State-Federal Vocational Rehabilitation in Traumatic Brain Injury: What Predictors Are Associated With Employment Outcomes?

Abstract

This study identified predictors of employment for individuals with traumatic brain injury (TBI). Data from 4,923 individuals with TBI were extracted from the Rehabilitation Services Administration’s Case Service Report (RSA-911) database. A multiple logistic regression model using demographics, disability-related variables, vocational rehabilitation (VR) service variables, and their interactions correctly classified 69.5% of the cases as successfully employed or not successfully employed. The model explained approximately 27.1% of the variance in employment outcomes. Results indicated that level of education, race/ethnicity, age at application, preemployment status, Supplemental Security Income (SSI), Social Security Disability Insurance (SSDI), comorbid depression, and case expenditure were significantly associated with employment outcomes (all p ≤ .05). VR variables that showed the most significant positive effect on employment outcomes were on-the-job support, job placement, and on-the-job training. Race/ethnicity moderated the effect of college training, supported employment, transportation, and extended evaluation or work trial assessment services on employment outcomes. The findings have implications for promoting the use of those VR services that are strongly related to employment outcomes for persons with TBI. They also point to the need for rehabilitation personnel to address some of the demographic and disability-related barriers to successful employment.

Keywords

vocational rehabilitation traumatic brain injury race ethnicity employment

Benefits of employment include increases in financial, physical, and psychological well-being, social participation, access to health insurance and health care, productivity, self-worth, provision of a sense of belonging, and quality of life (Brown, 2011; Chiu, Chan, Bishop, Cardoso, & O’Neil, 2013; Dutta, Gervey, Chan, Chou, & Ditchman, 2008). Unemployment, however, is associated with considerable stress that has social and health consequences. Personal stressors range from the anticipation of losing a job, to actual job loss, and seeking employment, but failing to find a job (Dooley, Fielding, & Levi, 1996). Social effects of unemployment include poverty, increased dependence on public support, increased crime rates, economic recession, reduction in spending power, and political instability. From a health perspective, unemployment can lead to poor physical and mental health, including a doubling of psychological problems in the unemployed versus the employed (Paul & Moser, 2009); a higher incidence of depression, anxiety disorders, alcohol use, and low self-esteem (Dutta et al., 2008); and increased risk of suicide (Dooley et al., 1996; Pawel & Kateryna, 2017).

The unemployment rate is not constant across the U.S. population; it varies by race/ethnicity, age, gender, health status, and disability. In 2017, the unemployment rate among working-age (16–64 years) Americans with disabilities was 9.2% compared with 4.2% in those without disabilities (Bureau of Labor Statistics [BLS], 2018). A 2012 BLS survey showed that 80.5% of working-age persons with a disability reported their disability as a barrier to employment (BLS, 2013). Among people with sensory/communicative, physical, and mental disabilities, people with mental impairments have the worst employment outcomes even after rehabilitation (Rosenthal, Chan, Wong, Kundu, & Dutta, 2001). Mental impairment can be as a result of substance abuse, biological factors, environmental factors, or brain trauma such as traumatic brain injury (TBI), which is the focus of this article.

TBI is a serious personal, public health, and societal concern in the United States (Noone, 2011; Tucker & Degeneffe, 2017). TBI is defined as a disruption in the normal function of the brain because of an externally inflicted trauma, such as a blow, jolt, or penetrating object to the head (Centers for Disease Control and Prevention, National Center for Injury Prevention and Control, 2019). According to Gary et al. (2009), many working-age people with disabilities due to TBI, especially those who were employed prior to TBI, look forward to returning to work in their communities after discharge from the hospital. Bloom et al.’s (2018) review found that 50% and 80% of individuals with mild TBI return to work 1 and 3 months, respectively, after TBI. However, a systematic review (van Velzen, van Bennekom, Edelaar, Sluiter, & Frings-Drescen, 2009) found that only about 40% of persons with disabilities related to TBI who were employed prior to their injury returned to work 1 or 2 years after the injury. The sequelae of TBI (i.e., physical, cognitive, emotional, and functional limitations and pain) present a barrier to social and cognitive functioning and employment (Catalano, Pereira, Wu, Ho, & Chan, 2006; Wehman, Bricout, & Targett, 2000).

Employment plays an important role in enhancing a meaningful life for individuals with TBI (Catalano et al., 2006), often serving as a sign of recovery (Levack, McPherson, & McNaughton, 2004). Aggressive vocational rehabilitation (VR) geared toward return to work plays a key role in TBI rehabilitation, and it has the capacity to provide direction, focus, and meaning to other treatments within the TBI continuum of care (Cardoso, Romero, Chan, Dutta, & Rahimi, 2007). The value of state-federal vocational rehabilitation (SFVR) program as a viable vocational intervention for people with disabilities, including TBI-related disabilities, is well documented (Dutta et al., 2008; Pete et al., 2015; Szymanski & Parker, 2010). For example, a review of SFVR participants with TBI whose cases were closed after the receipt of VR services between 1993 and 2001 found an increase in the rate of successful employment outcomes between 57% and 60% (Wehman, Targett, West, & Kregel, 2005).

Traumatic Brain Injury is one of the fastest growing disability groups that is being served by the SFVR program (Wehman et al., 2005). Because of the high cost (case expenditure) associated with treatment and rehabilitation and the importance of work to the health and well-being of persons with TBI, considerable research efforts have been devoted to studying the employment challenges of these individuals. Previous studies have examined the relationship of the receipt of SFVR services and employment among individuals with TBI. For example, employment is positively predicted by the provision of the following: job placement and job search assistance (Cardoso et al., 2007; Catalano et al., 2006; Schonbrun, Kampfe, & Sales, 2007), on-the-job training (Catalano et al., 2006; Johnstone, Vessell, Bounds, Hoskins, & Sherman, 2003b), counseling and guidance (Cardoso et al., 2007; Catalano et al., 2006; Gamble & Moore, 2003a; Johnstone et al., 2003b), college education/training (Cardoso et al., 2007; Catalano et al., 2006; Gamble & Moore, 2003a), on-the-job support, assistive technology (AT), and maintenance (Cardoso et al., 2007; Catalano et al., 2006). Supported employment (Gamble & Moore, 2003b), diagnosis and treatment of impairment (Schonbrun et al., 2007), and transportation (Catalano et al., 2006) also predicted employment. Supplemental Security Income (SSI), Social Security Disability Insurance (SSDI), and case expenditure are other factors that studies found to be predictive of employment outcome (Cardoso et al., 2007; Catalano et al., 2006).

