Housing as Healthcare
Abstract Format : Research has found that chronically homeless individuals in permanent supportive housing (PSH) are able to maintain their housing, improve their quality of life, and decrease emergency department use, substance use and days incarcerated. Still, the effects of different PSH and HIV Housing models and resident characteristics on health outcomes have not been examined. This is despite the fact that, in less than five years, PSH units in the US have increased from 176,830 to 236,798. The research reported here sought to describe components of different PSH models, compare differences in housing models between people living with HIV (PLH) and those without HIV, and the effects of different PSH models and other factors on optimal ART adherence among PLH.
Methods : In phase 1, we conducted in-depth interviews with 32 directors and 33 case managers from 30 agencies that provide PSH in the Chicago area. Interviews followed a semi-structured guide to explore how services were provided and the type of housing provided (i.e. in rental units that the participant found, scattered site, or in a project owned building, project based). Qualitative content analysis was performed using an iterative consensus based approach. We then recruited 888 supportive housing residents to complete surveys at baseline, 6-, 12- and 18-months.The sample was stratified in order to sample sufficient numbers of participants from different housing types. Survey measures included: demographics, health quality of life, psychiatric and anxiety symptoms, length housed and longest period of homelessness, arrest, social support, and, for those with HIV, ART adherence and HIV medical care within the last year. Chi-square analyses were run to determine the relationship between having HIV and housing type and health quality of life. For PLH, chi-square analyses were conducted to determine the relationship between demographic characteristics, social support, housing type, depression, housing satisfaction and anxiety symptoms and ART adherence. Factors that were significant in bivariate analyses were included in linear regression model with suboptimal ART adherence as the outcome of interest.
Results : Qualitative data indicated 6 basic types of supportive housing models in Chicago based on housing configuration (project-based or scattered site) and service provision (low intensity case management with case worker to client ratios of 1:50, high intensity case management with a case worker to client ration of 1:15, and behavioral health models which used Assertive Community Treatment or other clinical team based approaches). . We recruited 888 participants from 23 supportive housing agencies, of which 8 have significant experience and histories of HIV housing, and are currently completing follow-up interviews at 6-, 12- and 18 months. PLH were significantly more likely to live in fixed site residences, and to report better access to health care, higher health quality of life and lower reported pain compared to those without HIV (p<.05). PLH in supportive housing were, in general, highly adherent to ART (with 65% reporting taking over 95% of their medications in the last month) and HIV medical care. Significant associations with suboptimal HIV adherence included African American race, lower age, fewer people who can help them with transportation,, and depression. PSH type, housing configuration and supportive service type were not significantly associated with ART adherence. In regressions, younger age, Black race, and depression remained significant predictors of suboptimal ART adherence.
Conclusions : While PSH residents with HIV generally were on ART and saw their HIV doctors regularly, nearly 40% still were not optimally adherent. Results suggest that additional interventions are needed to address depressive symptoms and lack of social support, such as barriers to transportation.