Testing community-based support for people experiencing homelessness who leave hospital early or do not attend hospital after an ambulance is called
| ISRCTN | ISRCTN15217482 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN15217482 |
| Sponsor | University of Edinburgh |
| Funder | Legal and General |
- Submission date
- 24/07/2026
- Registration date
- 17/08/2026
- Last edited
- 08/09/2026
- Recruitment status
- No longer recruiting
- Overall study status
- Ongoing
- Condition category
- Other
Plain English summary of protocol
Background and study aims
People experiencing homelessness often have much poorer health than the general population. They also often face many difficulties with getting healthcare, including difficulty registering with services, attending appointments, or getting follow-up care after hospital.
For various reasons, some people feel the need to leave hospital wards or emergency departments before their care is finished and others also decide not to go to hospital after an ambulance has been called. At the moment, there is little evidence about the best way to support people experiencing homelessness after these situations.
This study is testing whether it is possible to run a larger trial of a community-based support service. The support will be provided by charities specialized in offering support for people experiencing homelessness. The support is designed to help people reconnect with health and social care, housing support, accessing income and other activities or services that may improve wellbeing.
The study will look at whether it is possible to:
1. Recruit people to the study
2. Keep in touch with people during the study
3. Collect study information at the start and after 6 months
4. Deliver the support as planned
5. Understand whether people find the support and the study processes acceptable
This is a pilot study. It will help the research team understand whether the support is acceptable and whether a larger study could be run in the future. It is not designed to prove whether the support works.
Who can participate?
People may be able to take part if they:
1. Are aged 18 years or over
2. Are experiencing homelessness, such as rough sleeping, staying in temporary or hostel accommodation, sofa surfing, or living in insecure or unsuitable housing
3. Have recently left hospital or emergency care early, or decided not to go to hospital after an ambulance attended
4. Are able to give informed consent
What does the study involve?
The study will involve 30-40 adults experiencing homelessness in Edinburgh and Dundee.
People who agree to take part will be randomly put into one of two groups. This is done so that the groups can be compared fairly in a future trial. In this trial we're testing whether this process is acceptable for people.
One group will receive usual care. Usual care means the normal health, social care, housing, homelessness and support provided by charities that are already available.
The other group will receive usual care with the addition of a support worker to help them when they are out of hospital. This support may include help with:
1. Health need
2. Housing
3. Income support and money issues
4. Getting appointments or reconnecting with services
5. Taking part in activities
6. Other practical or social needs identified by the participant
The support will be provided through weekly visits for up to 6 months but can involve more if a person needs.
All participants will be asked to provide information at the start of the study and again around 6 months later. This information will help the research team understand whether it is possible to collect the information needed for a larger trial. The study may ask about areas such as health, quality of life, use of health services, housing, income, drug use, overdoses, and confidence in managing health.
Some participants and staff may also be invited to take part in interviews. These interviews will ask about their experiences of the study and, where relevant, the support provided. This will help the research team understand what worked well, what was difficult, and how the support or study procedures could be improved.
What are the possible benefits and risks of participating?
People who receive the extra support may benefit from help to access health and social care, housing support, benefits advice or other services. However, this cannot be guaranteed.
People in either group may value the opportunity to share their experiences and contribute to research that may improve services for people experiencing homelessness in the future.
There are some possible risks or disadvantages. Some questions may feel personal or upsetting, for example, questions about health, housing, drug use, overdose, or difficult experiences with services. People that take part do not have to answer any question they do not want to answer and can take a break or stop taking part at any time.
Taking part will involve giving some time to complete study questions and, for some people, additional interviews. The research team will take steps to protect participants’ privacy and keep their information confidential, unless there is a serious concern about someone’s safety or the safety of others.
Taking part is voluntary. Choosing not to take part, or choosing to leave the study later, will not affect the usual care or support that someone can receive.
Where is the study run from?
The study is run from the University of Edinburgh (UK). Participants will be recruited in Edinburgh through homelessness services, third-sector organisations and community outreach.
When is the study starting and how long is it expected to run for?
August 2026 to April 2027
Who is funding the study?
