The durability of supported employment effects
The original New Hampshire Supported Employment Study was a two-site, controlled, clinical trial with random assignment to Group Skills Training (GST) or Individual Placement and Support (IPS) (within site) and 18-month follow-up. Both the GST and IPS programs were implemented in two New Hampshire cities and surrounding regions with populations of 166,000 and 119,000. Implementation data supported the fidelity of both interventions, and clients received approximately the same number of direct contact hours and amount of service costs in the two interventions. Following 18 months in the experimental phase, clients were allowed to leave their assigned vocational condition and were asked to participate in a 2-year extension phase. Guided by advice from providers, they pursued additional vocational services at their own discretion. Those who gave written informed consent were reassessed after 1 and 2 years (30 months and 42 months from original baseline) with a composite interview that was administered by a research interviewer who was independent of the clinical or vocational programs.
The purpose of this study was to examine the persistence of supported employment outcomes and the influence of continuing vocational services following the experimental phase of the New Hampshire Supported Employment Study. In the original study, one form of supported employment, Individual Placement and Support (IPS), was found to be more effective than another form, Group Skills Training (GST), in improving clients' competitive employment. IPS clients worked approximately twice as much and earned twice as many wages
The original New Hampshire Supported Employment Study was a two-site, controlled, clinical trial with random assignment to GST or IPS (within site) and 18-month follow-up. Both the GST and IPS programs were implemented in two New Hampshire cities and surrounding regions with populations of 166,000 and 119,000.
The original study included 143 unemployed adults with severe mental illness from two community mental health centers in New Hampshire. Of the 140 participants who completed 18 months in the original study (experimental phase), 126 (90.0%) consented to participate in the 2-year extension phase. At the start of the original study, the 126 extension phase participants had an average age of 36.8 years (SD = 9.5); 50% were female; 52.4% were never married, and 8.7% were currently married; 27.8% had not completed high school or received a GED; and 96% were Caucasian. Their primary psychiatric diagnoses were heterogeneous: schizophrenia and related psychotic disorders, 46.8%; bipolar and other severe mood disorders, 44.4%; and other disorders (primarily severe personality disorders), 8.7%. During the 18-month experimental phase, 61.1% of the extension phase clients (77/126) worked in at least one competitive job, and they worked an average of 430.8 (SD = 716.5) hours.
Competitive employment was as work in the competitive job market at prevailing wages that was supervised by personnel employed by the business. Employment was assessed regularly by employment specialists in GST and IPS during the 18 months of the experimental phase and by direct interviews with clients at the 1-year and 2-year points of the extension phase (30-month and 42-month interviews) using the Employment and Income Review. To assess vocational services during the extension phase, clients reported on the vocational services received in the previous 2 months, including type of service, the provider, number of days, and the average amount of time each day. Consequently, estimates of vocational service utilization during the extension phase are based on 4 of the 24 months.
Other variables reported in this paper derived from the structured interview that was conducted with clients at regular intervals throughout the study period. This interview, which is described more fully elsewhere contained measures from the following domains: demographics, psychiatric symptoms, income and benefits, quality of life, drug and alcohol use, self-esteem, recent work and school history, and residential history. This hour-long interview was conducted by research staff members, who had been trained in standardized research interviewing and who were supervised throughout the study.
GST was a vocational rehabilitation program located in a private agency outside the mental health centers that offered individualized intake, 8 weeks of pre-employment skills training in a group format, individualized job placement, liaison with mental health providers, and follow-along supports. During the pre-employment training, clients were encouraged to explore values, preferences, strengths, and weaknesses, as well as to discuss and practice skills in choosing, getting, and keeping a job.
IPS offered an integrated and direct approach in which employment specialists joined the case management teams in the mental health centers and immediately helped clients to begin searching for a job on an individualized basis. The IPS employment
specialists assumed that clients would learn about the job world, and about their skills and preferences, on the job rather than through pre-employment training.
The comparison conditions were Individual Placement and Support; and the Group Skills Training and other Vocational Rehabilitation Services model; and No Services.
This 2-year extension of an experimental study of supported employment showed persistence of the experimental effects on competitive employment. Overall, clients experienced no significant deterioration in amount of competitive employment, despite the fact that only 60% continued to receive vocational services. This finding is in contrast to that found in earlier studies of supported employment although it is consistent with the persistence found by Bond and colleagues.
Moreover, differences between the original experimental groups that emerged during the 18-month experimental phase persisted throughout the 2-year extension phase with only moderate attenuation. The evidence indicates a continuation of the momentum gained during the experimental phase of the study despite minimal continued vocational supports overall. This finding was also contrary to expectations, as earlier studies have often found a decrease in group differences following the termination of a formal intervention. The results also showed that receiving services during the extension phase was related to amount of employment for clients in the original IPS group. More than half of the IPS clients received IPS-like services during the extension phase, and together with those few who received other services, they were more than twice as likely to work than clients who received no services.
It appears that continued vocational services, even if minimal, are critical to the durability of the elevated vocational outcomes from an IPS program. As IPS becomes implemented more widely, it will be important to design and to build in mechanisms that provide continued services in order to maintain the higher rates of competitive employment that IPS provides.
