Abstract
BACKGROUND:
Adults with neurodevelopmental disability (NDD) have poor employment outcomes when compared to their peers without disabilities. Examining employment outcomes beyond common dichotomous descriptive metrics (i.e., employed versus unemployed) depends on the use of standard measures and structured procedures.
OBJECTIVE:
This review of vocational research literature focused on identifying measures of employment outcomes for adults with NDD.
METHODS:
Searches were conducted across five databases - ERIC, MEDLINE, CINAHL, HaPI, and PsycINFO. Screening was conducted in duplicate, with all disagreements adjudicated by the senior researcher.
RESULTS:
A total of 45 articles met inclusion criteria, and data extraction revealed that 64 different employment measures were used in these vocational research studies.
CONCLUSIONS:
This work summarizes the employment measures for people with NDD utilized in the literature. Descriptions of these measures were provided and coding by person and environment themes, which is a useful resource for planning future vocational research for people with NDD.
Introduction
Global issue of neurodevelopmental disabilities and employment
In the industrialized world, approximately one in six children have neurodevelopmental disabilities (NDD), such as, cerebral palsy, autism, attention deficit hyperactivity disorder, and developmental delay (Dietrich et al., 2005). With an increase in the lifespan of children with NDD, recent reports, including the World Health Organization (WHO), have emphasized the importance of a lifespan approach to care and community integration, including employment (WHO, 2013). Data from the WHO from across 51 countries provides clear evidence of lower employment rates for persons with disabilities (52.8% males, 19.6% females) than for their peers without disabilities (64.9% males, 29.9% females) (WHO, 2011). For persons of working age with disabilities, data from the developed world show rates are as low as 50 to 70%, at least twice the rate of persons without disabilities (International Disability Rights Monitor, 2004; WHO, 2011). In developing countries, these rates are much worse since the focus of employment for people with NDDs was largely unaddressed (Durkin, 1991), and recent estimates of unemployment may be as high as 80 to 90% (Zarocostas, 2005).
In 2006, the United Nations (UN) adopted the Convention on the Rights of Persons with Disabilities (CRPD), which specifies the right to equal participation in work and employment. This document calls for access to vocational guidance, placement and training programmes, as well as ensuring opportunities for job placement, accommodations and supports (CRPD, article 27, items a-k) (UN, 2006). To address the issues related to employment for people with NDD, it is important to collect valid, reliable and relevant data that can be uniform around the world (WHO, 2013). The WHO Global Disability Action Plan (2014–2021) calls for member states to collect data that is relevant and internationally comparable to support research on disability and related services (WHO, 2015). A fundamental step in this global action plan involves identifying and developing sound measures, including standardized instruments and procedures, especially in the area of employment.
Vocational rehabilitation services and research tools
Over the last 20 years, there have been initiatives around the world aimed at providing vocational rehabilitation to persons with disabilities (Chan et al., 1997; UN, 2006; WHO, 2011; WHO, 2013). Specific examples from North and South America, Europe, Asia and Africa have demonstrated how employment agencies have responded to the needs of job seekers with disabilities by expanding their activities for improving employment opportunities (WHO, 2011). Vocational rehabilitation (VR) training programmes are ultimately designed with the goal of enabling better rates of participation in the labour market. The foundation of VR services involves information gathering about the client’s interests and abilities, in order to establish a good job matching and facilitate an on-site training plan (Hall et al., 2014).
Related work opportunities have historically been available through sheltered workshop environments, supported community settings, or competitive placements within the broader community (Jiranek & Kirby, 1990; Kober & Eggleton, 2005; Cimera et al., 2012). Disability researchers have traditionally examined the success of VR programs through descriptive employment status metrics. In many studies, the most commonly explored metrics include job status (i.e., employed or not) and wages (i.e., hourly rate) (Cimera et al., 2010, 2011). These are important metrics contributing to the global picture of employment for persons with disability; however, to better measure the sensitivity of factors that impact on employment, both “personal” and “environmental” factors for work functioning must be considered (WHO, 2011b; WHO, 2011). When planning future interventions and studies, researchers should think beyond monitoring typical employment outcomes of job rate and compensation. While these data are essential for informing disability policy, they are not sufficient to informing new directions of practice towards meeting local needs.