This research was funded by the Legal and General Group (research grant to establish the independent Advanced Care Research Centre at the University of Edinburgh). The funder has no role in the conduct of the study. The views expressed are those of the authors and not necessarily those of Legal and General.
Who is the main contact?
Christian Newman, christian.newman@ed.ac.uk
Contact information
Principal investigator, Public, Scientific
Old Medical School - Doorway 6
Teviot Pl
Edinburgh
EH8 9AG
United Kingdom
 ORCID ID |
0009-0000-4674-2059 |
|---|---|
| Phone | +44 (0)131 650 1000 |
| christian.newman@ed.ac.uk |
Study information
| Primary study design | Interventional |
|---|---|
| Allocation | Randomized controlled trial |
| Masking | Open (masking not used) |
| Control | Active |
| Assignment | Parallel |
| Purpose | Health services research, Supportive care |
| Participant information sheet | 49991_PIS_Trial EDI_V5_09June2026.pdf |
| Scientific title | Hospital Admission support for people who are Underserved, No fixed abode and Early/iRRegular discharge Study (HAUNERRS): a pilot randomised controlled trial with qualitative investigation |
| Study acronym | HAUNERRS |
| Study objectives | Primary research question: Does the pilot RCT achieve a priori progression criteria for a subsequent definitive randomised controlled trial? Progression criteria will be assessed across four feasibility domains: recruitment, retention, data collection, and intervention fidelity: 1. 30 participants recruited within 2 months 2. At least 80% of participants retained in the study (receiving follow-up data collection in person) at 6 months after recruitment 3. At least 70% of data collected from participants (combining baseline and follow-up) 4. At least 70% of participants in the intervention arm receiving the intervention as planned? Progression criteria will be assessed using a traffic-light system of green (go), amber (proceed with modification), and red (do not proceed) across the four feasibility domains: 1. Recruitment (number randomised within 2 months): ≥30 (green), 15–29 (amber), <15 (red) 2. Retention (proportion with any follow-up data at 6 months)*: ≥80% (green), 60–79% (amber), <60% (red) 3. Data collection proportion of planned outcome measures completed across all randomised participants at baseline and follow-up: ≥70% (green), 50–69% (amber), <50% (red) 4. Intervention fidelity (proportion of intervention‑arm participants receiving the intervention as planned, according to pre‑specified minimum dose)**: ≥70% (green), 50–69% (amber), <50% (red) * Denominator: all randomised participants excluding deaths during follow-up, which are recorded separately. In-person and phone follow-up both count; in-person proportion is reported as a secondary metric. **Minimum dose defined as 24 contacts over 6 months, including ≥1 contact within 14 days of randomisation, assessed via contact logs and fidelity checklist. Reported by site and pooled. Progression to a definitive RCT will be recommended if all four criteria are green or if ≤1 criterion is amber and the Trial Steering Committee agrees a clearly specified plan. Progression will not be recommended if any criterion is red, or if ≥2 criteria are amber and qualitative evidence indicates systemic feasibility barriers. The Trial Steering Committee's recommendation will also integrate qualitative findings on acceptability and implementability and will not recommend progression in the presence of unexpected safety signals regardless of quantitative feasibility outcomes. Secondary research questions: A number of secondary research questions will explore key uncertainties relating to the feasibility of trial procedures and the intervention. These include: 1. Are recruitment, consent, randomisation and follow-up procedures feasible and acceptable to participants, staff and stakeholders? 2. How often are potentially eligible individuals not approached at first contact or not able to consent, and what are the reasons for this? 3. Is the HAUNERRS intervention acceptable to participants, staff and stakeholders? 4. Can HAUNERRS be delivered as intended by third-sector staff, and what barriers and facilitators affect delivery? 5. How do participants experience patient-directed discharge or ambulance refusal and subsequent access to healthcare? 6. What contextual factors and perceived mechanisms may influence intervention engagement and outcomes? 7. Is it feasible and acceptable to collect candidate health, healthcare-use, and social-care outcomes at baseline and six months? Data collection will also involve a number of outcomes to test their feasibility as primary outcomes for a future definitive trial and to inform future sample size calculations. Candidate outcome data will be summarised descriptively; this study is not designed or powered to test intervention effectiveness. Candidate outcomes include GP registration status, unscheduled secondary care use, non-fatal overdose, engagement with mental health treatment or support, alcohol use, non-prescribed substance use, housing status and stability, health-related quality of life (EQ-5D-5L), Health Confidence (HCS), all-cause mortality and cause-specific mortality where ascertainable. |
