Three of the sites provided services exclusively to persons with mental ill-. Vocational Outcomes Among Formerly. Homeless Persons With Severe Mental.
Vocational Outcomes Among Formerly Homeless Persons With Severe Mental Illness in the ACCESS Program Judith A. Cook, Ph.D. Susan A. Pickett-Schenk, Ph.D. Dennis Grey, B.A. Michael Banghart, B.S. Robert A. Rosenheck, M.D. Frances Randolph, Dr.P.H.
Objective: This study examined the vocational outcomes of 4,778 formerly homeless individuals with severe mental illness who were enrolled in the Access to Community Care and Effective Services and Support (ACCESS) program, a multisite demonstration project designed to provide services to this population. Methods: Participants were interviewed at the time of enrollment and again three months and 12 months later by trained researchers who were not part of the treatment team to determine their employment status. At 12 months, participants were also asked about the types of services they had received during the past 60 days. Multiple logistic regression analysis was used to predict employment at 12 months. Results: ACCESS participants reported receiving relatively few job-related services. Nonetheless, modest but significant increases occurred between baseline and three months and between three months and 12 months in the total proportion of participants who were employed and who were employed full-time and in hourly earnings and estimated monthly earnings. The number of hours worked per week increased significantly between three months and 12 months. When the analysis controlled for site, study condition (whether the ACCESS site received or did not receive extra funds to improve service integration), minority status, addiction treatment, and mental health treatment, participants who were employed at 12 months were more likely to have received job training and job placement services. Conclusions: Programs that work with homeless mentally ill persons may better serve their clients by placing as great an emphasis on providing employment services as on providing housing and clinical treatment. (Psychiatric Services 52: 1075–1080, 2001)
Dr. Cook is director and professor, Dr. Pickett-Schenk is assistant professor, Mr. Grey is project coordinator, and Mr. Banghart is a research data analyst in the mental health services research program of the department of psychiatry at the University of Illinois, 104 South Michigan Avenue, Suite 900, Chicago, Illinois 60603 (e-mail, cook@ ripco.com). Dr. Rosenheck is director of the Veterans Affairs Northeast Program Evaluation Center in West Haven, Connecticut, and clinical professor of psychiatry and public health at Yale School of Medicine. Dr. Randolph is acting chief of the homeless programs branch of the Center for Mental Health Services in Rockville, Maryland.
PSYCHIATRIC SERVICES
♦ August 2001 Vol. 52 No. 8
T
he problems that link homelessness and unemployment in America are exacerbated by the fact that many homeless persons who have a severe mental illness do not receive clinical or vocational services (1). Furthermore, national health and disability policies often hamper rather than promote employment among this population (2). Despite the existence of employment programs for homeless persons with mental illness, such as Comprehensive Opportunities to Assist Consumers Who Are Homeless (COACH), Project HOME, and InCube (3), few studies have focused on vocational outcomes for this population. Existing research consists primarily of studies of vocational rehabilitation among generic populations of homeless persons that may have included subsamples of individuals with severe mental illness. The Job Training for the Homeless Demonstration Program, a federal program that operated between 1995 and 1998, provided assessment and employability planning; vocational training; job development; and placement, postplacement, and follow-up services to homeless persons (4). In addition, each of the program’s 63 sites provided outreach services, case management, housing assistance, substance abuse counseling, mental health treatment, child care services, transportation, and life skills training. Three of the sites provided services exclusively to persons with mental ill1075
ness, who constituted 3 percent (N=1,232) of the 45,192 program participants. Thirty-three percent of the participants who had a mental illness were placed in jobs, compared with 35 percent of all participants; 60 percent of those with mental illness were employed 13 weeks later, compared with 50 percent of all participants. These results suggest that the employment potential for persons with mental illness is significant. In an initiative called Next Step: Jobs, which was conducted from 1995 to 1998 by the Corporation for Supported Housing, 20 nonprofit organizations in San Francisco, New York City, and Chicago provided job development, training, placement, and follow-up services to 3,200 supportedhousing residents who previously had been homeless (5–7). More than a third of the program participants were found to need mental health services, and five of the programs in the initiative served this population exclusively. Results released to date indicate that, over time, equivalent proportions of participants who did and did not have a mental illness became employed. However, participants who had a mental illness were more likely to be working in part-time positions and held fewer jobs overall; furthermore, these participants were more likely to hold jobs within their service delivery program (6). Findings from studies of both of these programs suggest that job training and placement services may be most effective when they are combined with the additional services that homeless persons need to overcome obstacles to employment. These obstacles include poverty, poor physical health, inadequate housing, low education levels, trauma from victimization, and substance abuse. Service needs are even greater for homeless persons who have a severe mental illness, for they must also cope with debilitating psychiatric symptoms and social stigma (8). The study reported here assessed the vocational outcomes of participants in the Access to Community Care and Effective Services and Supports (ACCESS) program, a federal service demonstration project that operated between 1994 and 1998 and 1076
