Supported accommodation for people with mental health problems: the QuEST research programme with feasibility RCT

Recruitment of research team

• Ms Sarah Dowling: Project Manager, University College London – started 18 September 2012.

• Dr Sima Sandhu: Researcher, Queen Mary University of London – started 1 September 2012.

• Dr Rose McGranahan: Researcher, Queen Mary University of London – started 1 January 2016.

• Dr Peter McPherson: Research Associate, University College London – started 4 October 2012.

• Dr Joanna Krotofil: Research Associate, University College London – started 1 January 2013.

• Ms Isobel Harrison: Research Associate, University College London – provided maternity cover for Joanna Krotofil from January to July 2014 and remained involved with the project through North Thames Clinical Research Network funding.

No-cost extension

The National Institute for Health Research kindly agreed a 6-month no-cost extension to our contract to acknowledge the time lag between the official contract start date and the date when the relevant subcontracts were agreed that enabled recruitment of the research team.

Ethics approval

Application for ethics approval for WP1–3 of the programme was made in November 2012 and approval was received on 4 February 2013 from the Harrow Research Ethics Committee (reference 12/LO/2009). Ethics approval for WP4 was sought on 12 March 2015 and approval was received on 7 April 2015 from Liverpool Central Research Ethics Committee (reference 15/NW/0252).

Research governance

The research was conducted in keeping with usual research governance guidance and processes. The chief investigator and researchers prepared appropriate standard operating procedures for all WPs. The researchers were trained in the use of the study materials by Helen Killaspy and piloted these prior to use. In 2014, the research was audited by North London Central Research Consortium and no concerns were identified.

Programme management

The chief investigator and programme manager managed the day-to-day running of the programme, overseen by the Programme Management Group (PMG), which met quarterly to review study progress and address managerial and scientific issues as they arose. The Programme Steering Group (PSG) provided an objective ‘quality assurance’ process, reviewing progress and advising the PMG if problems arose that might have an impact on the successful completion of the research. We also recruited a Service User Reference Group to provide an independent view on aspects of the research that were of particular relevance to users of supported accommodation services. The PMG also consulted with an independent group of clinical and policy experts in the field of supported accommodation at relevant stages of the programme, particularly in relation to the adaptations to the service quality tool in WP1 and recruitment in WP4. In addition, a dissemination event and stakeholder roundtable event were held at the end of the programme, attended by members of this expert group, to discuss the implications of the programme findings for future practice and policy.

Patient and public involvement

We recognised that the involvement of service user expertise was key to the success of the programme and included PPI throughout, from design to dissemination. The co-investigator Gerard Leavey co-ordinated these activities. We consulted with the North London Service User Research Forum (SURF) about the focus and design of the study prior to submitting the proposal for funding and incorporated suggestions into the application (e.g. having an independent service user reference group to consult throughout). We further consulted the SURF in relation to the adaptation of the tools in WP1, the interpretation of the results in WP2, the development of the topic guides in WP3 and WP4 and addressing recruitment issues in WP4.

One of our co-applicants (MA) has lived experienced of severe mental illness and of living in supported accommodation. He has worked closely with our group on previous studies in the field of complex psychosis and contributed to our lay summary at the design/application stage and throughout the programme as a member of the PMG. He was also the service user expert on our PSG. After funding for our programme was agreed, we recruited a second service user expert to join Maurice Arbuthnott on the PMG. Service user representatives in the PMG were actively involved in commenting on the delivery of the research programme, reviewing progress and assisting in the development of plans to disseminate the research findings.

We also recruited three lay members with experience of severe mental health problems and supported accommodation services to our Service User Reference Group. This group was facilitated by Gerard Leavey and met every 6 months to provide an independent view on all aspects of the research that were of particular relevance to users of supported accommodation services.

The PSG meetings were held every 6 months to provide oversight to the research programme, reviewing its progress and advising the PMG if problems arose that might have an impact on its successful completion. The service user representative on this group contributed to these discussions.

All service user experts were paid £50 per meeting plus travel expenses.

Work package 1: project months 1–12

Data collection in WP1 was completed on time. However, it was decided that further analysis of the psychometric properties of the adapted QuIRC using the larger sample of service managers and service users recruited in WP2 was indicated. The results were published in April 2016 in BMC Psychiatry. 20 Adaptation of the CAT scale was completed on time and the results were published in March 2016 in BMC Psychiatry. 21

Work package 2i: project months 13–30

The first component of WP2 (national survey) was completed on time and the main quantitative results comparing the three main types of supported accommodation were published in November 2016 in Lancet Psychiatry. 22

Work package 2ii: project months 45–60

The second component (30-month follow-up of service users’ progress) commenced in June 2016 and data collection was completed in August 2017. Researchers tracked service users through 3-monthly contact with their supported accommodation service manager to minimise loss to follow-up. The results were published in The British Journal of Psychiatry. 23,24

Work package 3: project months 13–24

Work package 3 was completed on time and the results were published in July 2017 in BMC Health Services Research. 25

Work package 4: project months 31–60

Recruitment for WP4 started in April 2015 and data collection was completed in August 2017. The results were published in The British Journal of Psychiatry. 23,24

Tool content review and adaptations

The content of the QuIRC was first reviewed by the research team to identify obviously problematic items (e.g. irrelevant items or those requiring rephrasing). Three focus groups with staff and service users recruited from North London were held, one of each focus group from each of the three main types of supported accommodation in England (residential care, supported housing and floating outreach) to gain participants’ views on the relevance of individual QuIRC (staff) and CAT (service user) items. Focus groups were facilitated by the researchers under the supervision of Gerard Leavey. The relevant QuIRC items (staff focus groups) and all seven CAT items (service user focus groups) were used to structure and focus the discussion, which aimed to identify items that required amendment/deletion and additional items. The focus groups were recorded and transcribed and the researchers collated responses.

The findings were supplemented by the advice of three panels of experts who also reviewed the QuIRC and CAT. The first panel comprised five members with expertise in supported accommodation (two senior clinicians, a service manager, a senior policy advisor and a Care Quality Commission senior adviser). The second expert panel was our QuEST study service user reference group, which comprised three members with lived experience of specialist mental health supported accommodation and services. The third expert panel was the North London SURF, which comprised 12 members with lived experience of mental health problems and expertise in mental health services research. All three expert panels were sent the original QuIRC and CAT and a document summarising the comments from the focus groups. They were asked for their opinion about the suggested amendments and any additional amendments. The researchers collated all responses from the focus groups and expert panels, identifying items where there was consensus for adaptation, deletion or a new item. These were reviewed by the QuEST PMG to gain final agreement on changes. The revised QuIRC was then piloted with three service managers (one of each of the three types of supported accommodation) in North London and final amendments to wording were made.