| Ethics approval(s) |
Approved 20/07/2026, Health in Social Science Research Ethics Committee, University of Edinburgh (Old Medical School, Elsie Inglis Quadrangle, Teviot Pl, Edinburgh, EH8 9AG, United Kingdom; +44 (0)131 651 3969; ethics.hiss@ed.ac.uk), ref: CAHSS2603/16 |
| Health condition(s) or problem(s) studied | Community follow-up and access to health and social care for adults experiencing homelessness who leave hospital or emergency care early, or decline ambulance transfer to hospital |
| Intervention | Participants in this two-arm pilot randomised controlled trial will be randomly allocated in a 1:1 ratio to either (a) usual care or (b) usual care plus HAUNERRS, using simple randomisation with no stratification or blocking. Simple randomisation is considered appropriate because the primary aim of the pilot is to assess feasibility rather than estimate intervention effectiveness, and because block randomisation at this sample size risks allocation predictability in an open-label trial. Prior to the commencement of recruitment, a single random allocation sequence will be generated by an independent researcher affiliated with the University of Edinburgh, who will have no involvement in participant recruitment, eligibility assessment, intervention delivery, or outcome assessment at any stage of the trial. The allocation sequence will be generated using R statistical software and will cover the full anticipated sample across both sites. The randomisation schedule will be produced in full before the first participant is recruited at either site. Allocation concealment will be maintained using sequentially numbered, opaque, sealed envelopes prepared in advance by the same independent researcher. Envelopes will be identical in external appearance, sealed with tamper-evident tape, and signed across the seal. Prepared envelopes will be divided between the Edinburgh and Dundee sites prior to recruitment and stored securely at each site in a locked filing cabinet. Following confirmation of eligibility, completion of written informed consent, and completion of all baseline measures, the recruiting researcher will open the next envelope in the local numerical sequence. The recruiting researcher will have no access to the allocation sequence in advance; allocation concealment is maintained by the tamper-evident sealing of envelopes and the pre-prepared sequential numbering. Participants will be informed of their allocation immediately following the opening of the envelope. For participants allocated to the active arm, the HAUNERRS intervention will commence as soon as practicable following allocation. All opened envelopes will be retained for audit. Any baseline imbalance between trial arms arising by chance will be described transparently in the pilot trial report. Usual care (control arm): Usual care is the standard health and social care support that participants would ordinarily receive from statutory services and third-sector organisations, and differs between the two trial sites. In Edinburgh, hospital homelessness in-reach is provided by Cyrenians, a third-sector organisation with practitioners working across three acute hospital sites and supporting approximately 150–200 unique individuals per month. During inpatient admission, Cyrenians practitioners support people experiencing homelessness by building relationships, providing practical support (e.g., access to a phone or personal items), offering income maximisation and clinical advocacy, coordinating care between hospital and community services, and planning for discharge. Participants in Edinburgh may also be referred to Milestone House, which provides step-down intermediate care following hospital discharge. While Cyrenians deliver some post-discharge outreach in the community, capacity constraints mean this does not occur routinely. When a participant self-discharges, coordinated discharge planning is disrupted; Cyrenians or hospital staff may subsequently refer the individual to third-sector outreach support in the community, but there is no standardised post-PDD follow-up pathway in Edinburgh. Dundee does not have a dedicated hospital homelessness in-reach service. Following discharge, community support may be facilitated by a care navigator, who supports individuals to re-engage with community-based services. Community health support for people experiencing homelessness in Dundee is also available through Parish Nursing Dundee, based in the city centre. Parish Nursing operates twice-weekly drop-in sessions available to anyone seeking support with