days if data for participants admitted to long-term rehabilitation units were excluded) over the course of a year. However, these figures should be interpreted with a large degree of caution. As figures for post-engagement hospitalisations were based on self-report, these may be unreliable. Additionally, future hospital utilisation is difficult to predict and while previous utilisation is a strong predictor of future utilisation [13], it is likely that individuals’ hospital bed days will vary from year to year. Additionally, due to the high level of variance between participants’ hospital bed days in both pre- and post-engagement phases, the change in average yearly bed days did not reach statistical significance. If the 11 day reduction in hospital bed days is an accurate reflection of actual reduced utilisation, then this would represent a potential saving of $12,034 per participant, based on a cost of $1094 per day [1].

In addition to the potential savings in terms of hospital bed days, participants also reported improvements in functional and clinical recovery and several domains of wellbeing: intellectual (moving from complacent to inquisitive), social (moving from alone to connected) and psychological (moving from unadaptive to adaptive). Each of these represent positive changes experienced by participants over the time they were engaged with H2H. These results suggest that, at the conclusion of the program, participants were engaged in more meaningful activities and felt more able to manage their own mental health. In the period following hospitalisation, these are important outcomes. Supporting individuals to engaging in meaningful activities provides daily structure, opportunities to make social connections and can provide a sense of purpose, all of which have been associated with positive wellbeing [14, 15]. While improvements in clinical recovery initially seem an obvious outcome in the period following discharge, data on increased rates of suicide and re-hospitalisation following discharge suggest this is not always the case. Further, clinical recovery as defined and measured by RAS-DS refers not simply to reduced symptoms, but to reduced impact of symptoms on daily life and also to the sense of mastery over or positive management of those symptoms. Sharing of self-management strategies is one of the many benefits associated with peer-based interventions [16] and may be one of the reasons for the reported improvements in this domain of recovery.

Despite improvements in functional and clinical recovery, personal and social recovery domains of the RAS-DS did not demonstrate significant improvements over time. This is likely related to the fact that improvements in these areas may take longer to emerge. There are, however, substantial indicators that positive changes had occurred in these areas. Firstly, average scores in each of these domains were higher at the conclusion of the program than at commencement. Secondly, positive relationships with support workers with lived experience of mental illness and recovery were highlighted by interview participants as helping to ignite their hope for the future and provided valuable social contact during the vulnerable post-discharge period. This is also reflected in significant improvements in the domains of psychological and social wellbeing in the Personal Wellness Wheel.

Lack of social connectedness is a commonly identified issue for individuals living with mental illness [17] and is challenging to overcome [18]. While positive relationships with H2H workers were helpful, it may be useful to increase focus on family and other natural and thus sustainable relationships. A Tomita, EP Lukens and DB Herman [2] reported positive outcomes associated with family-focused interventions in a transitional support program; improved satisfaction with family relationships was associated with a modest reduction in psychiatric re-hospitalisation rates. Additionally, having a variety of friends and being engaged in relationships where individuals provide support to others have been reported as important indicators of more advanced recovery [19, 20].

The value of the lived experience of support workers was highlighted by participants in both the satisfaction questionnaire and interviews. Numerous advantages of peer-support interventions have been highlighted in previous research [16, 21, 22] and are reflected in the increased employment of peer workers by government and non-government mental health services in Australia and internationally. Similar to results from other studies of peer-delivered transition support programs [5, 7], participants in this study highlighted that support workers were understanding, demonstrated empathy and had more credibility due to their own lived experience of mental illness. Additionally, participants identified that support workers were positive role models, supporting an increased sense of hopefulness for the future.

Finally, supports provided through the H2H program were valued by participants. Information from satisfaction questionnaires and interviews highlighted those features most valued by participants. Participants valued support to connect with community-based resources, emotional supports offered and practical supports provided. Notably, participants highlighted the value of assistance provided in terms of transportation, particularly being transported from hospital to home at the point of discharge and when completing necessary activities in the period following discharge. This is an important finding, as these practical supports are often viewed as unnecessary and a poor utilisation of resources. Additionally, these benefits associated with these practical supports may not be captured by typical service evaluation measures or by clinical or recovery-oriented outcome measures. Finally, the prompt actioning of referrals, frequent face-to-face contacts and positive results in terms of personal goal attainment further demonstrate the responsiveness and usefulness of this program.

Overall, evaluation data suggest that the H2H program has resulted in positive outcomes for participants. Positive outcomes were achieved in recovery and wellness and participant feedback suggests that the use of support workers with their own lived experience of mental illness was a particularly powerful aspect of the program. Additionally, the measures of service efficiency suggest that the program provided timely and efficient services to participants. Finally, although it is not possible to draw definitive conclusions, there is some preliminary data that suggest the program may have supported reductions in hospital utilisation.

For future research in this area, it would be valuable to complete studies that include a control condition and complete more rigorous analyses of post-engagement service utilisation (both community-based and hospital-based resources). Additionally, specific impact analyses including cost-benefit analyses would be useful. This would provide a more detailed understanding of the overall contribution of these programs.