VR professionals and researchers have a wide array of employment outcome instruments from which to choose because of the different sectors involved, such as health, education and social services. These sectors often operate independently, and to date there is no centralized list of employment outcome measures utilized across sectors. Knowing what other instruments are available in the literature is useful because this saves work and the necessity of constructing new measures (Norman & Streiner, 2008). Furthermore, understanding the properties of existing employment outcome measures across NDD can provide insights on the common and distinct constructs measured for different groups and in different contexts. These insights can be helpful with identifying common tools to be used for better surveillance and service development, as well as inform specific diagnostic groups within NDD about different constructs that can better inform employment outcome evaluation. To identify the employment outcomes across sectors in the field of VR, a review was conducted to survey the standard scales and systematic procedures used for employment outcomes measured in neurodevelopmental disability research.
Methods
The objective of this study is to identify the peer review literature for employment outcome instruments used by VR disability researchers for adults with NDD. Inclusion criteria consisted of studies involving adults who had to have exposure to VR services or employment experience, and were evaluated beyond traditional employment outcomes (i.e., job status and wages) with assessment tools or systematic procedures. The research question is the following: for adults with NDDs, what employment outcome measures are utilized in the peer reviewed literature?
Search strategy
The search strategy involved three domains: population (i.e., people with neurodevelopmental disabilities), exposure (i.e., employment/vocational programs), and outcome (i.e., employment-related evaluation). To be representative of individuals across all disabilities, population terms were derived from NDD understood as disabilities acquired or manifest during youth, as defined by the Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5) and supplemented with information from the ICF-10 (APA, 2013). There were no limits for date of publication; however, non-English studies were excluded. The final list of search terms covered individualized terms for each database for population (40), exposure (58) and outcomes (16).
One reviewer (HV) conducted the searches in five international databases – Education Resources Information Center (ERIC), MEDLINE, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Health and Psychosocial Instruments (HaPI), and PsycINFO – for all publications up to March 2015.
Eligibility criteria and screening
Prior to screening, eligibility criteria were developed as in Table 1. Using these predetermined criteria, two reviewers (HV and RP) conducted independent screening in two stages. First, after duplicate removal, title and abstract screening was conducted for all records. Second, with the refined number of selected articles, full texts were screened and presented for data extraction using the same pre-determined eligibility criteria as outlined in Table 1. At the end of each screening stage- title and abstract, and full text screening- reviewers submitted unresolved disagreements for adjudication by the senior investigator (BDR).
Eligibility and screening criteria for journal article selection process
Eligibility and screening criteria for journal article selection process
In Fig. 1, an outline of the stages of our search, screening, and selection process are presented. For all journal articles in the included article pool, and using a standardized form, one reviewer (HV) extracted all data according to the following criteria: authors, year of publication, country where research was conducted, target population characteristics (i.e., age, gender, type of NDD diagnosis, and employment or VR exposure), and study characteristics (i.e., research study type, comparison groups included, and rater/assessor groups included).

PRISMA flow diagram of journal article selection process. Source: Adapted from PRISMA (Moher et al., 2009) flow diagram.
Both reviewers (HV and RP) conducted data extraction for the measures and procedures utilized to collect employment outcomes. For those measures that were not fully described within the selected journal article, the original source document was searched for a detailed description and a copy of the tool, if publically available on the world-wide-web.
In Table 2, a list of included articles is provided with extracted data on target populations, as well as the employment measures. For each measure, a brief description about major domains and items are provided. One reviewer (HV) scanned tool domains and individual items from each measure to code into work themes describing the essence of work being measured. The wording utilized for each theme was derived based measurement constructs described by the domains and items of each measure. Validity of the coding themes was confirmed through ongoing consultations between this reviewer and with the senior investigator (BDR). Common terminology was utilized where possible to improve the comparability of measures.