Previous VR employment outcomes research, however, utilized single state VR data (Gamble & Moore, 2003a; Gamble & Moore, 2003b; Gamble & Satcher, 2002). Other researchers utilized the same small sample (n = 78) of participants (Bounds, Schoppa, Johnstone, Unger, & Goldman, 2003; Johnstone et al., 2003a; Johnstone et al., 2003b; Skeel, Bounds, Johnstone, Lioyd, & Harms, 2003) to analyze the relationship of demographic and VR service variables on employment outcomes. Only a few studies (Cardoso et al., 2007; Catalano et al., 2006; Pete et al., 2015; Schonbrun et al., 2007; Tucker & Degeneffe, 2017) utilized the national Rehabilitation Service Administration Case Service Report (RSA-911) data set, which contains VR service variables thought to enhance employment outcomes. Because treatment variables are potentially malleable factors and knowledge of their relationship to employment outcomes could be used to improve rehabilitation services, more studies that utilize this data set to identify predictors of successful employment are needed. In addition, in the few studies that utilized the RSA-911 data set to examine the receipt of VR services, there is little consistency in the choice of VR service variables chosen from the set of 24 VR services provided to persons with disabilities by the SFVR programs.

Several studies have investigated the effects of demographic and disability-related variables on employment outcome of individuals with TBI. Education, gender, race/ethnicity, and co-occurring psychiatric disabilities (e.g., depression) were found to influence employment (Cardoso et al., 2007; Catalano et al., 2006; Gamble & Moore, 2003a). When age was found to be a significant predictor, 16 to 34 years was reported as being positively related to employment (Cardoso et al., 2007). However, findings from meta-analysis and systematic and critical reviews are mixed for the relationship of race/ethnicity and employment after TBI (Crisp, 2005; Ownsworth & McKenna, 2004; Schonbrun & Kampfe, 2004). Similarly, findings from the extant VR literature on the impact of race/ethnicity on VR experiences have also been mixed with little consensus about its effect on VR services and employment outcomes (Moore et al., 2016).

Research on racial/ethnic variation in employment is critical for improving VR policy and practice (Martin, 2010). Although some VR studies examined the interactions of certain demographic and work disincentive variables in relation to rehabilitation outcomes for individuals with disabilities (Harley, Wilson, & DeShea, 2002; Rosenthal et al., 2001; Wheaton, Wilson, & Brown, 1996), few studies (Cardoso et al., 2007; Catalano et al., 2006) have examined the interaction effects of demographics, work disincentives, and VR service on employment outcomes of individuals with TBI. Little is still known about how race/ethnicity and the presence of comorbid depression or substance abuse disorder interact with each VR service variable to predict employment outcomes of individuals with TBI. In summary, there is no study that has examined, in a comprehensive manner, the unique and potentially moderating effects of these variables on employment outcomes of persons with TBI.

The purpose of this study was to investigate the factors that predict successful employment in individuals with TBI-related disabilities. Specifically, this study looked at the relationship between demographic characteristics of the individual, comorbid psychological disorders, and a given set of VR services received and interaction effects of these variables on successful employment outcomes. This study posed the following research questions:

Research Question 1: What set of demographics (e.g., race/ethnicity) and disability-related characteristics (e.g., depression) predict the employment outcomes of SFVR participants with TBI-related disabilities? How does the relative contribution of demographic and disability-related characteristics compare with that of the receipt of VR services?

Research Question 2: After controlling for the contributions of demographic and disability-related characteristics, what set of VR service variables are associated with employment outcomes at closure for SFVR participants with TBI-related disabilities?

Research Question 3: How does the relation between each predictor variable in the set of VR service variables investigated in Research Question 2 vary with respect to each of the following variables: race/ethnicity and comorbid psychiatric disabilities (e.g., depression)?

Method

Data Source and Participants

Data were extracted from RSA-911. This U.S. Department of Education data set contains information on all persons receiving VR services in the United States. It includes personal history, types of VR services, and employment outcome information for all SFVR participants whose cases were closed during a given year. Each SFVR agency provides RSA with these data for the persons that they serve for the year under review. The Fiscal Year (FY) 2012 RSA-911 data were used because it was the most recently available data set that the authors could gain access to at the time this study was initiated. Since 2014, and especially beginning FY 2017, there have been some changes to some RSA-911 database content, coding schemes, and data formatting. These changes, however, were mainly for compliance with other government programs and for reporting purposes. From FY 2017, RSA-911 data reporting has changed from annual to quarterly. One limitation with this change is that an individual could now have multiple closure status in a given year. For the current study, FY 2012 data provided variables and closure status relevant to our research questions. There were data for 4,923 individuals whose primary cause of impairment was TBI and whose cases were closed as successfully employed (i.e., placed in competitive employment) or unsuccessfully employed (i.e., not placed in competitive employment) after receiving VR services. The demographic/disability-related characteristics of this sample and the most commonly received VR services during FY 2012 are presented in Tables 1 and 2, respectively.

Table 1.

Summary of Selected Demographic and Disability-Related Characteristics of Persons With TBI.

Characteristics	n	%	M (SD)
Special education/less than high school	1,149	23.3
Completed high school	1,784	36.2
Some postsecondary/associate	1,390	28.2
College degree or higher	600	12.2
Age (15–74)			34.75 (13.17)
Gender: Male	3,347	67.9
Race/ethnicity: White	3,818	77.6
Black	655	13.3
Hispanic/Latino	333	6.8
Other races/ethnicities	117	2.3
Preemployment status (Yes/No)	677/4,246	13.8/86.2
Depression and other mood disorders	510	24.0
Substance use disorder	143	6.7
Significant disability (Yes/No)	4,856/67	98.6/1.4
SSI (Yes/No)	1,058/3,836	21.5/77.9
SSDI (Yes/No)	1,561/3,333	31.7/67.7
Case expenditure			US$5,728.87 (9.972.44)

Note. Twenty-nine (0.6%) individuals in this study had missing data for SSI/SSDI. TBI = traumatic brain injury; SSI = Supplemental Security Income; SSDI = Social Security Disability Insurance.

Table 2.

Commonly Provided Vocational Rehabilitation Services to SFVR Clients With TBI.

