Assessing Programmatic and Demographic Risk Factors for COVID-19 Infection Among People Experiencing Homelessness in the United States

Background and Purpose: People experiencing homelessness are at high risk of COVID-19 infection and related adverse outcomes such as hospitalization and death due to the prevalence of congregate shelters and the advanced age and chronic health conditions facing many people experiencing homelessness. Research assessing incidence of COVID-19 infection have been limited to studies from early in the pandemic examining a few shelters in major U.S. cities, with little focus on rural communities or programmatic or demographic risk factors for infection. We leverage administrative data from homeless service providers across the country to assess programmatic and demographic risk factors for COVID-19 infection among people experiencing homelessness and those in permanent supportive housing.

Methods: This is an observational study using Homeless Management Information System (HMIS) data from a sample of homeless service providers across the United States that includes service use, demographic, and COVID-19 status information for adults who accessed emergency shelter, transitional housing, or permanent supportive housing following the onset of the COVID-19 pandemic. We conduct bivariate analyses and multivariate logistic regression to understand predictors of COVID-19 infection, including race, ethnicity, age, gender, household type, size of the Continuum of Care, and the type of shelter or housing being accessed. Our study includes records for more than 5,000 adults who accessed emergency shelter, transitional housing, and permanent supportive housing in 20 Continuums of Care scattered across urban and rural communities between March 1, 2020, and June 30, 2022.

Results: Emergency shelter users were at the highest risk of COVID-19 infection, being about 7 times as likely as their peers to become infected. Transitional housing was more protective than permanent supportive housing. Individuals using adult-only shelters had a risk of COVID-19 infection twice as high as those accessing shelters for families with children. Individuals identifying as American Indian or Alaskan Native and Native Hawaiian or other Pacific Islander had the highest rates of COVID-19 infection, while Black shelter users had the lowest rates. Hispanic service users were 1.5 times as likely to contract COVID-19 compared to their non-Hispanic peers. There were no significant associations between a Continuum’s COVID-19 infection rate and its size or whether it was urban or rural. Older adults had the highest risk of COVID-19 infection; those 55 and older were 1.5 times more likely to be infected than adults under 35 years of age.

Conclusions and Implications: COVID-19 among homeless and formerly homeless adults is a concern across geographies and those at greatest risk of the worst outcomes, notably older adults experiencing stays in emergency shelters, were most likely to be infected. As COVID-19 persists and other infectious respiratory diseases circulate while homeless service systems return to traditional congregate shelter models, governments and homeless service systems must emphasize the safety of highly vulnerable populations. This can be done through priority placement into permanent housing and alternate, non-congregate shelter models like hotels that were prevalent and supported by the federal government during the height of the COVID-19 pandemic.