Included journal articles and data extraction of employment outcome measures (with brief description, as well as person and environment themes)
Within the first column, the square brackets contain the reference number for each included journal article and the subsequent lower case lettering is used when more than one measure is identified in a single article.
Literature search and selection process
The study selection process is represented by the PRISMA flow diagram in Fig. 1 (based on Moher et al., 2009). Through electronic searches of the final five databases resulted in a total of 2,151 records reviewed – ERIC (n = 790 records), MEDLINE (n = 665), CINAHL (n = 430), HaPI (n = 222), and PsycINFO (n = 44). After title and abstract screening against the predetermined eligibility criteria, and removing duplicate records, 167 articles underwent full-text review. In this process, only 17% of the identified papers underwent adjudication. In total, 45 journal articles met the inclusion criteria.
Included articles and population characteristics
Global study distribution
Table 2 provides descriptive information extracted from each of the 45 included articles. Articles were published within a 45-year span, from 1971 through to 2015. Rather than a global perspective, we have no papers retrieved from outside of developed countries. Of the 45 articles, 29 involved target populations recruited from within the USA (e.g., Bullis & Foss, 1986; Burnham & Housley, 1992; and Wehman et al., 2014). The remaining studies were conducted across the UK (n = 3 articles) (Howlin et al., 2005; Rose et al., 2010; Ruggeri-Stevens & Goodwin, 2007) [16, 34, 36], Australia (n = 3) [9,17,18], New Zealand (n = 2) (Fillary & Pernice, 2005, 2006), Hong Kong (n = 2) (Li & Tang, 2002; Tse, 1994), Israel (n = 2) (Gal et al., 2013; Gal et al., 2015), Spain (n = 1) (Martorell et al., 2008), Taiwan (n = 1) (Chen et al., 2012), and Canada (n = 1) (Cobigo et al., 2009). One study recruited participants from across the USA and Canada (Wehmeyer, 1994).
Population representation
The majority of studies included target populations of adults with intellectual disability (ID) (n = 16 articles) (e.g., Chen et al., 2012; Cimera et al., 2010, 2011; West & Patton, 2014), or mental retardation (MR) (n = 15 articles) (e.g., Bullis & Foss, 1986; Burham & Housley, 1992; Gaylord-Ross et al., 1995). The next largest target population included adults with autism spectrum disorder (ASD) (n = 14 articles) (e.g., Howlin et al., 2005; Parr & Hunter 2014; Taylor & Selzer, 2012). Persons with learning disabilities (LD) were included in five articles (e.g., Dipleolu, 2007; Rose et al., 2010; Rugerri-Stevens & Goodwin, 2007), persons with cerebral palsy (CP) in three articles (Novak & Rogan, 2010; Parent et al., 2013; West & Patton, 2014), and persons with attention deficit hyperactivity disorder (ADHD) in only one article (Painter et al., 2008).
Across the lifespan
At least five studies included a small sample of adolescents in their study populations: with a few participants as young as 13 years of age for one study (Malgady & Barcher, 1982), but with a minimum age of 17 years for the other four studies (Bullis & Foss, 1986; Farris & Stancliffe, 2001; Novak & Rogan, 2010; Rose et al., 2010). The highest age range for included participants was at the upper range of 69 years in only one study (O’Sullivan & Strauser, 2015).
Types of measures identified
As indicated in Fig. 1, 64 measures were identified across the 45 articles. In Table 2, we list each article with the target population according to NDDs, as well as all extracted tools along with a brief description. Of these measures, 22 are interview-administered questionnaires/scales, 19 are self-report surveys/measures (several of which can be administered verbally, if needed), 11 are rater-reported assessments, six are direct observation tests, five are rater-administered assessments, and four are longitudinal national database surveys.