Psychometric property assessment of the adapted Quality Indicator for Rehabilitative Care

To assess the psychometric properties (item response spread, internal consistency, inter-rater reliability, sampling adequacy) of the adapted QuIRC, supported accommodation services were randomly selected from a Microsoft Excel^® (Microsoft Corporation, Redmond, WA, USA) database of all supported accommodation services (residential care, supported housing and floating outreach) in each of 14 nationally representative areas of England (see Work package 2 for further details of area selection and scoping of services). Two services of each of the three types were randomly selected from each of the 14 areas with the aim of recruiting 20 managers from each type of service (60 in total). The researchers contacted service managers to gain their informed consent for participation. Two researchers attended a face-to-face interview with participating service managers. One researcher led the interview and both researchers independently rated the adapted QuIRC from the answers given by the unit manager.

Data analyses were conducted by Sarah White, statistician from St George’s University London, who was involved in the development of the original QuIRC. Items were considered to have inadequate response spread if > 90% of service managers gave the same response. Internal consistency of domain scores was assessed using Cronbach’s alpha and considered acceptable if > 0.6. Inter-rater reliability was assessed using kappa coefficients for categorical data (weighted kappa if more than two categories) and intraclass correlation coefficients (ICCs) for normally distributed, continuous data, and considered acceptable if > 0.7.

Psychometric property assessment of the adapted Client Assessment of Treatment

The psychometric properties of the adapted CAT were assessed in the 618 service users who participated in the national survey of supported accommodation services in WP2 (see Work package 2 for full details of the approach to sampling and recruitment). Internal consistency was assessed using Cronbach’s alpha. Convergent validity was assessed through correlation with ratings of participants’ subjective quality of life using the Manchester Short Assessment of Quality of Life (MANSA),29 which has a total maximum mean score between 1 and 7.

Adapted Quality Indicator for Rehabilitative Care

A total of 28 QuIRC items were rephrased, 20 items were deleted and 10 items were added. For example, item 77 from the original QuIRC was adapted to better reflect the nature of support within these settings (original item: ‘How many of your staff are trained in control and restraint techniques?’; amended item: ‘How many of your staff are trained in breakaway techniques?’), whereas item 116 was simply omitted (‘Are patients/residents free to send and receive uncensored mail or email?’). The final version had 143 items. It was also agreed that because floating outreach services are not ‘building based’ but provide support to people living in an independent tenancy, the items relating to the living environment of the service were not relevant and, therefore, the adapted QuIRC would not be able to provide a rating on this domain for these services.

Inter-rater reliability of the QuIRC was carried out with managers of 14 residential care homes, 21 supported housing services and 17 floating outreach services (52 services in total).

Only 16 out of the 143 items showed a poor response spread. Internal consistency was inadequate for all domains except therapeutic environment and treatments and interventions. However, the analysis was limited by a relatively small sample size and the lack of variability in response to some items. With regard to inter-rater reliability, 70 ICC analyses were conducted and only one item was found to be unreliable. A total of 186 kappa coefficient analyses were conducted and four items were found to be unreliable (κ < 0.7). In addition, there were 14 items where analyses could not be conducted because there were too few cases (five items), zero variance (two items) or where variables were constants (seven items). The full results of these assessments are available in Killaspy et al. 20

The PMG agreed amendments to the adapted QuIRC in response to the results. It was decided to keep items with inadequate variance because (1) to drop them would have disrupted the logical flow of the tool content and (2) greater variance might be achieved in future development of the tool for use in settings outside the UK. Additional explanatory information was added to improve the reliability of one item, one item was dropped completely and unreliable response options were dropped for three items.

It was also agreed that internal consistency would be reassessed using the larger sample of 87 services participating in WP2 along with an assessment of sampling adequacy (the proportion of variance among the variables that might be common variance) using the Kaiser–Meyer–Olkin (KMO) statistic. Hair30 suggested a KMO value of ≥ 0.5 to evaluate whether or not a sample of data has sufficient common variance to make exploratory factor analysis appropriate.

Internal consistency improved on all domains with the larger sample (Table 2). All domains met the KMO > 0.5 criterion of sampling adequacy. It should be noted however that both Cronbach’s alpha and the KMO statistic are influenced by sample size and in this analysis the sample size was still smaller than desirable for robust estimates of both properties (generally 300 observations). However, we were able to investigate whether or not there was adequate common variance within the domain items to assume that the domain had coherence. We replicated the approach taken in the original development of the QuIRC, where items were considered to load onto a factor (domain) if they scored > ± 0.3. As some items had zero variance, they were removed before analysis (living environment: two items; self-management and autonomy: three items; human rights: four items). All items loaded onto a factor within that domain at the > ± 0.3 level (i.e. each item was contributing to the common variance within the domain).

We engaged the same information technology specialist who developed the web-based version of QuIRC to develop a similar application for the adapted QuIRC for supported accommodation services. This increases its accessibility, reduces the time required to complete it compared with a face-to-face interview, and provides a built-in scoring algorithm. It has a similar facility to the original QuIRC application in producing a printable report for the service manager about the performance of their service on the adapted QuIRC domains, comparison benchmarking data for similar services and suggestions for how to improve performance. The final version of the adapted QuIRC was named the Quality Indicator for Rehabilitative Care: Supported Accomodation (QuIRC-SA).

Adapted Client Assessment of Treatment

All seven CAT items were considered relevant by the focus group participants and expert reference group members, with only slight modification of the wording required (e.g. ‘treatment/care’ was changed to ‘support/care’). No items were deleted or added. Cronbach’s alpha was 0.89. With regard to convergent validity, scores on the adapted CAT were positively correlated with the mean total MANSA (quality of life) score (r_s = 0.35; p < 0.001). Full results are available in Sandhu et al. 21

Quality Indicator for Rehabilitative Care: Supported Accommodation

One of the limitations of the QuIRC-SA was the inadequate internal consistency. Although the domains were coherent [as measured by sampling variance (KMO statistic)], it is possible that the estimates of internal consistency (Cronbach’s alpha) may have been influenced by our sample size; a sample size of 300 to 400 is desirable in order to obtain robust estimates of internal consistency. The decision to retain items with inadequate response variance may be considered a limitation of the tool. Although it could be argued that the retention of these items unnecessarily increases the size and administration time of the QuIRC-SA, omitting them would have disrupted the logical flow of the tool. We were not intervening to improve service quality during the QuEST programme and we therefore did not assess the tool’s sensitivity to change. Test–retest reliability was not assessed as we felt that this was too onerous for busy service managers (as the tool takes 1 hour to complete). Both should be assessed in future studies, but their omission had no influence on the findings of later phases of the QuEST programme as we used only baseline QuIRC-SA assessments.

Client Assessment of Treatment for Supported Accommodation

Although the development and testing procedures for the CAT-SA were rigorous, a number of limitations must be acknowledged. First, we did not collect data on symptoms or clinical profile when assessing the psychometric properties of the measure, which negates the possibility of investigating potential relationships between CAT-SA ratings and symptomology. Second, owing to the nature of our investigations, test–retest reliability and sensitivity to change have not been assessed. Third, additional validation procedures are needed to establish its validity in countries outside England. Finally, service user experience/appraisals on the CAT-SA may be influenced by support received outside the supported accommodation setting.

