Investigating the Comparative Effectiveness of Place-Based and Scatter-Site Permanent Supportive Housing for People Experiencing Homelessness During the COVID-19 Pandemic: Protocols for a Mixed Methods, Prospective Longitudinal Study

Abstract

Background Permanent supportive housing (PSH) is an evidence-based practice to address homelessness that is implemented using 2 distinct approaches. The first approach is place-based PSH (PB-PSH), or single-site housing placement, in which an entire building with on-site services is designated for people experiencing homelessness. The second approach is scatter-site PSH (SS-PSH), which uses apartments rented from a private landlord while providing mobile case management services. Objective This paper describes the protocols for a mixed methods comparative effectiveness study of 2 distinct approaches to implementing PSH and the patient-centered quality of life, health care use, and health behaviors that reduce COVID-19 risk. Methods People experiencing homelessness who are placed in either PB-PSH or SS-PSH completed 6 monthly surveys after move-in using smartphones provided by the study team. A subsample of participants completed 3 qualitative interviews at baseline, 3 months, and 6 months that included photo elicitation interviewing. Two stakeholder advisory groups, including one featuring people with lived experience of homelessness, helped guide study decisions and interpretations of findings. Results Study recruitment was supposed to occur over 6 months starting in January 2021 but was extended due to delays in recruitment. These delays included COVID-19 delays (eg, recruitment sites shut down due to outbreaks and study team members testing positive) and delays that may have been indirectly related to the COVID-19 pandemic, including high staff turnover or recruitment sites having competing priorities. In end-July 2022, in total, 641 people experiencing homelessness had been referred from 26 partnering recruitment sites, and 563 people experiencing homelessness had enrolled in the study and completed a baseline demographic survey. Of the 563 participants in the study, 452 had recently moved into the housing when they enrolled, with 272 placed in PB-PSH and 180 placed in SS-PSH. Another 111 participants were approved but are still waiting for housing placement. To date, 49 participants have been lost to follow-up, and 12% of phones (70 of the initial 563 distributed) were reported lost by participants. Conclusions Recruitment during the pandemic, while successful, was challenging given that in-person contact was not permitted at times either by program sites or the research institutions during COVID-19 surges and high community transmission, which particularly affected homelessness programs in Los Angeles County. To overcome recruitment challenges, flexible strategies were used, which included extending the recruitment period and the distribution of cell phones with paid data plans. Given current recruitment numbers and retention rates that are over 90%, the study will be able to address a gap in the literature by considering the comparative effectiveness of PB-PSH versus SS-PSH on patient-centered quality of life, health care use, and health behaviors that reduce COVID-19 risk, which can influence future public health approaches to homelessness and infectious diseases. Trial Registration ClinicalTrials.gov NCT04769349; https://clinicaltrials.gov/ct2/show/NCT04769349 International Registered Report Identifier (IRRID) DERR1-10.2196/46782