People with mental illness who lived in supportive housing and received a spectrum of psychiatric treatment services cost taxpayers $41,494, only $994 more per year.

"Supportive housing is a sound investment of public resources," said Culhane, an associate professor of social welfare policy at the University of Pennsylvania’s Center for Mental Health Policy and Services Research.

"We certainly hope these findings create momentum to create more housing for homeless people with mental illness—but one would hope that society doesn’t need a cost argument to accomplish this," he said.

The researchers tracked service use by 4,679 mentally ill homeless people in New York City for two years before they were placed in housing funded by the 1990 New York/New York (NY/NY) Agreement to House Homeless Mentally Ill Individuals. They then tracked them for two years after placement.

The NY/NY initiative came into being when New York state and New York City joined to fund and develop 3,600 community-based permanent housing units in New York City for homeless people with chronic mental illness.

The agreement provided for supportive housing with linkage to a variety of psychosocial services such as vocational training, group and individual therapy, and case management.

Besides tracking the sample of almost 5,000 homeless people who entered NY/NY housing, the researchers matched the group with an equal number of control subjects with mental illness who remained homeless.

They then sought to determine the extent of this population’s shelter use, inpatient hospitalization, and time spent in jails and prisons, for example. The total sample included almost 10,000 individuals.

To determine service use for the large sample, researchers gained access to data from seven databases on psychiatric, public health, and criminal justice systems in New York. These included the New York City Human Resources Administration, the New York City Department of Homeless Services, and the New York City Health and Hospitals Corporation.

Study findings revealed that after placement in NY/NY housing, there was an 86 percent drop in the number of shelter days per person, a 60 percent drop in state hospital use, and an 80 percent drop in the number of inpatient days spent in a public hospital. Researchers also found that housing the homeless mentally ill cut incarceration rates in half.

All in all, supportive housing resulted in a per-person cost reduction of $16,282 for the more expensive services used by homeless people with mental illness.

"This study is really important because it documents with good methodology what advocates have known all along—that it makes more sense to house homeless people with mental illness than to let them remain on the streets," remarked Hunter McQuistion, M.D., chair of APA’s Committee on Poverty, Homelessness, and Psychiatric Disorders.

Researchers did find one increase in costs—people living in the NY/NY housing increased their use of less-costly outpatient services by 95 percent.

McQuistion, who is also medical director of Project Renewal, a nonprofit agency serving homeless people in New York City, attributed this increase in services to the restabilization period of the homeless person. "This is related to the chronic medical and psychiatric conditions they have not been able to address while living in the streets—and these costs drop eventually," he said.

According to Culhane, the total cost estimates from the study are conservative; that is, the researchers didn’t measure some costs associated with homelessness. "We only tracked costs that could be associated with one particular individual," said Culhane, who added that costs accrued through outreach programs and costs to the police and court system were not accounted for.

"Had we tracked all those costs and estimated the benefits, we would have found that the housing actually saved money," Culhane pointed out.

McQuistion took a similar view, asserting that if the researchers tracked the mentally ill people placed in housing for an even longer period, the cost of housing homeless people with mental illness would fall even more.

Indeed, supportive housing holds an intrinsic value that extends beyond cost savings.

According to Culhane, the supportive housing movement came about because of patient preferences. "People with serious mental illness preferred to have ‘normalized’ housing in the community where they have a right to tenancy, privacy, and permanency—where they have their own key."

Culhane acknowledged that more research is needed to determine whether the results of the New York City study could be generalized to other cities with large homeless populations. "The point is that homelessness is not cost free and that housing reduces people’s use of expensive services. I think those general principles would likely stand up in many other places," he said.

A third of the people placed in housing ended up back on the streets, however, and Culhane said he would like to see further research on different types of housing and service-utilization patterns of homeless people with mental illness.

McQuistion agreed. "When you find the best fit between the clinical and social characteristics of a person and the type of housing—whether it be less structured or more, the person’s tenure in that housing will increase," he said. ▪