	Yes		No
	n	%	n	%
VR C&G	3,443	69.9	1,480	30.1
Assessment	3,414	69.3	1,509	30.7
Job placement assistance	2,122	43.1	2,801	56.9
Diagnosis and treatment	1,854	37.7	3,069	62.3
Transportation	1,615	32.8	3,308	67.2
Job search assistance	1,574	32	3,349	68.0
On-the-job support	1,322	26.9	3,601	73.1
Information referral	1,017	20.7	3,906	79.3
Job readiness training	889	18.1	4,034	81.9
Supported employment	866	17.8	3,996	82.2
Maintenance	749	15.2	4,174	84.8
College training	704	14.3	4,219	85.7
Assistive technology	451	9.2	4,472	90.8
On-the-job training	186	3.8	4,737	96.2
Technical assistance	84	1.7	4,839	98.3
Personal attendant	31	0.6	4,892	99.4
Interpreter services	7	0.1	4,916	99.9
Reader services	5	0.1	4,918	99.9

Note. VR C&G = vocational rehabilitation counseling and guidance. SFVR = state-federal vocational rehabilitation; TBI = traumatic brain injury.

Measures

The outcome variable used in the analysis is employment (successful or unsuccessful). Successful employment is defined in the 2012 RSA-911 as competitive employment in an integrated setting, self-employment, or employment in a state-managed Business Enterprise Program (BEP, which refers to vending facilities and small businesses operated by individuals with significant disabilities as well as home industry that falls under management of the SFVR agency). Furthermore, successful employment refers to either full-time or part-time work with compensation at or above the federal or state minimum wage, whichever is higher. Unsuccessful employment refers to individuals who exited the SFVR system without an employment outcome after completing their planned VR services.

Three sets of predictor variables were used for this study: demographics, disability-related characteristics, and VR services received. The demographic variables included the following: (a) level of education (special education, less than high school, high school, associate/some college education, and college degree or higher), (b) age at application, (c) gender (male or female), (d) race/ethnicity (White, Black/African American, Hispanic/Latino, Other), and (e) preemployment status (employed vs. unemployed). Disability-related characteristics included (a) public support (SSI, SSDI), (b) comorbid psychiatric disabilities (depression and substance use disorder), (c) severity of disability (significant vs. not significant), and (d) case expenditure. Case expenditure is included here because as with public support the amount spent in purchasing services to support the health and community reintegration of an individual with TBI is also disability related.

The third set of predictors included in this study was guided by the existing empirical VR literature on predictors of employment outcome (especially the more recent studies that utilized larger sample size and more rigorous methodologies). These include transportation, job search assistance, AT, maintenance, on-the-job training, job placement assistance, on-the-job supports, college or university training, and VR counseling and guidance. We also included three additional VR service variables—supported employment, assessment, diagnosis, and treatment of impairment. Except in the Catalano et al. (2006) study, these three VR service variables have rarely been examined together in previous studies even though they are often provided to individuals with significant disabilities such as TBI and may, based on face validity, be predictive of employment. According to the RSA-911 case Service Reporting Manual (2012), the definitions of the 12 VR services examined in this study are as follows:

Supported employment: This is an evidence-based intervention to individuals with significant disabilities who require intensive employment-related support to perform and maintain employment or work toward successful employment in an integrated setting.

Assessment: Services and activities to determine an individual’s eligibility for VR services and to determine the nature and scope of the VR services in the Individualized Plan for Employment (IPE). Examples are trial work experience and extended evaluation.

Diagnosis and treatment: Diagnosis and treatment for mental and emotional disorders, hospitalization (inpatient or outpatient) related to surgery/treatment, drugs and supplies, prosthetics, orthotics, physical therapy, occupational therapy, and speech or hearing therapy.

Transportation: Services and related expenses that enable an individual to participate in a VR service (e.g., purchase and repair of vehicles, use of public transportation training, relocation expenses, and employment-related personal care or aide services).

Job search assistance: Job search activities that support and assist a VR service recipient in their job hunt; may include assistance with preparing a resume, identifying job opportunities, developing interview skills, and contacting companies on behalf of the individual with a disability.

AT: Services to assist in the selection and acquisition of AT devices; it includes evaluation of an individual’s needs and functional evaluation in their home and community. Other services include leasing, customizing, and servicing of AT devices, and providing training and technical assistance for AT to an individual with a disability, family members, advocates, or authorized representative and trainings for professionals and employers who support the individual to obtain a successful employment outcome.

Maintenance services: This means financial support provided for those expenses (food, shelter, and clothing) that are more than the normal expenses of the individual and that are required for the individual’s participation in an assessment for determining their eligibility for VR services or while they are receiving services under the individual’s IPE.

On-the-job training: Refers to apprenticeship training programs and paid training in a specific job skill by a potential employer.

Job placement assistance: A referral to a specific job for a job interview, regardless of the employment outcome.

On-the-job support: Support services (such as job coaching, follow-up and follow-along, and job retention services) provided to an individual who has been placed in employment for the purpose of job stability and job retention.

College training: Full-time or academic training above the high school level leading to associate, baccalaureate, graduate, or professional degrees and provided by a 4-year college or university, community college, junior college, or technical college.

VR rehabilitation counseling and guidance: Discrete therapeutic counseling and guidance services required to achieve an employment outcome, including personal adjustment counseling, and any other form of counseling and guidance which is different from general counseling that exists between counselor and the individual during the entire rehabilitation process adapted from the RSA-911Case Service Reporting Manual (2012, p. 39).

Data Analysis

Data extracted from the 2012 RSA-911 data were analyzed using SPSS version 22. Multiple logistic regression analysis was employed to examine the effect of demographic characteristics, disability-related characteristics, and VR services on employment outcomes. Multiple logistic regression was chosen to address the primary research questions because multiple logistic regression allows for the prediction of a dichotomous dependent variable with covariates that are continuous, discrete, dichotomous, or categorical in nature (Keith, 2006; Tabachnik & Fidell, 2001).

In Step 1 of the multiple logistic regression model, the set of demographic characteristics was entered simultaneously. In Step 2, the set of disability-related characteristics was entered. The difference in −2 log likelihoods is shown in SPSS under block chi-square values (Tabachnik, & Fidell, 2001). The difference between the full model (set of demographic and disability-related predictors) and the reduced model (set of demographic characteristic variables) was examined to assess the significance of adding disability-related characteristic variables. In Step 3, the 12 VR service variables were entered together to determine their contributions to employment outcome over and above the set of demographic and disability-related variables.

In Step 4, the interaction variables were added to determine what they contribute to employment outcome over and above the sets of demographics, disability-related characteristics, and VR service contributions together. The predictive success of the multiple logistic regression models was assessed by looking at the classification tables, showing correct and incorrect classifications of the criterion variable. The significance of individual independent variables was analyzed by examining their coefficients (as with the Wald statistic) and associated odds ratios (ORs). Each categorical variable was coded dichotomously. The outcome variable, employment outcome, was also coded dichotomously (successfully employed or unsuccessfully employed).

