Effects of supported employment on the competitive integrated employment outcomes of transition age and young adults with intellectual disabilities: A non-experimental causal comparative study

1.1Supported employment: A means to CIE

The purpose of supported employment is to support individuals with the most significant disabilities in achieving CIE outcomes ( Wehman, 2012). To ensure efficacy, supported employment requires the professionals providing services to hold a distinctive set of skills. Employment specialists must know how to help identify meaningful consumer choice, arrange for funding, identify employment opportunities in the community, engage employers, work with parents and families, help with Supplemental Security Income (SSI) eligibility determination issues, arrange transportation, and most importantly, effectively train clients to achieve their work goals (Wehman, 2012). Supported employment services are intensive interventions offered to individuals with the most significant disabilities in a competitive work environment. This includes individuals who have never experienced work in a competitive employment environment and individuals who cannot function effectively at a job due to a disability ( Sung et al., 2022). Existing research has demonstrated a positive relationship between supported employment and health for individuals with intellectual disabilities (Iwanaga et al., 2021).

A key aspect of supported employment is rapid placement into CIE by using a job coach trained in how to teach vocational skills to the client while they are on the job and employed ( Wehman, 1981). This model has been used all over the world since the early 1980s with persons who experience intellectual disabilities. Central to the supported employment approach is the process of first securing CIE for an individual and then providing needed support, rather than waiting for “work readiness” to initiate job placement. In other words, supported employment adopts a “place then train” approach. Eliminating unnecessary preparatory training enables individuals to learn job skills in real work settings as soon as possible. With SE, there is a presumption of employment, which refers to the right of all individuals to work, regardless of disability.

1.2Purpose of the present study

The present study focuses exclusively on individuals with intellectual disabilities using the Rehabilitation Services Administration Case Report (RSA-911) database. This study builds upon our earlier research in 2014 ( Wehman et al., 2014). The 2014 study examined the effect of supported employment on the CIE outcomes of transition-age youth and young adults with intellectual and developmental disabilities served by state VR agencies. The 2014 study included 23,298 participants and the classification and regression tree (CART) method was used to estimate propensity scores and to adjust for selection bias on all prominent demographic covariates. Results yielded six homogeneous subgroups, and receiving supported employment was found to increase the rates of CIE across all groups. The effect of supported employment was especially strong for those who were SSI recipients, received special education, and individuals with intellectual disabilities or autism spectrum disorder who graduated from high school. The findings of that study suggest that supported employment significantly enhances CIE outcomes.

In the present study we examined whether supported employment is an effective intervention for transition-age youth and young adults with intellectual disabilities who were served by state VR agencies. Specifically, the present study was designed to answer two key research questions:

2.1Study cohort

Data for this study were extracted from the U.S. Department of Education, Rehabilitation Services Administration (RSA) Case Service Report (RSA-911) data base, which contains detailed information about demographics, primary causes of disability (e.g., intellectual disability), types of VR services, and employment outcomes and quality of employment for all clients receiving state VR services in the United States (RSA, 2017). In the present study, RSA-911 data for program year (PY) 2018 were used for data analysis. The inclusion criteria for the present study were: (a) ages between 16 and 36 (transition-age youth and young adults) at intake, (b) a primary diagnosis of intellectual disabilities at intake, (c) received special education in high school, and (d) case closed as successfully employed or not employed after receiving VR services. This study was composed of 9,291 transition-age youth and young adults who met the inclusion criteria and whose cases were closed in PY 2018. There were 2,191 individuals with intellectual disabilities who received supported employment as a VR service, whereas 7,100 individuals with intellectual disabilities did not receive SE.

2.2Variables

2.3Procedure

The purpose of the present study was to extract data from a large government database (RSA-911 data) to determine whether transition-age youth and young adults with intellectual disabilities who received supported employment as a VR service were more likely to achieve CIE outcomes than those who did not receive SE. Because data furnished to researchers from RSA have been stripped of personal information, our study received the exempt status from the institutional review board of a research-intensive university in the Mid-Atlantic region of the United States.

