Peer to Community (P2C) Model: A Pilot Randomized Controlled Trial

The investigators will conduct an open randomized controlled trial to pilot and evaluate the P2C model. The purpose of this pilot is to not only pilot the P2C intervention to determine the feasibility, acceptability, and initial effectiveness of the model, but also to pilot the study design to inform future, larger scale, implementation science research. While the study team will enter this pilot with the intention to complete a randomized controlled trial design, they will use a quasi-experimental design should the participants decline randomization. At the end of the trial, the study team will use the knowledge generated to inform refinements to the P2C model for future implementation science research.

Baseline data collection: Once all participants have been referred, research assistants will be provided with the names and contact information of potential participants. Research assistants will conduct interviews in person and read all interview components aloud to overcome threats to validity and reliability associated with illiteracy. These interviews will be composed of demographic elements (age, gender, sexual orientation, race, ethnicity, months housed/unhoused in the past year, health conditions, income), and a range of standardized measures that correspond with key constructs including community integration (primary outcome), meaningful activity engagement, mental well-being and substance use. Research assistants will use a visual stimulus when delivering these measures to reduce respondent fatigue. The team has used this strategy with a high degree of success in previous research when a large number of items is presented to individuals living with mental illness, substance use, or concurrent disorders and experiences of homelessness.

Qualitative interviews: In addition to demographic and quantitative items, the study team will deliver a baseline qualitative interview with participants in both UC and P2C groups to uncover participants' experiences of CI in the past and present. Interviews will be semi-structured, recorded on a digital recording device, and transcribed verbatim.

Recruitment, group assignment, and sample size estimation: The investigators will recruit participants through advertisements placed in the common areas of shelters, housing case management programs, and health care services in the recruitment city (Kingston, Ontario, Canada). The study team will also engage service providers in these organizations in our recruitment process by asking them to share information about this research with service users. Participants will be enrolled over a one-month period and will be included if they are: over the age of 16; have secured a tenancy within the last three months or are in the midst of moving into a tenancy; have experienced homelessness for at least one month in the past year; and acknowledge living with a mental illness and/or substance use disorder. The study team will assign participants to one of two groups: 1) usual care (UC); and 2) P2C. The UC group will receive typical services available in the community. The P2C group will be engaged in the P2C model in addition to any other supports available in their community. Participants will be assigned to groups that have been carefully matched based on gender, age, health condition, number of months of housed/unhoused in the past year, and income. The investigators will enroll 20 participants in each group (n=40 total). This sample size has been determined by modelling after sample sizes recommended for pilot research, with an attrition estimate of 15% [51].

Implementation: Once baseline interviews have been completed and participants have been assigned to P2C or UC groups, program staff will implement the P2C intervention with the P2C group. PSSs will be provided with the names and contact information for all participants in this group and will initiate the process of making contact with each within the first week to ensure that service for all participants begins at the same time. PSSs will initiate the development of relationships with each participant, while OT and SW at each site will ensure their availability for two days per week each for consultation. PSSs will meet with participants in the P2C group in community locations at least once per week. PSSs, OT and SW at each site will meet weekly for personal and professional support meetings as described in the overview of the P2C intervention in this proposal. The intervention and associated data collection will continue for a one-year period.

Data collection over the intervention period with UC and P2C participants: Research assistants will collect data every three months over the one-year intervention period with both UC and P2C groups. Quantitative interview components delivered with both UC and P2C groups at baseline will remain the same to facilitate comparison across time points and groups. Program staff will record encounters with P2C participants on a common reporting system associated with the collaborating organizations (Providence Care, Kingston, ON; Trellis HIV and Community Care, Kingston, ON) to enable our team to capture data related to frequency and nature of service interactions. Qualitative interviews will be delivered every six months with both groups (baseline, 6-months, 12-months). This will help us to uncover experiences of CI for the UC group over a one-year period without support, and to understand CI experiences and acceptability of the intervention for P2C participants. Qualitative interviews will be semi-structured, recorded on a digital recording device, and transcribed verbatim.

Data collection over the intervention period with P2C program staff: At baseline and every six months during the one-year intervention period, the study team will deliver qualitative interviews with PSSs, OTs and SWs. The purpose of these interviews will be to identify program staff experiences of delivering this approach, acceptability to program staff, and to determine any challenges encountered with implementation (feasibility). Interviews will be semi-structured, recorded on a digital recording device, and transcribed verbatim.

Data collection with service providers in community services: At the end of the one-year intervention period, 15 participants working in health and social service agencies that provide supports to individuals experiencing homelessness from each site will be recruited for qualitative interviews aimed at identifying community service provider perspectives on the P2C intervention, acceptability of the approach in the community, and feasibility of implementing the approach in the future. Interviews will be semi-structured, recorded on a digital recording device, and transcribed verbatim. This sample size has been modelled after sample sizes used in similar published research [61].

Fidelity measurement: Every three months, research assistants will observe one session delivered by each program staff to determine the extent to which services are aligned with the central tenets of the P2C model using the fidelity checklist developed in Phase II. These observations will continue over the 12-month intervention period. Research assistants will be asked to observe a session with a different participant during each encounter with program staff.

Strategies for reducing attrition and withdrawal among UC and P2C groups: The study team recognizes that individuals who live in poverty and have experiences of homelessness may struggle to remain in contact with our research team due to changing circumstances. To reduce attrition, the study team will ask interested participants at the stage of recruitment for a range of contact sources for follow up including social media contact information, phone numbers, email addresses, and contact information for family, friends and other health and social care professionals to facilitate follow-up. Research assistants will also remain in regular contact with UC and P2C participants and program staff to schedule interviews, provide reminders throughout the study and remind participants of the benefits of participation. These strategies have been described to support retention in previous implementation science research.

Analysis of quantitative data: The study team will calculate internal consistency scores for all standardized measures, and descriptive statistics for all demographic variables. To compare the effectiveness of the P2C intervention over time, the study team will conduct parametric (repeated measures and independent samples t-tests, multi-level linear models) and non-parametric (Wilcoxon signed rank and Kruskal-Wallis tests) statistics to compare participant scores on all standardized measures across time points with both UC and P2C participants. This will enable the study team to identify any mean differences and change on indices of CI, mental well-being, meaningful activity engagement, substance use, and service contacts across data collection points and groups, noting any statistically significant differences within and between groups. The investigators will also separate groups based on gender to conduct gender-based analyses to identify any differences in outcomes across gendered groups. Using a combination of parametric and non-parametric statistics (t-tests, Mann-Whitney U, and chi-square), the investigators will compare the demographic characteristics of any participants lost to follow-up with participants remaining in the study to determine the presence of any significant differences between these groups. This will enable us to uncover any demographic explanations for attrition. Significance will be set to p≤.05 for all tests.

Analysis of qualitative data: Qualitative interview transcripts conducted with UC and P2C groups, program staff and community service providers will be uploaded to Dedoose, a qualitative data management program for coding and analysis. Transcripts will be coded abductively, informed by a social ecological lens, using thematic analysis [64]. Themes representing each data set will be generated during this process. In line with thematic analysis, the investigators will generate a central essence describing each set of themes in concert with the approach described by Braun and Clarke. Using the same processes, the study team will separate qualitative interview transcripts by site and gender to conduct site and gender specific analyses for all groups.