Significance of the predictor variables was examined through p values ≤ .05, and the impact of these variables were determined by the unstandardized coefficients (B) at the 95% confidence interval (CI) and the standardized coefficients (β) (Chan, 2004). In addition, the Nagelkerke R² was used to measure the relationship of the observed values of the criterion variables and the predicted value based on the regression line (Chan, 2004). Prior to running the regression analysis, all 12 VR dichotomous predictor variable distributions were examined. Only “on-the-job training” exceeded the 10% to 90% rule (96.2% of individuals with TBI did not receive this service). However, this variable was retained in analysis because of its potential practical importance and its possible influence on other predictors in the multivariate analysis.

For multiple logistic regression, normality of continuous variables is not a requirement. However, because it is possible for outliers to affect the goodness of fit of the model, the continuous variables of age at application and case expenditure were checked for outliers. Descriptive measures showed that age at application was slightly skewed (skewness = 0.30) as was case expenditure (skewness = 0.63), although their distributions did not present outliers. We tested for multicollinearity with variance inflation factors (VIF). A multiple regression model with all the single independent variables and the outcome variable was run, and VIF were calculated; VIF values above 5 would be considered problematic and are worthy of concern (Cohen, Cohen, West, & Aiken, 2003). All variables of interest passed the criteria of no multicollinearity because no VIF values were above 5 (Cohen, Cohen, West, & Aiken, 2003).

Results

Demographic Characteristics

Table 1 summarizes the demographic and disability-related characteristics of the participants in this study. More than two thirds were males. There was a broad age range in terms of entry into the VR program with a mean age of 34 years. Most of the sample (77.6%) was White, followed by Black/African American (13.3%), and Hispanic/Latino (6.8%). More than a third had completed high school, somewhat less than a third of the sample had completed some postsecondary education/associate degree, and 12.2% completed a bachelor’s or higher degree. Most of the individuals with a history of TBI (86.2%) had had prior work experience.

Disability-Related Characteristics

Table 1 shows that close to a third of participants received SSDI and about 20% received SSI while their cases were still open. Almost all participants (98.6%) reported having a significant disability. Of the 61.5% of the participants who had a secondary impairment, psychosocial and cognitive impairments were the most widely reported. Among participants with TBI for whom cause of secondary impairment data was available, depressive and other mood disorders, accident/injury, other than TBI or spinal cord injury (SCI), and substance use disorder were the most commonly reported causes of secondary impairment.

VR Services Provided

Most of the 24 VR services provided by SFVR programs were received by individuals with TBI-related injuries. The average length of time an individual spent in SFVR program was 33.99 months (SD = 27.90). Table 2 shows a sample of the common VR services provided to participants.

Multiple Logistic Regression Analysis

Individuals with TBI-related disabilities with missing data on some VR services received, such as supported employment services, were removed from data analysis. The cases with missing information were approximately 1.9% of the sample. The final sample size N = 4,831 that was used for data analysis was still large enough to meet the minimum ratio rule of 10 cases per predictor (Peduzzi, Concato, Feinstein, & Holford, 1996). The model was computed to analyze the main effects of demographic, disability-related characteristics, and VR service variables on employment outcomes. In addition, the interactions of race/ethnicity, depression, and Substance Use Disorder × VR Services on outcomes were examined.

The omnibus test for the multiple logistic regression model was statistically significant, χ²(88, N = 4,831) = 1,095.75, p < .005. The Nagelkerke R² was .27, indicating that the predictors moderately explained the variance in successful employment. The model correctly classified 69.5% of cases as successfully employed or not successfully employed.

Research Question 1: What demographics (e.g., race/ethnicity) and disability-related characteristics (e.g., depression) predict the employment outcomes of SFVR participants with TBI-related disabilities? How does the relative contribution of demographic and disability-related characteristics compare with that of the receipt of VR services?

The contribution of demographic characteristics was investigated using multiple logistic regression analysis. To test the effect of demographic variables on employment outcomes, the first set of variables entered into the analysis were demographic variables (see Table 1). The Nagelkerke R² of the first regression model was .055, indicating that 5.5% of the variance in employment outcomes could be explained by demographic predictors alone. Table 3 shows that the odds of individuals finding successful employment who had some college (OR = 1.15; 95% CI = [1.03, 1.29]) or a college or higher degree (OR = 1.37; 95% CI = [1.17, 1.60]) were 1.15 and 1.37 times, respectively, greater than the odds of those with only a high school diploma/degree.

Table 3.

Relationship of Demographic/Disability-Related Characteristics and Employment Expressed as Odds Ratio.

Independent variable	Block 1		Block 2
Independent variable	OR (95% CI)	p	OR (95% CI)	p
Demographic variables
Completed high school	1 (Reference)
Special education	0.792 [0.624, 1.004]	.050
Less than high school	0.801 [0.702, 0.914]	.001
Some college degree	1.152 [1.026, 1.293]	.017
College degree or higher	1.366 [1.165, 1.601]	<0.001
Age at application	0.987 [0.983, 0.992]	<0.001
Gender^a	Male (Reference)
Female	1.033 [0.970, 1.099]	.311
Race/ethnicity
White	1 (Reference)
Black	0.850 [0.722, 1.000]	.050
Hispanic	1.123 [0.923, 1.366]	.245
Other races/ethnicities	0.936 [0.701, 1.251]	.656
Preemployment status^b	1.689 [1.542, 1.849]	<0.001
Disability-related variables
SSI^b			0.778 [0.722, 0.838]	<.001
SSDI^b			0.860 [0.805, 0.919]	<.001
Depression^b			0.848 [0.769, 0.936]	.001
Substance use disorder^b			0.910 [0.763, 1.084]	.291
Significant disability^b			0.813 [0.625, 1.056]	.120
Case expenditure			1.040 [1.031, 1.049]	<.001

Note. OR = odds ratio; CI = confidence interval; Some college degree = postsecondary/associate; SSI = Supplemental Security Income; SSDI = Social Security Disability Insurance.

Male is the reference group. ^bDichotomous variables with Yes/No response; No is the reference group.