Quality Indicator for Rehabilitative Care: Supported Accommodation

• The QuIRC-SA is the first standardised tool for quality assessment of specialist mental health supported accommodation services.

• Of 143 items, only 18 showed a narrow response range, and five had poor inter-rater reliability.

• The QuIRC-SA had excellent inter-rater reliability and exploratory factor analysis showed that items loaded onto the domains to which they had been allocated.

• The QuIRC-SA can be recommended as a self-report quality assessment tool for supported accommodation services.

• A digital version is available, free to use, at www.quirc.eu.

Client Assessment of Treatment for Supported Accommodation

• All items of the CAT were considered relevant to mental health supported accommodation services; slight changes to wording were made.

• The CAT-SA demonstrated good internal consistency and convergent validity.

• The CAT-SA can be recommended as a ‘patient-reported outcome measure’ for use in mental health supported accommodation services.

Methods

Our sample size was estimated to assess the difference in proportion of people moving on from each of the three types of supported accommodation 30 months after recruitment (assessed in the second component of WP2 – the cohort study). Our original estimate was that we needed to recruit 90 services (30 of each type) and 450 service users (five from each service) from 14 nationally representative areas based on an intraservice cluster correlation coefficient of 0.07 and a mean cluster size of five. We selected the 14 areas using an index developed by Priebe et al. 11 for their postal survey of supported accommodation, which ranks local authority areas on the basis of mental health morbidity, social deprivation, urbanicity, provision of community mental health care, supported accommodation, local authority mental health-care spend and housing demand. Recruitment was carried out between 1 October 2013 and 31 October 2014.

The researchers contacted key personnel working in housing departments in each of the 14 areas to first scope the local provision of mental health supported accommodation services and the number of places available in each. Residential care homes for adults with mental health problems in each area were also identified from the Care Quality Commission (the registration authority for care homes in England and Wales) website to ensure that none was missed (www.cqc.org). Services that were unusually large (> 80 service users) or small (< 6 service users) were excluded to increase generalisability of the sample. Five areas were replaced with another area closest on the sampling index: three because the area had no residential care services and two because the researchers were unable to clarify local supported accommodation provision (see Appendix 2, Table 5).

After 15 services of each of the three types were recruited, we reviewed our sampling strategy. We recalculated our intraservice correlation coefficient using responses to one item of the managers’ research interview [their estimate of the proportion of service users who had moved to more independent accommodation (residential care and supported housing) or required less support (floating outreach) in the last 12 months] and found it to be 0.18. A larger service user sample (624 rather than 450) was thus required. Owing to the relatively small number of residential care services identified, the number of services needed was adjusted to 21 residential care, 35 supported housing and 35 floating outreach, and the number of service user participants per service was increased to a maximum of 10.

The researchers randomly ordered the services in each area and within each of the three service types and contacted service managers sequentially to invite their participation. Service managers were given up to 4 weeks to reply. When the target number of services of each type had been recruited or the list exhausted for each area, the researchers moved on to the next area (see flow chart in Killaspy et al. 22).

Where service managers responded to the initial contact, the researchers arranged a time to discuss the study further by telephone. If the manager was willing for their service to participate, the researchers arranged a face-to-face meeting to explain the study to staff. A list of potentially eligible service users was then obtained from the service manager (excluding those they considered to be unable to give informed consent to participate and those absent from the accommodation for any reason). Each potentially eligible service user was allocated a unique identifier code. The service user list for each service was randomly ordered by the researchers in blocks of six. Service users in the first block were approached in any order by the researchers to seek their participation and subsequent blocks were generated until five service users had been recruited or the list was exhausted for each service. All potential participants received a participant information sheet about the study and had at least 2 days to read it and address queries to the researchers before giving their informed consent to participate.

Results

A total of 22 residential care, 35 supported housing and 30 floating outreach services were recruited. From these 87 services, 619 users were recruited (residential care, n = 159; supported housing, n = 251; floating outreach, n = 209). A total of 193 keyworkers completed ratings on service users.

The full characteristics of the three types of service and their users are shown in tables 1 and 2 of Killaspy et al. 22

In summary, floating outreach provided more places than the other two types of service. Both floating outreach and supported housing services expected to work with their users for a median of 2 years, whereas for residential care this was 5 years. All three types of service used similar processes to assess new referrals. Almost all services had clinical input from a community mental health team or community rehabilitation team, despite the fact that only one-third of floating outreach clients were subject to the Care Programme Approach (vs. almost all residential care and supported housing clients). Supported housing services scored highest on six of the seven domains of the QuIRC (floating outreach scored slightly higher for human rights).

Most service users were male (66%), single (66%) and unemployed (82%). Users of residential care services were older and had been known to mental health services longer than users of the other two types of service. Most (68%) had a primary diagnosis of psychosis but one-third of floating outreach service users had depression/anxiety. The proportion of service users with substance misuse problems was relatively small (16% alcohol, 12% drugs), with the lowest prevalence among residential care service users. Users of residential care and supported housing had slightly more previous admissions than users of floating outreach and more were subject to a community treatment/restriction order. Overall, 40% of service users had committed an act of violence at some time, but there were few serious incidents of risk to others in the last 2 years. More users of supported housing (26%) and floating outreach (21%) had self-harmed within the last 2 years than users of residential care (4%). Risk of serious self-neglect was reported for 57% overall (72% for users of residential care services and at least 50% for users of the other two types of services). Vulnerability to exploitation was reported for around one-third of users in supported housing and floating outreach services and for 41% of those in residential care. Overall, 67% to 78% of service users across the three types of accommodation were considered a risk to themself or others.

There were few differences in social function and challenging behaviours between service users in the three types of accommodation, but those in residential care, unsurprisingly, had more needs than those in supported housing or floating outreach. However, there were few unmet needs across the three types of accommodation. Users of supported housing and floating outreach were more likely to report having been a victim of crime in the last 12 months than those in residential care (residential care 8%, supported housing 25%, floating outreach 22%) and around half of these incidents involved physical assault. Residential care service users had higher ratings of satisfaction with their safety than users of the other two types of service.

Regression analyses that took account of clustering by area and service showed lower quality-of-life scores for those in floating outreach and supported housing than for those in residential care services, but higher levels of autonomy in both supported housing and floating outreach than in residential care (see table 3 in Killaspy et al. 22). In these analyses, the therapeutic milieu (which could not be assessed in floating outreach services) was rated higher by users of supported housing than by users of residential care [mean difference 0.973, 95% confidence interval (CI) 0.126 to 1.821; p = 0.024], yet there were no statistically significant differences between the three service types in terms of satisfaction with the care received.