health concerns, harm reduction, and recovery, as well as one-to-one support outside drop-in hours and support groups for people in recovery. These services are delivered by a team of three registered nurses, a recovery support worker, an assistant, and trained volunteers, working in collaboration with the NHS Homeless Health Team and the NHS Mobile Dental Team. The existing Parish Nursing drop-in and community services are available to all Dundee trial participants irrespective of allocation and are considered part of usual care in Dundee. There is no standardised post-PDD follow-up pathway in Dundee. HAUNERRS intervention (intervention arm): Participants allocated to the intervention arm will receive usual care plus HAUNERRS: up to six months of flexible, assertive outreach support delivered in the community following patient-directed discharge (PDD). In Edinburgh, HAUNERRS will be hosted and delivered by Simon Community Scotland from the Edinburgh Support Hub. In Dundee, HAUNERRS will be hosted and delivered by Parish Nursing Dundee from The Steeple. Usual care services at both organisations will remain available to all participants regardless of allocation; HAUNERRS support is in addition to usual care. Participants will be offered weekly planned one-to-one contact with a HAUNERRS support worker, with additional unscheduled support provided as required. Support is expected to be more intensive in the period immediately following PDD and to taper over time as participant stability increases and connection to longer-term community-embedded support is established. Support will be delivered primarily through in-person contact in community settings chosen by the participant (e.g., accommodation settings, cafés, day centres, community venues). Participants may also access ad hoc support by dropping in to the Edinburgh Support Hub (Edinburgh) or the Parish Nursing drop-in at The Steeple (Dundee). Flexible communication methods, including telephone calls and text messaging, will be used in line with participant preference and access. Support may additionally be delivered in health settings, including accompaniment and advocacy at outpatient or hospital appointments and, where relevant, within hospitals to support participants during readmission. The components of the HAUNERRS intervention are as follows, each delivered in a tailored and flexible manner according to participant need and preference: Following allocation, a support worker will conduct a brief non-clinical immediate needs assessment to identify urgent priorities and co-produce a support plan. Support plans will be revisited and updated iteratively throughout the intervention in line with participants' goals and changing circumstances. Participants will be supported to re-engage with appropriate health services (e.g., GP/primary care, community pharmacy, community nursing, homeless health services, specialist services) with accompaniment and advocacy at appointments. Practical barriers to engagement will be addressed, including transport, GP registration, phone access, and support to understand discharge documentation. Facilitation of return to emergency departments will occur when clinically indicated. Support workers will provide advocacy and navigation support in relation to housing needs, tailored to the individual's circumstances. Throughout the intervention, support workers will remain alert to safeguarding concerns (including abuse, neglect, and self-neglect) in line with Adult Support and Protection procedures in Scotland. Where concerns are identified, the support worker will take a person-centred approach and seek to agree next steps with the participant wherever possible. Support workers will follow their host organisation's safeguarding policies and will escalate to appropriate services (e.g., NHS services, local authority Adult Support and Protection pathways, Police Scotland, or emergency services) where risk is immediate or significant. Support workers will provide non-clinical medication support, including facilitating access to community pharmacy, supporting prescription replacement where lost, planning for safe storage, and encouraging adherence through reminders and check-ins. Support workers will not administer or store medicines on behalf of participants. Support workers will assist with banking, welfare, benefits and personal administration as required and will provide referrals for food and clothing bank vouchers and signpost participants to hot meal provision. Support workers will provide harm reduction advice and supply necessary materials, including Naloxone, injecting equipment, and foil. They will also support participants to access harm reduction services, including opioid substitute treatment, blood-borne virus medication, and drug checking services. Where participants express interest, support workers will signpost to social prescribing initiatives and