This 2019 report summarizes analyses from a multi-site study on aging homelessness in Los Angeles County, Boston, and New York City. Researchers in each jurisdiction linked administrative data across multiple domains to better understand the complex needs of this population, and estimate future costs across service systems over time.
Researchers at each site matched historic shelter data with corresponding data on health services use to identify four distinct subgroups within the study population. It must be noted that the NYC and LA County sites were able to match with the all-payer data in their respective states, whereas in Boston health records were limited to Medicaid fee-for-service and Medicaid managed care encounter claims. After determining these subgroups, they then recommend potential housing interventions based on the specific needs of each segment and also estimate the average annual per person cost of each intervention.
Finally, to illustrate the national impact of this growing population, the team used the local estimates to project costs and cost offsets nationally. Together, projected costs and offsets show that if proposed housing models for the four subgroups are brought to scale in line with the report’s recommended approaches, excessive costs associated with aging of the homeless population should significantly decline. The authors note that, because the Boston dataset does not include Medicare expenditures and potential offsets, the cost reductions are likely underestimated for this jurisdiction.
These findings, made possible only through the use of linked administrative data, are urgent. The authors conclude by compelling government agencies to take action and work across sectors to actualize solutions because, “Absent new housing solutions, substantial public resources will otherwise be spent unnecessarily on excess shelter, health, and nursing home use.” The data show that the need for action is critical.
Access the combined report here.