Appendix 2, Tables 6 and 7, show the results of the regression analyses that investigated service and service user characteristics associated with service users’ quality of life, autonomy, ratings of the therapeutic milieu (residential care and supported housing only) and satisfaction with care. The QuIRC domains therapeutic environment and recovery-based practice were found to be highly correlated. It was decided to keep recovery-based practice in the models as this domain had been found to be a predictor of successful discharge from inpatient mental health rehabilitation units in a previous study. 39 As data could not be collected for living environment domain scores for floating outreach services, this domain was also dropped from the regression models. The 19 variables included (see Appendix 2, Tables 6 and 7) were agreed by the PMG on the basis of those where differences were found in the summary statistics and where there was clinical justification.

In these analyses, which accounted for differences in service user characteristics, service users’ quality of life (MANSA) did not differ between supported housing and residential care but remained lower for those in floating outreach than for those in residential care. Supported housing was still predictive of higher autonomy (RCS) than residential care but floating outreach was not. In addition, positive associations were found between service users’ quality of life and the mental health morbidity/housing index of the local area, service user age, primary diagnosis (psychosis vs. non-psychosis) and problematic drug use. Negative associations were found with the number of places occupied per service, the adapted QuIRC recovery-based practice domain score and the number of unmet needs. However, the small size of the coefficients suggests that these associations had very little clinical impact. Service users’ autonomy was negatively associated with the QuIRC-SA domain treatments and interventions score and unmet needs, but, again, small coefficients suggest that these associations were of limited clinical significance. For each additional occupied place per service, there was a small reduction in service users’ ratings of satisfaction with care (CAT-SA) and the therapeutic milieu (GMI). The GMI ratings were positively associated with the QuIRC-SA human rights domain score, service users’ age, staff ratings of their social function (LSP) and risk history. Once again, small coefficients limit the clinical relevance of these findings.

Health economic component

Links to other work packages

The QuIRC-SA and CAT-SA, developed in WP1, were used to assess service quality and service user experiences in WP2i. Service users in supported housing and floating outreach had similar levels of risk, social functioning, challenging behaviours and autonomy, and received similar levels of input from their supported accommodation support workers, suggesting equipoise between the two service types, which supported the rationale for a comparison of these two service models in the feasibility trial (WP4).

Limitations

A number of limitations must be acknowledged. In spite of using a sampling strategy explicitly designed to minimise bias and produce a nationally representative sample, service users who declined to participate or who lacked capacity to consent may have introduced a sampling bias. In addition, our findings cannot be generalised to supported accommodation services and systems outside England. Finally, the cross-sectional nature of WP2i means that we cannot infer causality from our findings.

Key findings

• Compared with floating outreach, service users in residential care and supported housing had more severe mental health problems.

• Over half of all service users were at risk of serious self-neglect and over one-third had been vulnerable to exploitation over the previous 2 years.

• One-quarter of those in supported housing and one-fifth of those receiving floating outreach services had been the victim of crime in the last year, compared with 8% of those in residential care.

• Residential care was the most expensive service and floating outreach was the cheapest. Residential care and supported housing were both highly likely to produce better outcomes but had higher costs than floating outreach.

• As assessed by the QuIRC-SA, supported housing demonstrated the best quality of care compared with the other two service types.

• After adjusting for clinical differences between service users, quality of life was similar for those in residential care and supported housing, but lower for those in floating outreach. Autonomy was greater for those in supported housing than in residential care and floating outreach. Satisfaction with care was similar across service types.

Methods

Results

Participant flows in the cohort are shown in Figure 2. After accounting for withdrawals (n = 7) and deaths (n = 26), we followed up 586 out of 619 (95%) participants over 30 months (residential care n = 146; supported housing n = 244; floating outreach n = 196).

FIGURE 2.

Participant flow during the prospective study. FO, floating outreach; RC, residential care; SH, supported housing; SU, service user.

Health economic component

Links to other work packages

WP2ii followed the cohort of participants recruited in WP2i and our analyses to identify predictors of our primary outcome used the data gathered at WP2i recruitment.

Limitations

Findings from WP2ii must be viewed in the light of a number of limitations. First, successful move-on (the primary outcome) for floating outreach service users was operationalised as having fewer hours of support per week than at baseline. Arguably, this is a lower threshold for ‘success’ than that applied to service users of residential care and supported housing and thus the higher ‘move-on’ we found for floating outreach service users may have been overestimated. Nevertheless, our sensitivity analysis that reclassified floating outreach service users as having a successful outcome only if the number of hours of support they were receiving had reduced by more than half found similar results. Second, as we were unable to monitor service users directly, it is possible that participants who moved on to full independence (discharged from floating outreach services) may have returned to some form of supported accommodation without the research team knowing. However, we attempted to address this through case note reviews of all participants at the 30-month follow-up. Third, follow-up data for some service users was collected through case notes only; although this allowed us to assess both the primary and secondary outcomes (a strength of our design), some other data, such as contact with family for our health economic analysis, could not be collected. Fourth, service use data provided by staff and participants may have be prone to recall error. However, the period of interest was 3 months and it is assumed that this is short enough to mitigate against this possibility. Fifth, it is possible that in some cases there may have been confusion over service definitions and double counting between care co-ordinator and support staff contacts. However, if this occurred it is unlikely to have been a major issue because care co-ordinator contacts were far less frequent than support staff contacts. Finally, owing to the small sample size for the < 9-month subcohort, we were unable to conduct inferential analyses for this group.

Key findings

• We achieved a high follow-up rate, collecting primary outcome data on 95% of participants at 30-month follow-up.

• Successful move-on was achieved by 15 out of 146 service users (10.3%) in residential care, 96 out of 244 service users (39.3%) in supported housing and 132 out of 196 service users (67.3%) from floating outreach.

• Successful move-on was associated with the service’s quality scores at recruitment; QuIRC-SA domain scores for human rights and recovery-based practice were positively associated with the primary outcome and QuIRC-SA social interface domain score was negatively associated with it.

• More unmet needs, greater risks and longer length of time in the service were negatively associated with participants moving on successfully.

• After adjusting for these variables, floating outreach service users were more likely than those in residential care (OR 7.96, 95% CI 2.92 to 21.69; p < 0.001) and supported housing (OR 2.74, 95% CI 1.01 to 7.41, p < 0.001) to achieve successful move-on. The adjusted OR of achieving successful move-on for supported housing service users compared with those in residential care was 2.90 (95% CI 1.05 to 8.04; p < 0.001).

• Very few service users had an admission after moving on.

• For service users who had been in their accommodation for < 9 months at recruitment, satisfaction with care (CAT-SA) was highest among those in residential care. This group also had the lowest social functioning. There was little difference in quality-of-life (MANSA) and autonomy (RCS) scores between users of the three service types in this subgroup.

• Service costs for those who had moved on were significantly lower than for those who did not.