The odds ratio with age at application was 0.987, which indicated that the likelihood of being competitively employed at case closure decreases slightly with each additional year increase in age at the time of application. The overall effect of race/ethnicity—χ²(df = 3) = 10.483, p = .015—was significantly related to employment outcome. The likelihood (OR = 0.85, 95% CI = [0.72, 1.00]) of being successfully employed at closure was decreased by 15% for Black/African American individuals compared with White individuals. Individuals who had been employed in the past had a 69% increase in odds (OR = 1.69; 95% CI = [1.54, 1.85]) of obtaining successful employment at closure compared with individuals with no prior employment experience. The difference in the Nagelkerke R² between the second regression models with both demographic and disability-related variables together and the first model with only demographic variables was .047, indicating that an additional 4.7% of the variance in employment outcome could be explained by the addition of these disability-related characteristics.

Significant findings for disability-related variables are in Table 3. Specifically, individuals who received SSI and SSDI were found to have a 22% reduction in odds (OR = 0.78; 95% CI = [0.72, 0.84]) and 14% reduction in odds (OR = 0.86; 95% CI = [0.81, 0.92]), respectively, of becoming successfully employed in comparison with those who did not receive such benefits. Co-occurring depression was a risk factor for people with TBI, resulting in a 15% lower chance of employment success (OR = 0.85; 95% CI = [0.77, 0.94]). Demographic and disability-related characteristics together accounted for 10.2% of the total variance in employment outcomes. The difference in the Nagelkerke R² between the third (set of VR service variables, demographic, and disability-related variables) and the second regression models indicates a 15% contribution in employment outcomes by the set of VR service variables alone

Table 4 shows that 10 VR services were found to be statistically significantly related to employment outcomes. Of these 10 VR predictors, six were significantly and positively related to successful employment outcomes across all races/ethnicities.

Table 4.

The Association of VR Services/Interaction Variables With Employment Expressed as Odds Ratio.

Independent variable	Block 3		Block 4
Independent variable	OR (95% CI)	p	OR (95% CI)	p
Supported employment^a	0.728 [0.661, 0.801]	<.001
Assessment^a	0.905 [0.844, 0.970]	.005
Diagnosis and treatment^a	0.907 [0.848, 0.969]	.004
Transportation^a	0.926 [0.863, 0.995]	.035
Job search assistance^a	1.094 [1.016, 1.178]	.017
Assistive technology^a	1.123 [1.003, 1.257]	.044
Maintenance^a	1.171 [1.068, 1.284]	.001
On-the-job training^a	1.448 [1.206, 1.739]	<.001
Job placement^a	1.490 [1.390, 1.597]	<.001
On-the-job support^a	1.901 [1.744, 2.072]	<.001
College training^a	0.978 [0.886, 1.078]	.651
VR C&G^a	1.064 [0.991, 1.142]	.086
Transportation^a × Black^b			0.808 [0.656, 0.996]	.045
Transportation^a × Other^b			1.863 [1.251, 2.775]	.002
Supported Employment^a × Black^b			0.620 [0.441, 0.870]	.006
Supported Employment^a × Hispanic^b			0.625 [0.407, 0.960]	.032
Supported Employment^a × Other^b			3.261 [1.513, 7.025]	.003
College Training^a × Black^b			1.524 [1.114, 2.085]	.008
Assessment^a × Black^b			1.258 [1.016, 1.558]	.035
College Training^a × Substance Use^b			1.568 [1.167, 2.106]	.003

Note. VR = vocational rehabilitation; Other = Other races/ethnicities; Black = Black/African American; OR = odds ratio; CI = confidence interval; VR C&G = vocational rehabilitation counseling and guidance.

a, b

Dichotomous variables with Yes/No response; No is the reference group.

Most notably, on-the-job support services (OR = 1.90; 95% CI = [1.74, 2.07]), job placement assistance (OR = 1.49; 95% CI = [1.390, 1.597]), and on-the-job training (OR = 1.45; 95% CI = [1.206, 1.739]) showed the most positive effects on successful employment outcomes; VR clients with TBI who received on-the-job support services, job placement assistance, and on-the-job training had a 90%, 49%, and 45% increase in odds, respectively, of obtaining successful employment compared with those who did not receive these VR services.

Additional positive predictors of successful employment were maintenance, AT, and job search assistance, with 17%, 2%, and 9% increases in odds, respectively, of obtaining successful employment compared with individuals who did not receive these VR services. Counseling and guidance was not a significant predictor of employment outcome.

To answer this research question, the interaction terms between each level of race/ethnicity and VR services, the interaction terms between comorbid depression and VR services, and the interaction terms between substance use and VR services were entered separately in the fourth block of the final regression model. After controlling for other variables, Table 4 shows the eight interaction effects that this study found to be significant.

Three of these interaction effects could be considered risk factors and have to do with race/ethnicity; they were Transportation × Black/African American (OR = 0.81; 95% CI = [0.66, 1.00]), Supported Employment × Black/African American (OR = 0.62; 95% CI = [0.44, 0.87]), and Supported Employment × Hispanic/Latino (OR = 0.63; 95% CI = [0.41, 0.96]). Black/African American clients who received transportation and supported employment had a 19% and 38%, respectively, reduction in odds of being employed than did White individuals who received similar services. Similarly, Hispanic/Latino clients who received supported employment toward working to obtain and maintain employment had a 37% significantly lower odds of obtaining employment than did White clients who received supported employment.

In contrast, Black/African American clients who received a comprehensive assessment service (OR = 1.26; 95% CI = [1.02, 1.56]) to determine the nature and scope of the VR service provision they needed to achieve employment and who received college training (OR = 1.52; 95% CI = [1.11, 2.09]) had significantly higher odds (i.e., 26% and 52%, respectively) of obtaining successful employment than did White clients who received the same services. All other races in the “Other” group who received transportation and supported employment services had about 2 and 3 times, respectively, higher odds of obtaining successful employment than did White clients who received the same services.

Although the main effect of comorbid substance use was not significant (Table 3), Table 4 shows that the interaction of substance use disorder with the receipt of college training was statistically significant (OR = 1.57; 95% CI = [1.17, 2.11]). In other words, clients with comorbid substance use who needed college training services had 57% odds of obtaining successful employment than did clients with comorbid substance use who did not need this service.

Discussion

The current study sought to examine the potential influence of various demographic, disability-related characteristics, and VR service variables on the successful employment outcomes of individuals with TBI-related disabilities. The finding of significant disparity in employment outcome between Whites and Blacks/African Americans in the current study is supported by previous studies (Atkins & Wright, 1980; Wilson & Senics, 2005). The lack of significant disparity in employment outcomes between Hispanic/Latinos and Whites is also consistent with the Cardoso et al. (2007) finding of no major disparity in employment outcomes between these groups, after taking into consideration other demographic factors.

