Researchers at NYU have developed a Housing Health Index that would allow local governments to target building inspections with a focus on improving health. A peer-reviewed study published today (Feb. 5) in the February issue of Health Affairs reports on this innovation.
As detailed in the study, entitled “Housing-Sensitive Health Conditions Can Predict Poor Quality Housing,” the researchers linked New York City Medicaid utilization claims data to the NYC Public Advocate’s Landlord Watchlist data and, using machine learning, identified 23 health conditions correlated with a building’s presence on the Watchlist. The “housing-sensitive” health conditions are grouped into five broader categories: asthma, cardiovascular, mental health, substance use, and injuries.
Using the Housing Health Index, local government housing agencies would be able to rank residential buildings by the likelihood that possible code violations were negatively affecting residents’ health—and could target inspections accordingly.
“Although our study does not attempt to assess causality,” the study authors write, “we show that the existence of these health conditions is predictive of poor building quality. Buildings with residents who have a high number of housing-sensitive health conditions (measured using our housing health index) also have many housing quality deficiencies. They are not only more likely to end up on the Landlord Watchlist, but also have increased numbers of housing code violations, emergency repairs, and 311 complaints.”
The authors are from New York University’s Robert F. Wagner Graduate School of Public Service and NYU Furman Center for Real Estate and Urban Policy. The study was led by Professor Daniel B. Neill of NYU Wagner, the director of NYU’s Machine Learning for Good Laboratory.
Professor Neill is available to be interviewed; he can be reached through the NYU press officer listed with this news release.
[DOI: 10.1377/hlthaff.2023.01008, Health Affairs, 43:2 (2024).]