provided coordinated mental health, housing, substance abuse, and social services to homeless persons with severe mental illness. The ACCESS program examined the effect of service system integration on removing homeless persons with untreated mental illness from homelessness and improving their health status, service use, and quality of life (9,10). Each of the program’s 18 sites provided intensive outreach and case management services to 100 clients per year for each of its four years of operation. The sites were located in urban areas in nine states: Connecticut, Illinois, Kansas, Missouri, North Carolina, Pennsylvania, Texas, Virginia, and Washington. At each site, one service delivery agency received extra funds to improve service integration (enhancement condition) and one did not (comparison condition). Outreach services included making initial contact with homeless mentally ill persons in the community—for example, in homeless shelters, on the streets, at soup kitchens, and in dropin centers—followed by referrals to case management. After an individual was enrolled in the program, case management services were delivered in community settings, with the majority of sites (N=14) using an intensive case management model (11,12). The other four sites used a continuous treatment team (13) or a strengths model (14) to deliver case management services. All of these models provided an array of services that were designed to stabilize symptoms, prevent relapse, and improve functioning in the community. Rosenheck and Lam (10) analyzed the service needs reported by recipients of ACCESS outreach services. They found that 56 percent of clients identified job assistance—that is, help with job training or finding a job—as an important service need. The ACCESS service mix did not mandate the inclusion of vocational services; however, these services were available at all sites either directly or through referral. Moreover, improved vocational outcomes may have resulted from helping clients meet other needs that were indirectly related to employment.
The ACCESS project provided an opportunity to examine work outcomes and the predictors of those outcomes for a large group of recently homeless individuals during their first year of integrated treatment. Our study addressed three major questions. First, were the participants’ vocational outcomes—that is, employment status, full-time employment status, hourly salary, hours worked per week, and estimated monthly earnings—different at baseline, at three months, and at 12 months? Second, what predicted the likelihood that participants would be employed 12 months after they began receiving ACCESS services? And third, how did receipt of vocational and other services—for mental health and substance abuse, for example—affect the likelihood of their being employed at 12 months?
Methods Procedure
The study evaluated data for 4,778 individuals who were enrolled in the ACCESS program at some time between 1994 and 1998. Table 1 summarizes the characteristics of the study participants at baseline. Individuals were eligible to enroll in the ACCESS program if they were homeless, had a severe mental illness, and were not receiving ongoing treatment for their illness. To meet eligibility criteria for homelessness, they had to have spent at least seven of the past 14 nights in a shelter, in a public or abandoned building, or outdoors. A 30-item screening instrument (15) was administered by outreach workers to assess mental illness eligibility. Individuals who met the eligibility criteria were then invited to take part in the evaluation component. Those who consented to participate were interviewed at the time of enrollment (baseline) and again at three months and 12 months after enrollment by trained research interviewers who were not part of the treatment team (8). The face-to-face interviews lasted 60 to 90 minutes each, and participants received a $15 honorarium for each interview. The interview protocol assessed history of homelessness, current housing status, psychiatric history, drug and alcohol use, physical
PSYCHIATRIC SERVICES
♦ August 2001 Vol. 52 No. 8
Table 1
Table 2
Characteristics at baseline of 4,778 individuals who participated in the ACCESS program
Employment outcomes of 4,778 individuals who participated in the ACCESS program
Characteristic
N or mean
Male 2,901 Race African American 2,415 Caucasian 1,950 Hispanic 210 Native American 83 Asian or Pacific Islander 30 Mixed or other 84 Education1 Less than high school 1,778 High school or equivalent 1,731 Some college or college degree 1,248 Marital status1 Married or cohabitating 249 Not married 4,521 Mean±SD age (years)2 38±9.5 Mean±SD lifetime duration of homelessness (years)3 7±8.4 Diagnosis4 Major depression 1,318 Alcohol dependency 1,236 Schizophrenia 1,190 Drug dependency 1,115 Personality disorder 740 Other psychotic disorder 704 Bipolar disorder 570 Posttraumatic stress disorder 368 Anxiety disorder 337 Adjustment reaction 233 Other 464 Prior hospitalization For mental health problems 3,394 For alcohol or drug dependency 2,105 1 2 3 4
% 61 51 41 4 2 1 2 37 36 27
Outcome
N or mean
%
All study participants Any work in past 30 days 864 18 Employed participants Any work in past 30 days 206 24 Mean number of hours per week (median) 20 (18) Mean hourly wage (median) $5.56 ($5.00) Mean monthly earnings (median) $259 ($151) 1 2
At three months1
At 12 months2
N or mean
N or mean
%
%
931
19 ∗
1,053 22∗
299
32 ∗
383 36∗
24 (21)
26 (25) ∗
$5.76 ($5.00) ∗
$5.94 ($5.50) ∗
$381 ($240) ∗
$469 ($300) ∗
Asterisks indicate a significant difference (p