As expected, having a preemployment status, younger age, and having a degree higher than a high school diploma were positive predictors of successful employment outcomes. These findings are consistent with what is known about the larger population of individuals who sustain TBI injury in the United States and to findings from more recent VR studies (i.e., Cardoso et al., 2007; Catalano et al., 2006; Pete et al., 2015) that examined the relationship between VR service variables and employment outcomes, and Tucker and Degeneffe’s (2017) study that examined the predictors of employment following postsecondary education for VR participants.

Individuals who received SSI/SSDI support while in a SFVR program had significantly lower successful employment outcomes compared with individuals who did not receive SSI/SSDI. This finding is consistent with findings from the extant literature. However, care must be taken in interpreting the negative relationship between receipt of SSDI/SSI and employment success because a third unmeasured variable (e.g., lack of family support, fear of losing medical benefits, or SSDI/SSI eligibility issues) might be relevant in explaining this association. For example, it is possible that the way the policy of Social Security work incentive eligibility is structured might be a work disincentive for beneficiaries (Weathers & Hemmeter, 2011). Currently, SSI/SSDI beneficiaries run the risk of losing their benefits if they earn an income above a set threshold or do work that is considered substantial gainful activity. According to Hennessey’s (1997) finding, if SSDI recipients have benefit counseling, knowledge of Trial Work period, and assurance of continued Medicare coverage, the effect of work disincentive would have dissipated.

For persons with TBI-related disabilities, co-occurring depression reduced the odds of successful employment. In contrast, the main effect of substance use disorder was not significant. This finding is consistent with Catalano et al.’s (2006) finding. Like Catalano et al.’s study in which the incidence of comorbid substance use was relatively low (3.0%), the incidence of comorbid substance abuse disorder in the current was also low at 6.7%. However, both figures pale in comparison to rates of 36% to 66% substance use reported in systematic review studies (Corrigan, Rust, & Lamb-Hart, 1995; Parry-Jones, Vaughan, & Cox, 2006). The nonsignificant main effect of substance use on employment outcomes may be related to the very low incidence (6.7%) of substance abuse issues in the current study cohort.

After controlling for the effect of demographic and postinjury covariates, VR services significantly improved the employment outcome of individuals with TBI-related disabilities. An important finding from this study is that on-the-job support, job placement assistance, on-the-job training, maintenance, AT, and job search assistance services appear to be the most important predictors, and hence the most helpful vocational services to VR individuals with TBI. However, several of these services were less frequently provided to individuals with TBI-related injuries (see Table 2). Previous studies (Cardoso et al., 2007; Catalano et al., 2006) also found these VR services to be significant predictors of employment outcomes.

Of all VR services, on-the-job supports service most significantly (i.e., 90% increase in odds) improved employment outcomes. Conversely, even though VR counseling and guidance was the most widely provided (69.9%) VR service to this cohort of individuals with TBI, the main effect of VR counseling and guidance service on employment outcome was not significant. This result is consistent with findings from the Cardoso et al. (2007) study. It is possible that because of the presence of a powerful variable (i.e., supported employment) examined in the present study, the effect of VR counseling and guidance was no longer significant.

Although the nonsignificant association of VR counseling and guidance with successful employment for people with TBI is counter-intuitive to what you would expect among the general population of SFVR individuals with disabilities, it is an important finding that warrants further examination. According to Butler and Statz (1988), most professionals working with persons with TBI are not familiar with the special needs and issues of these individuals. Therefore, it is important that counselors are trained to meet the needs of individuals with TBI and understand how the sequelae of TBI affect school and work activities.

Another important finding from the current study is the significant racial differences in VR service provision and outcomes. Of all VR services, college training and assessment most significantly improved the employment outcomes of Blacks/African Americans. The positive interaction effects of assessment and college training suggest that for Blacks/African Americans in particular, college training and assessment improved employment outcomes more so than the same services for other ethnic and racial groups. It is possible that Blacks/African Americans received more extended evaluation and work trial services in the eligibility and IPE stages of the VR process and were therefore provided with more specific VR services tailored toward successful college training and employment outcome. Given the high cost of VR services involving college training (Catalano et al., 2006) and because only 14.3% (Table 2) of this cohort received college training, additional research is needed to examine the interaction of extended evaluation assessment, race/ethnicity, and college training.

Black/African Americans who needed supported employ-ment and transportation services had a higher risk of failure after receiving VR services. Supported employment is an evidence-based effective employment-related support; it is an expensive service and it is often externally purchased. Transportation is one of the basic/lower cost services provided through VR to enable an individual with transportation issues to participate in VR. A plausible reason exists for these findings. It is possible that Whites were more likely to receive externally purchased supported employment services and less likely to receive basic transportation services than Blacks/African Americans. This finding is supported by some of the preamble to the 1992 amendments to the Rehabilitation Act of 1973, and some consistency in findings from three decades of VR disparity research which shows that in comparison with Whites, individuals from racial/ethnic minority groups (a) are more likely to receive fewer services, (b) are more likely to receive lower cost services, and (c) are less likely to receive externally purchased services (Bellini, 2003; Feist-Price, 1995; Martin, 2010).

In contrast, for individuals with TBI who belong to the “Other” race category, supported employment and transportation services most significantly improved their employment outcomes. This finding was surprising, especially when the Whitfield and Lloyd (2008) disparity research found that fewer funds were expended on purchased VR services for American Indians/Native Alaskans (AI/NA) than for non-AI/NA. It is possible that a third unmeasured variable such as counselor multicultural competency (Bellini, 2003) may explain this variation. Future studies that disaggregate the receipt of VR services by race/ethnicity and that examine disparity in outcomes in the context of counselor multicultural competency are needed.

Another important finding is the significant interaction between college training and substance use disorder on employment outcome. College training had a significant positive impact on employment outcomes for individuals recovering from comorbid substance use disorders. This is consistent with Pete et al.’s (2015) findings that VR positively improves the employment outcomes of young Black/African American men recovering from substance use disorders. However, because relatively few individuals had comorbid substance use in the current study compared with rates in other TBI studies, this finding deserves further investigation.

Limitations of This Study

This study used archival data and employed an ex post facto design; therefore, causality cannot be inferred from the findings. It is important to note that the data are from FY 2012 RSA-911 state-federal database, which were the most current data that the authors could access at the time the study was initiated. Since 2014 and especially beginning 2017, there have been some changes in some variables, coding schemes, and data formatting in the RSA-911 database. The reason for the changes is mainly for compliance purposes with other government programs and to improve the reporting capabilities of VR agencies. We assume, but cannot guarantee, that these changes do not affect the generalizability of our findings to more recent cohorts.

