Targeting Services to Individuals Most Likely to Enter Shelter: Evaluating the Efficiency of Homelessness Prevention

Successful strategies for homelessness prevention must efficiently target people at greatest risk for experiencing homelessness. We developed a model to target homelessness prevention services to individuals and evaluate its results compared to a similar model we developed for families. We tracked 10,220 individuals who applied for community-based homelessness prevention in New York City from 2004 to 2010 using Cox regression to predict shelter entry over the next 2–8 years. Both the comprehensive model and a brief screening model based on seven variables are at least as efficient as worker judgments, increasing correct predictions by 77 percent and reducing unidentified cases of subsequent homelessness by 85 percent. Risk factors for homeless individuals were mostly a subset of risk factors for families. The evidence suggests that, despite limitations, empirical models increase the efficiency of prevention services for individuals. Investigators in other cities may improve the efficiency of local prevention programs with empirical targeting models.