Placing homeless people with severe alcohol-use problems into free housing without requiring abstinence or treatment reduced costs, medical use, and contacts with the criminal justice system, according to University of Washington researchers.

Such Housing First programs "can reduce the public burden associated with overuse of crisis services and reduce alcohol consumption," wrote Mary Latimer, Ph.D., and colleagues in the April 1 Journal of the American Medical Association. Other research has shown Housing First programs to be successful in getting homeless mentally ill people off the streets while reducing some costs.

Chronic public inebriates—sometimes referred to as "frequent flyers" by police and emergency medical system crews—disproportionately use publicly funded health- and justice-system resources. Conventional interventions, like shelters, treatment programs, and abstinence-based housing, have poor records in breaking those patterns.

The Seattle researchers developed a list of 388 individuals who incurred the highest costs in 2004 for use of alcohol-related hospital emergency services, the local sobering center, and the King County jail. A total of 134 participants were eventually recruited for the study from November 2005 to March 2007. There were 119 people in the interventional housed group and 39 in a wait-list control group. Their average age was 48, and they had first become homeless at an average age of 31. Nearly all (94 percent) were male, and 30 percent had attended college at some time. Many had serious medical problems such as hepatitis (40 percent) and tuberculosis (18 percent). They each had been treated an average of 16 times for alcohol abuse.

The trial was not randomized for ethical reasons. Also, because participants were approached in rank order of services used and consequent costs in the year prior to the study, the treatment group had incurred larger expenses than did the control group.

The treatment group was housed at a Housing First facility where participants were offered meals and on-site health services. They had no treatment requirements, but case managers did engage residents about substance-use issues. Costs for housing and services averaged $1,120 a month.

The researchers collected administrative data at six months and one year from area medical, social services, and correctional departments, as well as Medicaid. FIG1

The individuals who received housing had incurred more than $8 million in costs in the year preceding the trial, but only about $4 million in the first year of intervention. Monthly charges dropped for both housed and wait-list cohorts, but the average had dropped for the housed group at six months and still further at 12 months (see table). Treatment differences were produced mainly by fewer nights in shelters and reduced use of the sobering center.

All measures of service use (except days in detoxification) dropped more with longer time in housing. Cost reductions occurred in charges to Medicaid, Harborview Medical Center, and emergency medical systems.

At one year, average daily alcohol use dropped from an average of 15.7 drinks a day to 10.6 a day.

The study demonstrates the value of stable housing, which is seen as a greater need than alcohol treatment by homeless persons with alcohol problems, said the authors. The lower burden on shelters, hospitals, and jails could mean greater access to care for other people who need it and more attention by police and emergency services to more significant matters of public safety, they suggested.

The study was funded by grants from the Substance Abuse Policy Research Program of the Robert Wood Johnson Foundation, National Institute on Alcohol Abuse and Alcoholism, and National Institutes of Health and received support from Seattle's Downtown Emergency Service Center, which runs the housing program site.

An abstract of "Health Care and Public Service Use and Costs Before and After Provision of Housing for Chronically Homeless Persons With Severe Alcohol Problems" is posted at<http://jama.ama-assn.org/cgi/content/abstract/301/13/1349>.▪