Data analysis

Interviews were recorded digitally, transcribed and anonymised before being entered into NVivo 10 (QSR International, Warrington, UK) software for coding. Transcripts were subjected to inductive semantic thematic analysis41 with the main research questions forming the basis for the initial coding frame. Further analysis sought to identify differences in the experiences of staff and users of different service types and the aspects of care considered particularly helpful/unhelpful. Coding was based on the search for patterns in terms of things occurring in a similar way, things occurring in different ways, if things occur in a sequential way, frequency of occurrence and any ascribed causation. A second coder checked for consistency and validity of interpretations in a randomly selected 20% of the coded interviews. Once this verification procedure was complete, the categorisation and theme generation process commenced. Conceptually similar codes were grouped to form subthemes, which were then grouped to form larger, candidate main themes. The thematic structure was agreed through discussion between the researcher (Sima Sandhu) and the leads for WP3 (SP and GL). Further details of the approach to data analysis are available in Sandhu et al. 25

Participants

The length of time that service user participants had lived in the current supported accommodation service ranged from 3 months to > 11 years, with a mean of 3.3 years. Staff had worked in the current service for between 3 months and 20 years, with a mean of 6.5 years, and they had worked in mental health services from 1 to 31 years, with a mean of 12.75 years. Participants’ demographic characteristics are shown in Table 3.

Purpose/goals of the service

There was consensus between service users and staff about the main purpose of supported accommodation, which was most frequently cited as being to provide support for users to attain and sustain greater independence. Four related subthemes emerged: supporting people with their mental health; providing safety and stability; building confidence; and building independence:

I think the purpose is to support people that um, generally obviously have a mental health problem, support them from being at crisis stage, back into independence, and to um, bring about positive outcomes, so that they can become independent.

Staff: 116. Support worker – floating outreach. Female

To achieve this, services needed to provide practical support for users to gain confidence in managing activities of daily living, encouraging them to build on their existing capabilities, and to provide living environments that were safe, secure and stable:

And most of the people that come here have struggled to live independently on their own without any ongoing support. So um, so they’ve found it a lot safer and a lot securer.

Staff: 070. Manager – supported housing. Female

Staff, in particular, emphasised their role in supporting service users with their financial affairs (budgeting, paying bills, accessing appropriate benefits and managing debt) and in advocating for service users:

We would look at income and expenditure – particularly if they were worried about money, or found themselves in debt. If they found themselves in debt I would, and I have done, encouraged them to make the call to the debtor and explain their circumstances, and maybe a little bit of their own personal background, so that there’s an element of understanding, and try if necessary to come to an arrangement whereby, you know they are making repayments on a lower, at a lower rate.

Staff: 089. Support worker – floating outreach. Female

Service users reported that staff gave them the encouragement and help they needed to achieve the things that they wanted to do. Across service types, this included gaining independence with activities of daily living, and in floating outreach services, users valued being supported to work towards employment:

Well it’s confidence-building, helping you to deal with things that you need to know about, like it’s how to pay rent and your bills and how to keep yourself clean and . . . and where you should go.

Service user: 141. Supported housing. Male

I wish to one day you know become like that so I could live independently; I’d really love that – you know?

Service user: 065. Floating outreach. Male

Effective/helpful aspects of the service

Six interrelated themes reflected staff views about how to work effectively towards the goals: incremental steps to progress; knowing the individual well; the support of other service users; working together to avoid de-skilling and dependency; tailoring support for social and community engagement; and good liaison with other services.

Services that supported their users to gain skills incrementally were regarded as effective and helpful by staff and users alike. These services tended to ‘support with’ rather than ‘do for’ people:

. . . and that progress report is looking at what they achieved last month and what new things have been introduced this month. So the idea is to keep introducing new things, and to step away from the things they no longer need support with. So, you know, we have service users where we would be going to the cinema with them two or three times a week, who now go to the cinema by themselves. So what we would then do is introduce an activity that maybe they will need support with – and then they go and do that by themselves, and that’s the progress.

Staff: 033. Support worker – floating outreach. Female

Services were more likely to be considered effective when staff understood their clients well and formed trusting relationships:

Well . . . they’re there to support you, they’re there to support you and um, you can trust them. I mean they’re sincere, genuine, um, they have a proper attitude.

Service user: 025. Residential care. Male

Staff noted that service users also appeared to benefit from feeling supported by other clients living in the accommodation:

I think you can see them support . . . like especially [Name] and [Name]. They’re particularly supportive of each other, so . . . I feel with a lot of them they do come together and they will support each other. They’ll always ask the other one ‘oh are you OK’ and ‘how are you doing?’, and you know it’s nice.

Staff: 095. Support worker–supported housing. Female

There was an awareness from both staff and service users that individuals needed to be encouraged and supported to do things to encourage them to gain/regain skills:

I got a little bit of support with it yeah. Kind of . . . you know, a bit from me and a bit from the staff really . . . They’re . . . it’s not like they force you to do things but you’ve got to put, you know, you’ve got to input yourselves . . . because it does take, primarily it does take yourselves to do it, yeah.

Service user: 035. Residential care. Male

Staff were mindful that social and community-based activities needed to be tailored to the individual to maximise their appeal, with the aim of creating wider social networks as part of the person’s rehabilitation:

Well I think what we’ve done initially is we’ve actually took the time to sit down with people and actually talk to ‘em about what their interests are; so what they actually would like to do. Like we found out that one of the residents apparently was a keen cook, many years ago, so we got him engaged with a local mental health charity who do cooking lessons and bakery lessons, so he goes every week . . . he’s absolutely delighted with that.

Manager: 083. Residential care. Female

Um, we do . . . we run activities here, we do the gardening clubs, walking clubs to get people socialising sort of with other customers in the area and to get people active as well. Um, we’ll organise things; we organise trips for them to go out from here as well so that they, you know they’re not socially isolated.

Staff: 052. Support worker – supported housing. Female

Staff also noted the importance of appropriate training, regular supervision and being flexible in their approach. Joint working with community mental health services and other agencies was seen as crucial for an effective service, as was awareness of when involvement from other services was needed:

Yes, because if something’s not working, then at least it’s more than one head thinking right well, what else can we try, where else can we go with this?

Staff: 089. Support worker – floating outreach. Female

Service users noted that the practical and emotional support given when attending appointments with external agencies was particularly helpful:

So . . . I, I went for um, a scan . . . and ECG [electrocardiogram], and [Name] came with me to see the consultant for the results. And I’m so glad she did . . . So you see I have got very little faith in the medical profession. Um, and as I say, [Name] supported me through that . . . she helped me through it.

Service user: 076. Supported housing. Male

Challenges/unhelpful aspects of the service

Poor communication with other agencies was seen as a particular challenge by staff, as well as a lack of suitable training and supervision. In addition, staff and service users felt that staff were pressured and did not always have adequate time available to support service users with their activities (inside the accommodation and in the community), which limited their progress:

Well just . . . or . . . just mainly that it’s getting tied, where there’s only one staff and you can’t leave the building – for me that’s the . . . the major stumbling block.

Staff: 009. Support worker – residential care. Male

And I asked to go to Asda [www.asda.com] in the day, with [Name], because we were getting . . . I wanted a few things from there. He said, ‘oh, no you can’t go . . .’. Then I ended up not going . . . we, we ended up not going – couldn’t go . . . because it, because there wasn’t other people going . . . There’s not always enough staff on here to go out.