Because of the dynamic nature of employment, it is necessary to assess employment over time. Future studies should consider using more recent multiple RSA-911 databases to assess whether the important findings of racial variation in vocational service provisions and outcomes in this current study are in fact isolated to FY 2012 or a trend that reflects systematic concerns.

Because counselors enter data at different points during the VR process, the reliability of the data may be affected. Other limitations discussed earlier concern the ability of rehabilitation counselors in the SFVR system to accurately identify and provide services for injury-related factors such as substance abuse disorder that might be important predictors of employment outcomes. The relatively low rate of this diagnosis in the database points to concerns about the validity of the coding of the substance abuse variable. Given the exploratory nature of this study and the relatively small sample size (n = 117) of ethnic minorities belonging to the “Other” group, the significant findings of racial disparity in VR service provision and outcomes found in this study should be regarded as tentative, pending confirmation in future research. A study that examines trend over time and that adopts a different statistical model such as a randomized split-half cross-model validation research design is warranted. Finally, because the database does not include people with TBI in the civilian and Veteran Administration TBI Model System databases, some of the findings from the current study may not generalize to all individuals with TBI.

Conclusion

The SFVR program provides important interventions to assist persons with TBI history gain successful employment. Specifically, VR service provision that is geared toward individuals with TBI and that provide on-the-job support, job placement, and on-the-job training is more likely to lead to successful employment outcomes than general VR service provision. VR services provide viable direction and input to other therapies within the continuum of care for individuals with TBI-related disabilities. Collaboration between public SFVR and other health professionals to promote the use of VR services to improve public health outcomes for individuals with TBI is warranted.

Footnotes

Acknowledgements

The authors thank Dr. Susan, N. Beretvas and Dr. Michael J. Mahometa for their support.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Atkins

B. J.

Wright

G. N.

(1980). Three views: Vocational rehabilitation of Black or African Americans—The statement the response the comment. Journal of Rehabilitation, 46(2), 40–49.

Bellini

(2003). Counselors’ multicultural competencies and vocational rehabilitation outcomes in the context of counselor-client racial similarity and difference. Rehabilitation Counseling Bulletin, 46, 164–173.

Bloom

Thomas

Ahrensberg

J. M.

Weaver

Fowler

Bestwick

Pearse

(2018). A systematic review and meta-analysis of return to work after mild traumatic brain injury. Brain Injury, 32, 1623–1636.

Bounds

T. A.

Schoppa

Johnstone

Unger

Goldman

(2003). Gender differences in a sample of vocational rehabilitation clients with TBI. Neurorehabilitation, 18, 189–196.

Brown

(2011). Career information, career counseling, and career development (10th ed.). Boston, MA: Allyn & Bacon.

Bureau of Labor Statistics. (2013). Persons with a disability: Barriers to employment, types of assistance, and other labor-related issues. Retrieved from https://www.bls.gov/news.release/archives/dissup_04242013.pdf

Bureau of Labor Statistics. (2018). Persons with a disability: Labor force characteristics summary. Retrieved from https://www.bls.gov/news.release/disabl.nr0.htm

Butler

R. W.

Statz

(1988). Individual psychotherapy with head-injured adults: Clinical note for the practitioner. Professional Psychology: Research and Practice, 19, 536–541.

Cardoso

Romero

M. D.

Chan

Dutta

Rahimi

(2007). Disparities in vocational rehabilitation services and outcomes for Hispanic clients with traumatic brain injury: Do they exist? Journal of Head Trauma Rehabilitation, 22(2), 85–94.

10.

Catalano

Pereira

A. P.

Chan

(2006). Service patterns related to successful employment outcomes of persons with traumatic brain injury in vocational rehabilitation. Neurorehabilitation, 21, 279–293.

11.

Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. (2019). Traumatic brain injury & concussion. Retrieved from https://www.cdc.gov/traumaticbraininjury/index.html

12.

Chan

Y. H.

(2004). Biostatistics 201: Linear regression analysis. Singapore Medical Journal, 45, 55–61.

13.

Chiu

Chan

Bishop

Cardoso

O’Neil

(2013). State vocational rehabilitation services and employment in multiple sclerosis. Multiple Sclerosis Journal, 19, 1655–1664.

14.

Cohen

West

S. G.

Aiken

L. S.

(2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed.). Mahwah, NJ: Lawrence Erlbaum.

15.

Corrigan

J. D.

Rust

Lamb-Hart

G. L.

(1995). The nature and extent of substance abuse problems among persons with traumatic brain injuries. Journal of Head Trauma Rehabilitation, 10(3), 29–45.

16.

Crisp

(2005). Key factors related to vocational outcome: Trends for six disability groups. Journal of Rehabilitation, 71(4), 30–37.

17.

Dooley

Fielding

Levi

(1996). Health and unemployment. Annual Review Public Health, 17, 449–465.

18.

Dutta

Gervey

Chan

Chou

C. C.

Ditchman

N. A.

(2008). Vocational rehabilitation services and employment outcomes for people with disabilities: A United States study. Journal of Occupational Rehabilitation, 18, 326–334.

19.

Feist-Price

(1995). African Americans with disabilities and equity in vocational rehabilitation services: One state’s review. Rehabilitation Counseling Bulletin, 39, 119–129.

20.

Gamble

Moore

C. L.

(2003a). The relation between vocational rehabilitation services and successful employment of individuals with traumatic brain injury. Journal of Rehabilitation, 69(3), 31–38.

21.

Gamble

Moore

C. L.

(2003b). Supported employment: Disparities in vocational rehabilitation outcomes, expenditures and service time for persons with traumatic brain injury. Journal of Vocational Rehabilitation, 19, 47–57.

22.

Gamble

Satcher

(2002). Rehabilitation outcomes, expenditures, and the provision of assistive technology for persons with traumatic brain injury. Journal of Applied Rehabilitation Counseling, 33(3), 41–44.

23.

Gary

K. W.

Arango-Lasprilla

J. C.

Ketchum

J. M.

Kreutzer

J. S.

Copolillo

Novack

T. A.

Jha

(2009). Racial differences in employment outcome after traumatic brain injury at 1, 2, and 5 years post-injury. Archives of Physical Medicine and Rehabilitation, 90, 1699–1707.

24.

Harley

D. A.

Wilson

K. B.

DeShea

(2002). The exploration of social security disability insurance, race and work status at closure on successfully rehabilitated customers in the United States. Journal of Rehabilitation Administration, 25, 137–143.

