A Cornell-led study is the first to report national, annual rates of sheltered homelessness over time across race and ethnicity and finds “staggering” disparities, showing Black and Indigenous Americans are far more likely to experience homelessness than other groups.
From 2007 to 2017, the lowest rate of homelessness among Black Americans – nearly 1.4% – was more than four times the highest prevalence for white individuals and families, the research team found. The lowest rates for American Indian or Alaska Native and the Native Hawaiian and Pacific Islander communities were at least double the highest rates for white and Asian people.
Particularly concerning, the researchers said, was that the disparities persisted or increased over the 11-year study period dating to the Great Recession – and likely were exacerbated by the pandemic.
The findings highlight an urgent need for more comprehensive data to inform policies targeting those most at risk of homelessness, said , assistant professor in the Department of Public and Ecosystem Health in the College of Veterinary Medicine.
“This is the first time national data has been reported on trends in homelessness across many different demographic groups over time,” Willison said. “To successfully address homelessness and reduce health inequities, we must know the extent of the problem, how it varies across groups and be able to measure it.”
Willison is the lead author of “,” published Feb. 16 in the Journal of Racial and Ethnic Health Disparities.
Co-authors are Naquia Unwala, research associate in Cornell’s ; Phillip Singer, assistant professor of political science at the University of Utah; Timothy Creedon, research associate at Cambridge Health Alliance; and two members of Cambridge Health Alliance’s Health Equity Research Lab: Senior Data Scientist Brian Mullin and Director Benjamin Lê Cook, an associate professor at Harvard Medical School.
Their findings were published not long after the U.S. Department of Housing and Urban Development (HUD) conducted its annual count of people without homes, tallied by shelter systems across the country on a night in January. Because data is collected on a single day, Willison said, the so-called Point-in-Time counts fail to capture the full extent of homelessness, of racial disparities and of changes over time across multiple years, which HUD does not report.
“It obscures variability across an annual trend, which is the measure we use for all other types of health outcomes,” Willison said.
To address that issue, the researchers mined archived HUD data that tracks how many people visited a shelter at some point in a year, noting their race and ethnicity. The Թվless Management Information System (HMIS) data – which Willison said is difficult to find and not reported on longitudinally across all races – was paired with census data to produce estimates of the proportion of different populations experiencing homelessness annually.
While still an undercount since it is limited to people in shelters, the researchers said, the new estimates are a more comprehensive measure of the burden of homelessness nationally.
For example, in 2017, HUD counted more than half a million people experiencing homelessness on a single night in January. In contrast, the new study leveraging HMIS data shows that nearly 600,000 Black Americans alone visited shelters that year – out of more than 1.4 million total visitors.
One potentially encouraging trend: a declining rate of homelessness among Hispanic or Latino individuals. But concerns over immigration status might have limited shelter visits, leading to less visible forms of housing insecurity, the study cautioned.
The researchers said the 2021 federal infrastructure bill presents an opportunity to reduce disparities by expanding access to affordable housing – if the spending is focused on the populations most in need. They said restrictive zoning laws and policies also must be addressed.
The authors call for federal and state agencies to improve systematic data collection, saying homelessness deserves the same attention as other social determinants of health. Improving public reporting of HMIS data would be a good start, Willison said.
Finally, the authors propose incentives to reform a complex and highly fragmented governance structure that has diffused responsibility for homelessness policy across nongovernmental service providers – known as Continuums of Care – and local, state and federal governments.
Those issues are the focus of Willison’s 2021 book, “,” and of .
“If we want to promote health equity as a goal and make sure some groups aren’t disproportionately likely to experience these adverse outcomes that are harmful across the life course,” Willison said, “we need to design more equitable policies.”