Service user: 019. Residential care. Male

Both staff and service users also reported that conflicts between service users were particularly unhelpful and stressful:

But then you’ll get somebody say like [Name] who will come into the room and just start like throwing things, um, and then obviously tensions mount or you’ve got two people that wanna get in the lift at the same time together and neither of them will back down so they’ll just try and get in there together and then have a fight.

Manager: 083. Residential care. Female

Sometimes I think . . . the only sort of criticism if you wanna call it that, is that sometimes, you know another resident might be annoying you or there’s something going on . . . You know I’m not talking massive things, I’m just like, you know um . . . I dunno, say upstairs messing our kitchen up or something say. Just, you know nothing . . . um, or someone’s behaviour . . . you know . . . that you know is outside of the rules if you like.

Service user: 098. Supported housing. Female

Service users reported that a lack of choice and involvement in decisions about things that concerned them was unhelpful. Related to this, some staff also expressed a particular concern about creating dependency among service users:

I think that um, still . . . the relationship I think is very much a dependency relationship from the resident’s point of view; I think that the residents um, still look to the staff as . . . or do look to the staff as kind of their nursemaids and their housemaids and their [in-breath] um, you know everything to do with that really.

Manager: 083. Residential care. Female

Others noted the particular difficulties that arose when users did not engage with services:

The only thing that gets in the way is the client engagement side of things. I would say we . . . we can offer across . . . we can either offer the support or signpost to the support that that service user needs . . . The support services are there, the support that we can provide are there, the only barriers sometimes to clients being successful is their engagement.

Manager: 094. Floating outreach. Female

Facilitators of and barriers to moving on to greater independence

As noted earlier, staff felt that preparing for greater independence could be achieved by working incrementally on specific skills to help service users gain confidence. Staff emphasised the importance of planning for move-on a long time ahead. This included ensuring that there was good communication and close working with staff from the relevant community mental health team throughout the process to ensure that adequate support would be provided after the move. Staff also highlighted the need to have open conversations with service users throughout their time with the service about move-on, but without forcing the issue with them:

So it’s discussed at every support plan meeting – they happen monthly, 2-monthly, 3-monthly, depending on what the client’s needs are, um, so we just, we discuss about moving on, if it’s moving on from accommodation based, or whether it’s to another accommodation setting that they require, [in-breath] but was also discuss about the steps towards independent living and what services, or support needs to take place to achieve that.

Staff: 099. Manager – floating outreach. Female

Aside from a lack of suitable accommodation to move on to, staff reported that the biggest barrier to service users’ progress was anxiety about move-on (both the process and the likelihood of success), which could manifest in resistance from service users and staff. Service users expressed appropriate concerns about the location of any future home, if they would have enough support and if they would feel lonely if they moved to more independent accommodation:

Yeah I do yeah, because um, what’s in the back of my mind is that, you know, what I’ve been through, I don’t want to, I don’t want a repeat of that again. So, you know, if you take that support away – you know moving on from here to another place – you know, in the back of my mind, I don’t wanna, I don’t wanna go through again what I’ve been through ‘cos I don’t think I could come back from another one.

Service user: 085. Floating outreach. Male

Views on comparing the effectiveness of two models of supported accommodation through a randomised controlled trial

The majority of staff and service users felt that a study to compare the two models of supported accommodation would be useful and should go ahead but there was concern about the use of random allocation in such a study:

I can understand that it would be useful. The . . . the problem I’ve got really is that if I’m quite sure that if I was to go er, back to head office and say well we’ve got . . . we’re discharging someone, we’re just gonna flip a coin and say well you’re going here and you’re . . . that they would say well, it’s not person-centred and you’re not, you’re not addressing the needs of the individual.

Manager: 034. Residential care. Male

No I think they should really leave it down to the person, individual to say, d’you know what, I’ll give that a go or I’ll give that a go. And then they can . . . then they decide which they think is gonna be better for them – you know – and how they coped in both those situations.

Service user: 076. Supported housing. Male

However, many participants stated that as long as service users’ needs and preferences, and the views of the referring clinicians, were taken into account, randomisation could be possible in some cases. Participants suggested the following outcomes for assessment in the trial: mental health (including hospitalisations); physical health; quality of life; social inclusion; managing the tenancy; level of independence; substance misuse; incidents of risk; medication adherence; ability to manage activities of daily living/social functioning; and engagement and satisfaction with services.

Trial design

This was an individually randomised, parallel-group trial.

Sample size

As this was a feasibility study, a formal sample size calculation was not conducted. The original target was to recruit and randomise 60 participants from across the three study sites.

Setting

The feasibility trial was conducted in three sites that provided both types of supported accommodation and where the study team had good links (i.e. North London – Camden and Islington; East London – Tower Hamlets, Newham and Hackney; Gloucester and Cheltenham).

Inclusion/exclusion criteria

All service users in the three study sites referred to either supported housing or floating outreach services who had housing rights in the local area and were subject to the Care Programme Approach were eligible for inclusion. Those who lacked capacity to give informed consent were excluded.

Recruitment process

Each of the three sites had a system for referral of service users to local supported accommodation services. All those referred to supported housing or floating outreach were considered for potential participation in the study. The research team first met with the relevant staff at each site to explain the purposes of the study and clarified the local referral processes. A researcher who liaised with the key personnel involved in co-ordinating the referrals system and the local clinicians making referrals was based at each study site. They gained details of any new referrals and contacted the referring clinicians to discuss whether or not the individual might be appropriate for participation in the study and, if so, the researcher contacted them to gain informed consent for participation. When the clinical team felt that it was not appropriate for the person to be approached about the study, that individual was not contacted by the researcher. Recruitment took place over 12 months from June 2015.

After 6 months, we decided to adjust our approach to try to increase recruitment. In addition to the processes described above, researchers met with the managers of acute inpatient wards and community mental teams operating across the three sites to identify any individuals being considered for referral to supported housing who had not yet been referred. Service users who did not consent to randomisation were offered to participate in a naturalistic cohort whereby they gave informed consent to complete the same research interviews as trial participants but their supported accommodation was not allocated randomly.

Allocation of participants to trial arms

Participants in the trial were randomly allocated on an equal basis to receive either a local supported housing or a floating outreach service. Computer-generated randomisation was conducted independently of the research team by a statistician from the Pragmatic Clinical Trials Unit based at Queen Mary University of London and stratified by site. The outcome of randomisation was communicated to the QuEST project manager who then communicated it to the local referrals co-ordinator, and referrer, to process the participant’s supported accommodation allocation accordingly.

Comparison services

Supported housing services provided a constant level of staffing on-site to a number of service users living in individual or shared tenancies with the expectation of move-on within 2 years. Floating outreach services provided visiting support of flexible intensity to service users living in a permanent (non-time-limited) independent tenancy.

Content of care was assessed using the QuIRC-SA,20 completed with the service manager once for each service, 6 months after the participant was randomised (assuming that they had moved to the service). This was completed only once per service.

