Bad housing conditions such as lead exposure, poor insulation, mold, and pests have been linked with elevated risks of physical and mental illnesses. City governments set maintenance standards for existing rental housing and enforce these standards through inspections and penalties - largely via reactive approaches – by responding to complaints. In a new paper, authors at the P4A Research Hub at New York University Robert F. Wagner Graduate School of Public Service developed an algorithm using health insurance claims to identify buildings of poor quality, allowing city governments to take a proactive approach in identifying low quality residential buildings.                  


The authors used random forest and logistic regression and identified twenty-three specific housing-sensitive health conditions that are correlated with poor-quality housing. The conditions are grouped into five broad categories, asthma, cardiovascular illness, mental illness, substance use and injuries. The authors then used these housing-sensitive health conditions and created a housing health index, which identifies substandard buildings. 

Implications for policy and practice

The housing health index has the potential to help local governments identify poor quality buildings, taking more of a proactive approach to housing code enforcement rather than solely relying on complaints from tenants.  Local governments can use the housing health index to prioritize inspections, which could help address biases from the complaints.  Additionally, community-based advocacy organizations can use housing health index to provide targeted supportive services for affected residents and connect them with health care providers.