3.1Descriptive statistics

Seven variables, namely, age at application, gender, race, long term unemployment (i.e., unemployed for 27 or more consecutive weeks), receipt of SSI, receipt of work incentives and benefits counseling, and referral source, were used for the PSM analysis. Prior to PSM, there were 2,191 participants in the treatment group (i.e., received SE) and 7,100 participants in the control group (i.e., did not received SE). Information regarding the demographic characteristics of the treatment and control groups before and after PSM is presented in Table 1.

Before matching, the average age of VR clients in the treatment group was 22.42 years old (SD = 5.01), which was significantly younger than the control group (23.60 years old [SD = 5.29], t = –9.52, p < 0.001). For gender and race, there were no significant differences between the two groups (gender, χ²[1, N = 9291] = 0.547, p = 0.55, n.s.; and race, χ²[1, N = 9291] = 0.063, p = 0.06, n.s.). Individuals in the treatment group were more likely to be unemployed on a long-term basis than individuals in the control group, χ² (1, N = 9291) = 27.68, p < 0.001, and they were more likely to be receiving SSI benefits at the time of enrollment in the VR program, χ² (1, N = 9291) = 40.22, p < 0.001. Participants in the treatment group were more likely to receive work incentives and benefits counseling compared to participants in the control group, χ² (1, N = 9356) = 309.26, p < 0.001. Individuals in the treatment group were also more likely to be referred to the VR program by community-based rehabilitation programs and less likely to be referred by educational institutions, χ²(3, N = 9356) = 119.11, p < 0.001. These results showed that the treatment and control groups differed significantly on five out of the seven prominent demographic covariates.

3.2Propensity score matching analysis

PSM analysis using the seven prominent covariates identified 2,162 transition-age youth and young adults who did not receive supported employment (i.e., the control group) who match the prominent characteristics of the 2,191 individuals who received supported employment (i.e., the treatment group). Chi-square test and t-test results indicated that, after matching, there were no statistically significant differences between the treatment and control groups on most of the prominent covariates and the work incentives and benefits counseling variable, except the referral sources variable. Thus, PSM created a treatment group and a control group with similar demographic characteristics.

3.3Employment outcome and quality of employment analysis

For employment outcome, we used a chi-square test for independence to determine whether there was a statistically significant difference between the treatment group and control group on employment outcome (i.e., employed vs. not employed). For quality of CIE outcomes, we included only individuals who were employed at case closure. We used independent-samples t-tests to determine whether there were statistically significant differences between the treatment group and control group in hourly wages and hours worked per week. These between-group results are presented in Table 2.

A chi-square test of independence revealed significant between-group differences in the rates of CIE outcomes, χ² (1, N = 4353) = 363.92, p < 0.001. Specifically, youth and young adults with intellectual disabilities who received supported employment (i.e., the treatment group) had significantly higher rate of CIE (n = 1570; 71.7%) at the time of case closure than did the control group who did not receive supported employment (n = 931; 43.1%).

For participants with successful employment outcomes, an independent-samples t-test was conducted to compare hourly wages between the treatment and control groups. The treatment group (n = 1754; M = $8.83, SD = $3.62) had a significantly higher average hourly wage than the control group (n = 1643; M = $5.36, S = $4.91), t (3395) = –23.60, p < 0.001.

Another independent-samples t-test was conducted to compare average hours worked per week between the two groups. The treatment group worked more hours per week on average (n = 1642, M = 18.43 hours, SD = 9.71 hours) than did the control group (n = 1186, M = 16.17 hours, SD = 12.25 hours), t (2826) = –5.46, p < 0.001.

4.1Implications

Supported employment is a heavily used modality of VR intervention, especially with individuals with intellectual disabilities as indicated by the recent Rehabilitation Services Administration (RSA) annual expenditures report (U.S. Department of Education, 2021). Return on investment has become extremely important in each state VR agency, where expectations for CIE success are increasingly high. Therefore, it is essential to accumulate research evidence to demonstrate the effectiveness of supported employment on CIE, wages, and hours worked per week.

