By Ingrid Gould Ellen, Kacie L. Dragan, and Sherry Glied, Housing Matters, April 22, 2020
In 2015, the New York City Housing Authority transferred six public housing properties to private developers for rehabilitation and operation as subsidized housing. The repairs and improvements included fixing elevators, installing new lighting, fixing stairway handrails, upgrading heat and hot water systems, installing emergency generators, repairing roofs, renovating bathrooms, and installing new appliances. A new study used New York’s experience with a conversion comparable with the Rental Assistance Demonstration (RAD) to explore how renovations may improve the health and well-being of residents in public housing. The findings have national relevance, as the US Department of Housing and Urban Development estimates that the 1.1 million public housing units across the country require $50 billion in capital repairs, and many housing authorities intend to fund such repairs through RAD.
Using developer information and New York state Medicaid claims data, the authors constructed a treatment group of 1,644 Medicaid beneficiaries who were living in the six renovated housing developments. They compared this group with a matched set of Medicaid enrollees in the same neighborhoods but whose buildings were not renovated. The authors then used a difference-in-differences regression model focused on five health conditions that they identified as housing sensitive based on prior research: asthma, respiratory infections, hypertension, mental health, and injuries. In addition, they designed two composite indices of the five housing-sensitive conditions for comparison.
There are a few limitations to the authors’ analysis. First, although the authors matched the groups using individual, building, and neighborhood variables, there could still be unobserved differences between residents of the renovated housing and those in the comparison group. Second, Medicaid claims data may over- or underestimate true rates of health conditions. More generally, health outcomes data do not take into account more holistic meanings of well-being; self-reported measures of health and quality of life are important outcomes as well. Finally, the authors were only able to track health outcomes for three years after the renovation; other health impacts may take longer to appear.
Comments (0)