activities offered by the host organisation and will accompany participants to activities where needed to reduce barriers to access. Participants will also be signposted to employment and volunteering opportunities via relevant local schemes. Support workers will assess participants' digital access needs and provide tailored support within their remit, including signposting to devices and digital skills services, to facilitate access to online resources such as banking, appointment booking, and welfare services. Intervention materials will include co-produced support plans, information on local services, small discretionary funds (e.g., for hot drinks, basic amenities, or transport to enable appointment attendance), and digital inclusion resources (e.g., access to phones, tablets, or laptops). Harm reduction materials, including Naloxone kits and injecting equipment, will also be provided. The intervention will be delivered by third-sector homelessness support workers from Simon Community Scotland (Edinburgh) and Parish Nursing Dundee (Dundee). Both organisations are experienced in delivering low-threshold, trauma-informed, person-centred support to people experiencing homelessness. Support workers will hold Disclosure Scotland checks and will have completed training in psychologically informed environments, trauma-informed care, Naloxone provision, and injecting equipment provision. |
| Intervention type | Behavioural |
| Primary outcome measure(s) |
2. Retention. The trial retention rate will be measured as the proportion of randomised participants alive and with capacity at 6 months, who have not withdrawn consent at any point during the study period or been lost to follow-up, reported overall and separately by arm. 80% target. Secondly, the follow-up completion rate: the proportion of randomised participants who participate in a follow-up outcome assessment at 6 months, whether in person or by telephone, reported overall and separately by arm. |
| Key secondary outcome measure(s) |
1. Intervention fidelity will be assessed primarily by a researcher reviewing session logs and case records (both completed by HAUNERRS practitioners) for a randomly selected 50% of intervention participants and compare these against a checklist of HAUNERRS core components and minimum dose criteria. This will be conducted iteratively throughout the 6 months to inform intervention refinement and trial progression. |
| Completion date | 30/04/2027 |
Eligibility
| Participant type(s) | |
|---|---|
| Age group | Mixed |
| Lower age limit | 18 Years |
| Upper age limit | 116 Years |
| Sex | All |
| Target sample size at registration | 40 |
| Key inclusion criteria | 1. Aged 18 years or over 2. Currently experiencing homelessness, including rough sleeping, temporary or hostel accommodation, sofa surfing, or insecure/inadequate housing 3. Self-reported self-discharge or early departure from hospital inpatient care or an emergency department, or declined ambulance transfer to hospital after ambulance attendance, within the past 2 months 4. Able and willing to provide informed consent to participate |
| Key exclusion criteria | 1. Has a permanent accommodation that they consider suitable and fit to return to after the relevant hospital, emergency department, or ambulance event 2. Is currently accommodated in a facility that provides 24-hour medical care 3. Lacks capacity to provide informed consent 4. Is currently in prison or other custodial detention 5. For the qualitative interview component only: lacks sufficient English comprehension to take part in an interview in English |
| Date of first enrolment | 07/09/2026 |
| Date of final enrolment | 30/10/2026 |
Locations
Countries of recruitment
- United Kingdom
- Scotland
Study participating centres
Edinburgh
EH8 8AF
Scotland
Dundee Law Parish Church
Nethergate
Dundee
DD1 6DG
Scotland
Results and Publications
| Individual participant data (IPD) Intention to share | No |
|---|
Study outputs
| Output type | Details | Date created | Date added | Peer reviewed? | Patient-facing? |
|---|---|---|---|---|---|
| Participant information sheet | version 5 | 09/06/2026 | 27/07/2026 | No | Yes |
| Participant information sheet | version 5 | 09/06/2026 | 27/07/2026 | No | Yes |
| Protocol file | version 6.1 | 14/07/2026 | 14/08/2026 | No | No |
Additional files
- 49991_PIS_Qual Interview (Participant)_V5_09June2026.pdf
- Participant information sheet
- 49991_PIS_Trial EDI_V5_09June2026.pdf
- Participant information sheet
- 49991_Protocol_V6.1_14Jul2026.pdf
- Protocol file
Editorial Notes
08/09/2026: The following changes were made to the study record:
1. The date of first enrolment was changed from 25/08/2026 to 07/09/2026.
2. The date of final enrolment was changed from 28/09/2026 to 30/10/2026.
3. The completion date was changed from 15/02/2027 to 30/04/2027.
12/08/2026: Study's existence confirmed by the University of Edinburgh.