25.

Hennessey

J. C.

(1997). Factors affecting the work efforts of disabled-worker beneficiaries. Social Security Bulletin, 60, 3–20.

26.

Johnstone

Mount

Gaines

Goldfader

Bounds

Pitts

(2003a). Race differences in a sample of vocational rehabilitation clients with traumatic brain injury. Brain Injury, 17, 95–104.

27.

Johnstone

Vessell

Bounds

T. A.

Hoskins

Sherman

(2003b). Predictors of success for state vocational rehabilitation clients with traumatic brain injury. Archives of Physical Medicine and Rehabilitation, 84, 161–167.

28.

Keith

T. Z.

(2006). Multiple regression and beyond: An introduction to multiple regression and structural equation modeling. Boston, MA: Pearson.

29.

Levack

McPherson

McNaughton

(2004). Success in the workplace following traumatic brain injury: Are we evaluating what is most important? Disability Rehabilitation, 26, 290–298.

30.

Martin

F. H.

(2010). Racial variation in vocational rehabilitation outcomes: A structural equation modeling approach. Rehabilitation Counseling Bulletin, 54, 26–35.

31.

Moore

C. L.

Wang

Johnson

Manyibe

E. O.

Washington

A. L.

Muhammad

(2016). Return-to-work outcome rates of African American versus White Veterans served by state vocational rehabilitation agencies: A randomized split-half cross-model validation research design. Rehabilitation Counseling Bulletin, 59, 158–171.

32.

Noone

(2011). Global health, social participation, disability and employment. Occupational Medicine, 61(1), 71.

33.

Ownsworth

McKenna

(2004). Investigation of factors related to employment outcome following traumatic brain injury: A critical review and conceptual model. Disability and Rehabilitation, 26, 765–784.

34.

Parry-Jones

B. L.

Vaughan

F. L.

Cox

W. M.

(2006). Traumatic brain injury and substance misuse: A systematic review of prevalence and outcomes research (1994–2004). Neuropsychological Rehabilitation, 16, 537–560.

35.

Paul

I. K.

Moser

(2009). Unemployment impairs mental health: Meta-analyses. Journal of Vocational Behavior, 74, 264–282.

36.

Pawel

Kateryna

(2017). Short-run and long-run effects of unemployment on suicides: Does welfare regime matter? European Journal of Public Health, 27, 1038–1042.

37.

Peduzzi

Concato

Feinstein

A. R.

Holford

T. R.

(1996). Importance of events per independent variable in proportional hazards regression analysis: II—Accuracy and precision of regression estimates. Journal of Clinical Epidemiology, 48, 1503–1510.

38.

Pete

J. P.

Diallo

Kaya

Brooks

Allen

Bezyak

Chan

(2015). Vocational rehabilitation as a public health intervention for young African American men with substance use disorders. Journal of Vocational Rehabilitation, 43, 149–157.

39.

Rehabilitation Services Administration (2012). Reporting manual for the case service (Report RSA-911, RSAPD United States). Department of Education, Office of Special Education and Rehabilitative Services. Retrieved from https://www2.ed.gov/programs/rsabvrs/rsa-911-case-service-report-manual-2016.pdf

40.

Rosenthal

D. A.

Chan

Wong

D. W.

Kundu

M2. M.

Dutta

(2001). The effects of gender, race, disability type, work disincentives, and service patterns on employment outcomes of vocational rehabilitation consumers. Retrieved from https://pdfs.semanticscholar.org/c43e/b848488e77fe8815bdd231772b427f545ef1.pdf

41.

Schonbrun

S. L.

Kampfe

C. M.

(2004). Variables relating to successful employment in individuals with traumatic brain injury. Journal of Applied Rehabilitation Counseling, 35(3), 14–21.

42.

Schonbrun

S. L.

Kampfe

C. M.

Sales

A. P.

(2007). Rehabilitation services administration services and successful employment in individuals with traumatic brain injury. Journal of Rehabilitation, 73(2), 26–31.

43.

Skeel

R. L.

Bounds

Johnstone

Lioyd

Harms

(2003). Age differences in a sample of state vocational rehabilitation clients with traumatic brain. Rehabilitation Psychology, 48, 145–150.

44.

Szymanski

E. M.

Parker

R. M.

(2010). Work and disability: Contexts, issues, and strategies for enhancing employment outcomes for people with disabilities (3rd ed.). Austin, TX: Pro-Ed.

45.

Tabachnik

Fidell

L. S.

(2001). Using multivariate statistics (4th ed.). Needham Heights, MA: Allyn & Brown.

46.

Tucker

M. S.

Degeneffe

C. E.

(2017). Predictors of employment following postsecondary education for vocational rehabilitation participants with traumatic brain injury. Rehabilitation Counseling Journal, 60, 215–226.

47.

van Velzen

J. M.

van Bennekom

C. A.

Edelaar

M. J.

Sluiter

J. K.

Frings-Drescen

M. H

. (2009). How many people return to work after acquired brain injury? A systematic review. Brain Injury, 23(6), 473–488.

48.

Weathers

I. I. R. R.

Hemmeter

(2011). The impact of changing financial work incentives on the earnings of Social Security Disability Insurance (SSDI) beneficiaries. Journal of Policy Analysis and Management, 30, 708–728.

49.

Wehman

Bricout

Targett

(2000). Supported employment for persons with traumatic brain injury: A guide for implementation. In Fraser

R. T.

Clemons

D. T.

(Eds.), Traumatic brain injury rehabilitation: Practical vocational, neuropsychological and psychotherapy interventions (pp. 201–240). Boca Raton, FL: CRC Press.

50.

Wehman

Targett

West

Kregel

(2005). Productive work and employment for persons with traumatic brain injury: What have we learned after 20 years? Journal of Head Trauma Rehabilitation, 20, 115–127.

51.

Wheaton

J. E.

Wilson

K. B.

Brown

S. M.

(1996). The relationship between vocational rehabilitation services and the consumer’s sex, race, and closure status. Rehabilitation Counseling Bulletin, 40, 116–133.

52.

Whitfield

H. W.

Lloyd

(2008). American Indians/Native Alaskans with traumatic brain injury: Examining the impairments of traumatic brain injury, disparities in service provision, and employment outcomes. Rehabilitation Counseling Bulletin, 51, 190–192.

53.

Wilson

K. P.

Senics

(2005). Exploring the vocational rehabilitation acceptance rates for Hispanics versus non-Hispanics in the United States. Journal of Counseling Development, 83, 86–96.