Homelessness is a pervasive public health issue, and one that has had a significant impact on Los Angeles County: The county has the highest rate in the United States of unsheltered individuals who experience homelessness, and this rate has continued to increase during recent years. Individuals experiencing homelessness are more likely to have serious and chronic health conditions than the general population. Studies also have shown that individuals living on the street utilize health and other social services at a significantly higher rate than similar individuals who are not experiencing homelessness.
Permanent supportive housing (PSH), a program model that provides long-term housing coupled with case management services, is a promising approach for both improving housing stability for ill people experiencing chronic homelessness and reducing their use of costly emergency and inpatient health care. In 2012, Los Angeles County’s Department of Health Services (DHS) launched Housing for Health (HFH), an initiative designed to reduce homelessness, reduce the inappropriate use of emergency room (ER) and inpatient health care, and improve the health of the population experiencing homelessness. The HFH program includes both interim housing (e.g., recuperative or transitional housing) and PSH. Since its inception, the program has created over 3,400 housing placements.
In 2014, Los Angeles County commissioned the RAND Corporation to evaluate the PSH component of the HFH initiative. RAND conducted a formative evaluation of program implementation and then evaluated outcomes, specifically health and social service use and associated costs among program participants. RAND examined data on service use and costs from several county departments: DHS; the Department of Mental Health (DMH); substance use treatment services provided by the Department of Public Health (DPH); General Relief assistance provided by the Department of Public Social Services (DPSS);1 shelter services provided by the Los Angeles Homeless Services Authority (LAHSA); and law enforcement services provided by the Sheriff’s and Probation Departments. The outcomes evaluation addressed the following questions:
The formative evaluation found that the different entities involved with administering the HFH PSH program had a consistent and shared understanding of the program’s purpose, boding well for future program implementation. However, representatives expressed some logistical concerns, such as communication barriers between the multiple entities and uncertainty regarding program expansion and sustainability due to limited operating capacity. Key program staff recognized these concerns and were working to address them at the time of our study.
The outcome evaluation included data from 890 individuals placed in PSH during the first
2.5 years of the program. RAND used a pre-post study design that compared HFH PSH recipients’ service use during the year prior to receiving housing compared to the year following receiving housing. A large proportion (83 percent) were experiencing chronic homelessness (i.e., met the U.S. Department of Housing and Urban Development definition of continuous homelessness for a year or more or four episodes of homelessness equal to a year in the last three years) and had co-occurring medical and mental health or substance use conditions (88 percent). Among the analytic sample, the time from initial application to receipt of PSH was, on average,
6.9 months; the time from initial application to receipt of case management was, on average, 4.3 months. More than 96 percent of HFH PSH recipients were stably housed for at least one year.
The outcome evaluation found that clients’ use of medical and mental health services dropped substantially, including ER visits and inpatient care. Costs, correspondingly, also decreased. After moving into PSH, participants made an average of 1.64 fewer ER visits in the ensuing year; inpatient hospital stays decreased by more than four days. In addition, outpatient visits were reduced by an average of four visits. PSH recipients’ time receiving General Relief declined by an average of 1.38 months. Although the number of individuals arrested and the number of jailed arrests decreased during the year after receiving housing, the number of jail days increased following PSH entry by an average of 2.76 days. The number of HFH PSH recipients using the other services for which RAND had data (i.e., emergency shelters, substance use treatment, and probation services) was quite small both before and after housing receipt.
Across all the services RAND examined, the associated costs for public services consumed in the year after receipt of PSH declined by close to 60 percent. The average public service utilization cost per participant for the year prior to housing totaled $38,146; in the year after receiving housing, it totaled $15,358. When taking into account PSH costs, RAND observed a
20-percent net cost savings, suggesting a potential cost benefit of the program.
The health functioning survey found that participants’ mental health functioning improved after receiving housing, though physical health was largely unchanged. At housing entry, participants reported significantly lower physical and mental health functioning than the general population norms, based on national data. One year after being housed, participants reported a
significant improvement in mental health functioning, though scores were still lower than general population–normed values. Physical health functioning scores remained lower than population-normed values and were no different from scores reported at housing entry. Physical health functioning values were consistent with those reported by older individuals experiencing chronic conditions, similar to the population enrolled in the HFH PSH program. These findings suggest that the program serves a population with chronic physical and mental health needs who are likely to benefit from long-term supportive housing.
These findings suggest that HFH PSH could save money for Los Angeles County. However, the cost results have an important limitation: they measured only services associated with six county departments over a two-year period and are not a full accounting of all potential costs and benefits from the HFH PSH program. Research that employs more rigorous causal methods (i.e., that includes a comparison group) is needed before we can state conclusively that the dramatic changes observed in county service utilization prior to and following supportive housing are solely attributable the HFH PSH program.
In summary, our findings suggest that DHS succeeded in implementing the HFH PSH program. Hundreds of individuals who formerly experienced homelessness, many with complex chronic physical and mental health conditions, have been stably housed at least for one year. Los Angeles County data demonstrate a dramatic reduction in service use across the medical and mental health departments. Overall, the cost reductions more than covered the year’s worth of supportive housing costs, as we observed a net cost savings of 20 percent.
As this program is considered for future implementation and sustainability, the ability to scale up with the appropriate level of oversight and collaboration among the different partnering entities will need to be monitored for success. Thus far, the program has successfully enrolled large numbers of individuals and has kept almost all of them in housing for a year while reducing their utilization of costly medical and mental health care.