Data collection

We collected the following metrics to inform the feasibility of a larger trial: number of referrals to the trial; number recruited; attrition (i.e. number who withdrew consent to continue with the research, declined to move to the allocated service or could not be located at follow-up); and time from recruitment to moving into either type of supported accommodation. We assessed the feasibility of potential outcome measures through collection of data from service users, support staff and service managers, at recruitment and 6 and 12 months after recruitment (Table 4).

Masking of researchers

The feasibility of using a telephone interview to collect follow-up data from service users was also explored. At the 12-month follow-up, the researcher met with the service user participant, took informed consent, then telephoned a second researcher (blind to the participant’s supported accommodation allocation) who completed one instrument from the interview battery (MANSA) with the participant. This measure was selected as all others would have invalidated the blinding by revealing the participant’s allocation.

Qualitative evaluation

We aimed to recruit five participants who had agreed to randomisation and five who had not but who had agreed to be followed up naturalistically, along with their respective referrers, to assess their experiences of the trial, the process of randomisation and their views on the usefulness and feasibility of a larger-scale trial. The aim was to elicit and detail common themes systematically that could inform the feasibility of a larger trial. lnterviews were recorded, independently transcribed and anonymised. The text data were entered into a software package (NVivo 10) to assist management and coding. The interviews were analysed using thematic content analysis; a coding frame was developed by one of the researchers (Rose McGranahan), with supervision from Stefan Priebe and Gerard Leavey, which was expanded and modified to include further codes as new themes and subthemes emerged in the course of interviews and analysis.

Data analysis

We followed Consolidated Standards of Reporting Trials (CONSORT) guidelines on the analysis of feasibility trials for the presentation of our results. 46 However, our analysis was mainly descriptive and focused on the recruitment rate, acceptability of randomisation to participants and staff, ease of collection of data, characteristics of participants, other baseline and outcome variables, the feasibility of masking outcome assessments, loss to follow-up and any adverse events.

Feasibility metrics

We screened 1432 potential participants, of whom 87 were ineligible (not on a Care Programme Approach, n = 63; no recourse to public funds/housing, n = 24), 456 were deemed inappropriate for participation by the research team (no plans for move-on/new admission, n = 194; already housed, n = 99; no response from clinical team, n = 60; own accommodation, n = 51; referral withdrawn, n = 22; previously screened, n = 13; eviction in process, n = 8; moved out of area, n = 5; and clinician refused access, n = 4) and 851 were deemed inappropriate for participation by the clinical team. The most common reasons were that the individual had a high level of support needs and was felt to be inappropriate for floating outreach (n = 524) or that they had low support needs and were thus inappropriate for supported housing (n = 137). In total, 17 service users consented to participate, with eight service users randomised and nine participating in the naturalistic arm. The CONSORT flow diagram is presented in Figure 3.

FIGURE 3.

The WP4 CONSORT flow diagram. CMHT, community mental health team; DNA, did not attend; FO, floating outreach; SH, supported housing.

Attrition

Of the 17 recruited participants, three were lost to follow-up [one randomised (one refused follow-up); two naturalistic (one death; one refused follow-up)].

Time from recruitment to moving

Of those randomised, three out of eight participants (38%) moved to their allocated accommodation. The median time from recruitment to moving was 4 months [interquartile range (IQR) 1.5–5.5 months].

Collection of outcome data

Demographic data for the naturalistic and randomised cohort are presented in Appendix 4, Table 22. A high level of data collection was achieved for each outcome, with completion ranging from 76% to 100% across all follow-up periods. At baseline, 100% of measures were completed, but completion rates dropped as the study progressed, primarily because of participant attrition (n = 3) described above. Of the measures, the Time Use Survey and the Scale to Assess Therapeutic Relationship (STAR) had the lowest completion rates. The completeness of data collection, organised by time point, is presented in Appendix 4, Table 22.

Our blinding procedures, tested at the 12-month follow-up, were successful. Telephone administration of the MANSA, by a researcher who was unaware of the participants’ allocation (supported housing/floating outreach), was completed without revealing the participants’ experimental condition.

Harms/unintended consequences

No harms or unintended consequences occurred during the course of WP4.

Economic evaluation

The CSRI and EQ-5D data were collected at baseline and 12-month follow-up. Owing to the very small numbers, it was not feasible to conduct the health economic analyses.

Qualitative findings

We carried out in-depth qualitative interviews with 11 service user participants and 10 staff [including care co-ordinators who referred participants to the study (n = 6), and those who knew of the study but did not refer participants (n = 4)]. Four main themes emerged from the service user and staff interviews that shed light on the impediments to recruitment: preference for a certain type of supported accommodation, rejection of randomisation, complexity of randomisation and value of the trial.

Preference for a certain type of supported accommodation

Staff interviews revealed a strong belief that supported housing and floating outreach offered quite different support to individuals and they therefore struggled to consider an individual as potentially suitable for either service. They voiced concerns that service users would be insufficiently or inappropriately supported in floating outreach and might be vulnerable to exploitation or at risk of relapse:

I suppose ultimately, when a decision is made to move someone into a . . . to allow them to go into an independent council flat with floating support, clinically we’ve already made the decision that you don’t think . . . it’s going to be a waste of resources for . . . Because there’s clinical reasons why you’d refer someone to a 9 to 5 project. I’d be slightly worried about medication compliance or maybe slightly worried about safeguarding issues.

Staff: 2998. Referrer (unsuccessful). Male

They held a clear conviction that the two models operated as sequential components of a ‘step-down model’, whereby service users move from more highly supported to less supported accommodation as they gain the skills to live more independently. They spoke of how this system provided the opportunity for staff and service users to be confident that the person would be able to manage in an independent tenancy by the time they reached the point of referral to floating outreach:

. . . the structure we’ve got does work quite well because they are in [supported accommodation provider], they stay with the staff, they are tested in the 24 hour, they are tested in the low, and then off to their own flat. It’s not a bad programme really.

Staff: 0020. Referrer (unsuccessful). Female

Although some service users had a clear preference for either floating outreach or supported housing, others appeared to see advantages and disadvantages for both types. This observation applied to participants who agreed to randomisation and those who joined the naturalistic group. Service users who expressed a preference for floating outreach felt that this model would allow them to be more independent and maintain control over their lives:

I’m [forties] years of age, I’m fed up of being monitored. I’m quite able, I can cook. I can clean. I can look after myself. I can wash my clothes. I can have a bath. I can do everything on my own.

Service user: 5010. Naturalistic follow-up (fully independent). Female

Some consented to randomisation simply to increase their chance of being allocated to this form of support. For others, the preference for floating outreach was based on having more control over where they would live as they would be able to ‘bid’ for a tenancy in their preferred area. Sometimes this was in order to be close to friends and family, and sometimes it was part of their ‘risk-management strategy’, to avoid areas with known individuals who might exploit them or offer them illicit substances:

I like to be close to my family, you know, my daughter round, you know, my grandchildren, things like that. I thought [borough] or somewhere like that I’d like to live, if it was like that.

