“Housing First” Reduces Drinking, Health Costs for Homeless Alcoholics

Eliminating requirements for homeless individuals with severe alcohol problems to quit drinking before they are eligible for supportive housing is associated with reduced alcohol use, decreased utilization of healthcare and social and legal services, and results in significant cost savings, new research suggests. By removing requirements for sobriety, mandatory attendance at alcohol-treatment programs, and other typical barriers to supportive housing — an approach known as housing first — investigators at the University of Washington, in Seattle, found individuals who were housed accrued a 53% reduction in costs compared with wait-listed control subjects.

Further, in the year prior to the study, 95 housed study participants accrued median costs of $4066 per month per individual, with a total of cost of almost $8.2 million. Six months after receiving housing, individual median costs per month declined to $1492 and again at 12 months, to $958. Total costs for the housed group for the year after study enrollment were just under $4.1 million, for a reduction in total costs of more than $4 million.

The researchers also found that individuals who were housed significantly reduced their alcohol use over time, with the study results indicating that the total amount of daily alcohol consumption was reduced by approximately one-third over a 1-year period. In addition, residents also drank to the point of intoxication on significantly fewer occasions after they were housed vs. when they were homeless.

"The take-home message is that the housing-first program for current public inebriates costs taxpayers less money than leaving these individuals on the streets — the savings are substantial and continue for at least 12 months," principal investigator Mary E. Larimer, PhD, told reporters attending a press conference where the study results were presented.

"In addition, reduced use of healthcare services and reduced burden on criminal justice resources may also increase efficiency of healthcare and improve public safety. Further, individuals in the housing-first program drink less over time when they have housing, even though they are not required to stop drinking or receive treatment as part of this program. These findings really support housing first as an important component of a coordinated response to the challenge of serving this difficult population," Dr. Larimer added.

The study is published April 1 in the Journal of the American Medical Association.

Controversial Initiative

To date, housing-first approaches have primarily targeted homeless people with serious mental illnesses and co-occurring substance-use disorders. The housing-first program in Seattle — known as 1811 Eastlake — is a permanent housing model that targets homeless adults with severe alcohol problems who use local crisis services at the highest levels.

According to study investigators, the housing-first project was extremely controversial. Originally approved in November 1999, the project ultimately was not completed until December 2005, said study author William G. Hobson, executive director of the Downtown Emergency Service Center (DESC) in Seattle.

"It took us 6 years to get this under development. A typical housing DESC housing project takes 28 to 30 months. The project was sued and held up in court, because it was perceived by the public that we were simply opening a party house where individuals addicted to alcohol were going to be allowed to drink, run amok, and generally set their hair on fire. I think this research shows that this is not the case and that this project [and others like it] can be operated in a responsible way," said Mr. Hobson.

Although residents are not required to stop drinking or participate in treatment for alcohol addiction, there are on-site case managers available to engage residents in discussions about substance abuse and life goals. Meals and on-site healthcare services are also offered. Per-person costs for housing and services average $1120 per month.

Important Public-Health Policy Implications

The study included 95 study participants who incurred the highest total costs in 2004 for alcohol-related hospital emergency services, the sobering center, and the local county jail.

The investigators' primary objective was to examine the impact of housing-first on use and cost of healthcare, criminal justice, and social services over a 6-month period for residents of the 1811 East Lake project compared with 39 wait-listed participants.

Individuals in both groups were chronically homeless and had experienced their first episode of homelessness an average of 18 years prior to study enrollment. In addition, they spent the majority of nights on the street, in the sobering center, or in shelters in the 3 years prior to study enrollment. All had extensive histories of alcohol abuse, with an average of 16 addiction-treatment attempts before enrollment.

In addition, the majority had chronic health conditions and comorbid mental-health conditions.

According to the researchers, the study found that permanent, rather than temporary, housing may be necessary to fully realize these costs savings, because, as Dr. Larimer pointed out, the cost offsets continued to grow over time.

"Other studies of housing-first or similar housing programs have shown a reduction in use of other services when people are housed. However, most of them have not been able to show that the cost offset is larger than the cost of administering the housing, staffing, and support services, etc. In our study, we actually showed a substantial cost offset above the administrative costs of providing housing," Dr. Larimer told Medscape Psychiatry in a follow-up interview.

Furthermore, she added, the study also showed that those who remained in housing longer had larger cost reductions over time.

-Adapted from: “Supportive Housing Without Conditions Reduces Drinking, Health Costs in Homeless Persons with Severe Alcoholism”, by Caroline Cassels for Medscape Medical News, http://www.medscape.com/viewarticle/590490 retrieved 4/3/09. Original research appeared in JAMA. 2009;301:1349-1357.

 

4/09