The findings of this study could be used for design of Medicaid waiver applications in different states tailored for ongoing payment of supported employment services for individuals with intellectual disabilities. With the rapid growth of Employment First states (U.S. Department of Labor, n.d.), the increased number of states where sheltered workshops are being ordered to close by the Department of Justice, and supported employment replacing segregated services (e.g., Oregon, Rhode Island, and North Carolina), supported employment research must be conducted to show it is an evidence-based practice. The importance of the role of VR in this process of supported employment implementation cannot be overstated. Youth and young adults with intellectual disabilities must be deemed eligible for VR services, have their cases opened, and have their IPEs developed before supported employment vendors can be found and supported employment services can be put in place. Therefore, rapid VR engagement in the earliest stages of career planning is critical.

The data reported in this article demonstrate the strong differential effects of supported employment services on employment outcomes for young people with intellectual disabilities, but the strongest-possible effects of supported employment can be realized only if the intervention model is implemented with fidelity. Implementation of supported employment involves four phases: 1) getting to know the job seeker, 2) job development and matching, 3) training and support, and 4) job retention services ( Schall et al., 2015). To briefly summarize, an employment specialist first uses a variety of methods including home visits, a review of relevant records, interviews, and situational assessments to better understand the job seeker’s strengths, preferences, interests, and needs related to employment. This information is integrated and analyzed in a meaningful way to help inform the job development process. During job development (Phase 2), the employment specialist reaches out to community businesses in industries that align with the job seeker’s interests. Once the worker is hired, on- and off-the-job training and support (Phase 3) is provided by the employment specialist to help the person learn job tasks and acclimate to the work environment. Funding for on-the-job services frequently comes from state VR services. The employment specialist fades from on-the-job supports over time as the worker becomes productive and independent. Finally, ongoing services are put in place to promote long-term job retention (Schall et al., 2015). Typically, an employment specialist will continue to visit the worksite and check in with the employee and employer monthly. This allows for any emerging problems to be addressed quickly.

4.2Limitations of this study

There are several limitations of the present study that must be noted. First, the grouping variable of receipt of supported employment services versus non-receipt of VR services was dichotomously coded; therefore, researchers could not consider the quality, fidelity, intensity, or duration of the supported employment services that the treatment group received. This is an especially important consideration because the WIOA does not require state VR agencies and vendors to assess the fidelity of supported employment implementation (Rubin et al., 2016). However, Wehman et al. (2021) asserted that measuring the quality and intensity of interventions provided throughout the supported employment process is imperative. Though it was not possible in this study using the RSA-911 database, additional research is needed to investigate the quality of supported employment services being provided by state VR agencies.

Second, although we employed a case control design, this study is not experimental and causation between the independent (receipt of supported employment services or not) and dependent (employment outcome, wages, and hours worked per week) variables can only be inferred with caution. Although we have included all the relevant matching variables (i.e., age at application, gender, race, long-term unemployment, SSI, work incentives and benefits counseling, and referral sources), this study was limited by the fact that only demographic and VR service variables were available in the RSA-911 database for PY 2018. There may be other demographic variables that could influence CIE, wages, and hours worked per week.

Third, successful VR case closure is solely defined as whether the individual is employed 90 days from initial employment. The RSA-911 database does not provide longer-term tracking to monitor job retention. Also, the coding of the CIE at case closure variable does not take into consideration the client’s satisfaction with the employment outcome or the client’s prospects for advancement in that job. Future research should extend the frame of observation beyond the conventional 90 days case closure criterion to examine the long-term effects of supported employment services for young people with intellectual disabilities and clinical assessment instrument must be developed and validated specifically for transition-age youth and young adults with intellectual disabilities with competitive integrated employment outcomes.