Service user: 4014. Naturalistic follow-up (supported housing). Female

Well I was worried that I would end up in a bad area of town . . . I might get involved in drugs again.

Service user: 5050. Naturalistic follow-up (fully independent). Male

Some service users acknowledged being influenced in their preference by family and staff and some reported previous negative experiences of other forms of supported accommodation and a lack information about supported housing that made them feel more positive about floating outreach as the preferred option.

Service users who expressed a preference for supported housing reported feeling concerned about the risks associated with more independent settings. Often they felt underprepared and expressed anxiety about relapse and ‘moving backwards’ if they were to move to a tenancy with floating outreach support:

I’m not ready for my flat yet, but everyone is saying I’m ready for it, but I’m not ready. I’m ready for it, but I’m ready for it now, but I just want that extra 6 months to make sure that I’m stable. I don’t want to get my flat and become unwell again. It costs the government so much money.

Service user: 2049. Randomised (supported housing). Male

Many seemed influenced by their family’s or support workers’ views that they were not ready for floating outreach and some had previous negative experiences of floating outreach services. Some held a clear preference for a particular supported housing service in a specific location, close to family or friends. Others felt that the lack of available tenancies would mean that they would wait longer for a floating outreach option that supported housing. Some expressed a preference to live among others rather than alone:

I think supported housing is better for some people . . . I prefer supported because you’re surrounded with people.

Service user: 0033. Randomised (supported housing). Male

Service user preferences, a lack of availability of independent tenancies leading to delays in individuals moving to floating outreach services and a perceived lack of staff resources to facilitate service users taking part in the feasibility trial were also cited by staff as impediments to recruitment into the study.

Rejection of randomisation

A number of service users and staff voiced ethics concerns about having accommodation decided by randomisation. Many service users suggested that housing was too important to decide by chance, and staff often reflected that a (perceived) lack of equipoise between supported housing and floating outreach services made random allocation inappropriate:

It’s a bit . . . We’re talking about someone’s home here, do you know what I mean? It’s a base need. It seems like something quite serious to flip a coin about, if you know what I mean?

Service user: 0033. Randomised (supported housing). Male

So, yes I understand the randomisation process, but I would hate to think that it was to the detriment of the well-being of a client in a sense. There must be some clinical judgement based on where that client goes.

Staff: 5010. Referrer (successful). Female

Specifically, staff suggested that the level of support and oversight in supported housing may be oppressive for an individual with relatively limited support needs, whereas a person with higher support needs may not receive the level of support necessary to facilitate recovery if they were randomised to their own tenancy with floating outreach. Service users also indicated that the individual and clinician should have the final say over housing and support arrangements. Similarly, staff voiced concerns about the randomisation aspect of the trial that was felt to negate the need for clinical judgement in deciding on supported accommodation options. Staff also felt that randomisation inappropriately excluded the service user from the process of decision-making about their accommodation.

Complexity of randomisation

Despite providing informed consent for participation at recruitment, one or two service users found it difficult to recall the processes relating to the process of randomisation when undertaking the qualitative interviews some weeks later. Some staff felt that the process of randomisation was too complicated for service users to understand and that this could lead to confusion or disappointment if they were allocated to a service they did not wish to move to. However, some staff also seemed to struggle with understanding the process and acknowledged this difficulty when they had tried to explain the study to potential participants:

The first time I heard about [the trial] I thought maybe it was a platform to find a way of how our clients can get accommodation easily. That’s what I initially thought, but obviously, as you indicated, it’s not about them, it’s about basically the support they can get once they get that accommodation. Yeah. That’s what I thought.

Staff: 0033. Referrer (successful). Male

Value of a trial

Despite the many obstacles to recruitment we encountered, all those who participated in the qualitative interviews felt that it would be worthwhile to carry out a larger trial. Several service users felt that the results from such a study would be of use in understanding which models of supported accommodation worked best and many appeared to gain some satisfaction from feeling that they had contributed to this cause:

It’s helpful; you need to find out things about people who are unwell and to better things in the future to come through us who are unwell. I don’t mind helping that.

Service user: 2017. Naturalistic (supported housing). Male

The majority of service users spoke positively about their experience of taking part in the trial, saying that they had enjoyed it. Staff also indicated that further research had potential to benefit service users, but were vague about how this might happen.

Research priorities

• Assessment of the sensitivity to change and test–retest reliability of the QuIRC-SA.

• International surveys of mental health supported accommodation services and service user characteristics to provide comparative data on the different models in operation in different settings and their target service user group(s).

• Cohort studies to identify the aspects of care and support associated with better outcomes to assist our understanding of how best to focus staff training and resources.

• Qualitative research to complement the findings from surveys and cohort studies to further enhance our understanding of the specific approaches and components of care that constitute ‘best practice’ in supported accommodation from staff and service user perspectives.

• Results from these studies will inform the development of complex interventions to enhance the quality of care delivered in mental health supported accommodation services. These should then be evaluated in terms of their clinical effectiveness and cost-effectiveness through RCTs.

Work package 1

Killaspy H, White S, Dowling S, Krotofil J, McPherson P, Sandhu S, et al. Adaptation of the Quality Indicator for Rehabilitative Care (QuIRC) for use in mental health supported accommodation services (QuIRC-SA). BMC Psychiatry 2016;16:101. https://doi.org/10.1186/s12888-016-0799-4

Sandhu S, Killaspy H, Krotofil J, McPherson P, Harrison I, Dowling S, et al. Development and psychometric properties of the Client’s Assessment of Treatment Scale for Supported Accommodation (CAT-SA). BMC Psychiatry 2016;16:43. https://doi.org/10.1186/s12888-016-0755-3

Work package 2i

Killaspy H, Priebe S, Bremner S, Dowling S, Harrison I, Krotofil J, et al. Quality of life, autonomy, satisfaction, and costs associated with mental health supported accommodation services in England: a national survey. Lancet Psychiatry 2016;3:1129–37. https://doi.org/10.1016/S2215-0366(16)30327-3

Work package 2ii

Killaspy H, Priebe S, McPherson P, Zenasni Z, Greenberg L, McCrone P, et al. Predictors of moving on from mental health supported accommodation in England: national cohort study. British J Psychiatr 2019;10:1–7. https://doi.org/10.1192/bjp.2019.101

Work package 3

Sandhu S, Priebe S, Leavey G, Harrison I, Krotofil J, McPherson P, et al. Intentions and experiences of effective practice in mental health specific supported accommodation services: a qualitative interview study. BMC Health Serv Res 2017;17:471. https://doi.org/10.1186/s12913-017-2411-0

Work package 4

Killaspy H, Priebe S, McPherson P, Zenasni Z, McCrone P, Dowling S, et al. Feasibility Randomised trial comparing two forms of mental health supported accommodation (supported housing and floating outreach); a component of the QuEST (Quality and Effectiveness of Supported Tenancies) study. Frontiers in Psychiatry 2019;10:258. https://doi.org/10.3389/fpsyt.2019.00258