There was little overlap in the use of similar employment instruments across the included studies. The following assessments were used in two studies with independent samples: the combined use of the Career Thoughts Inventory, the Mature Career Beliefs Inventory, and My Vocational Situation instrument were reported in two studies (Dipeolu, 2007; Painter et al., 2008); the Co-Worker Involvement Index (Rusch et al., 1995; Farris & Stanacliffe, 2011); the Nowiki-Strickland Locus of Control Scale (Jiranek & Kirby, 1990; Wehmeyer, 1994); and versions of the Inventory for Client and Agency Planning (ICAP) (Martonell et al., 2008; West & Patton, 2014).
Person and environment themes
As described in Table 2, there was a wide range of measurement constructs represented by the identified measures that could be demarcated as person (i.e., characteristics of the employee) and environment themes (i.e., physical, social institutional or cultural aspects of the workplace) (WHO-ICF, 2011). Person-focused themes included, work readiness, personality, motivation, adaptation, skills, performance, engagement, preferences, and satisfaction. Examples of environment-focused themes included, organizational commitment, social inclusion, social integration, workplace support, workplace culture, social acceptance and coworker involvement.
Discussion
This review sought only those studies that used standardized measures and clearly described procedures for examining employment outcomes beyond common metrics of job status (i.e., employed or unemployed) and compensation (i.e., wages and hours). While such outcomes are important to monitor, they are common in the research literature. For this review, the full array of instruments assessing employment outcomes in VR training and research amounted to 64 employment outcome measures. From this review, vocational researchers working with adults across various NDDs (including ID, MR, LD, ASD, CP, and ADHD) have developed and/or utilized measures that evaluate a variety of employment outcomes. There was little-to-no overlap in use of these instruments across the included studies. In this review we found an over representation of certain NDDs over others. The largest target population included persons with ID, and this was followed by persons with ASD. It is unclear if these differences are related to diagnostic prevalence, client services availability, publication bias, or even stakeholder group advocacy, but these differences are worthy of further investigation (Bishop, 2010).
The compiled list of outcome measures provided in this paper could be used as a starting point or repository of measures to trial in future studies. The measures identified represented various constructs that could be relevant to be tested across different disability groups, and age cohorts. How these tools are used (i.e., job matching, evaluating outcomes, or predicting employment success) and in what combination will be important to examine in future work. Future reporting on outcome measures could also be improved by researchers to better describe the work context or industry where the measures are best suited, which could complement the population profile.
The WHO has identified the critical importance of both “individual” and “environmental” factors for functioning in the ICF model (WHO, 2011b) and in the report on disability (WHO, 2011). By examining measures specifically for person (i.e., employee) and environment (i.e., workplace) constructs, the identification of measures for each construct highlight the diversity of employment outcomes researchers can assess beyond typical metrics. A variety of person and environment themes covered by the measures identified in the pool of included articles. The six most common person themes include job motivation, preference, personality, readiness, skills, and satisfaction. The three recurrent themes for environment measures include workplace culture, social integration, and supports. These are overlapping themes, and most measures cover more than a single theme.
Fewer studies were focused on the environment, especially in relation to the population of people with ASD. A recent systematic review by Ellenkamp et al. (2016) highlighted the importance of several work-environment related factors for people with Intellectual Disabilities. This could be further explored in ASD to determine the other work environment factors that could be relevant to this population. Using this repository of measures to test new constructs in ASD would be a good starting point.
This work represents the first time that employment outcome measures are reviewed from the literature across NDDs. It is important to look at measures across NDDs, because of the common employment challenges that exist in this population across various diagnoses. Rigorous methods were utilized in this review, such as throughout the selection process of relevant studies with two reviewers. Stratifying identified measures that were person versus environment focused was another strength of this paper in describing unique properties of the measures that are important for future researchers to consider.
This review had a few limitations, such as the inclusion of adolescents within some of the study samples, when the search criterion specified adult-focused population. However, given that the majority of participants were 17 or older across studies, the inclusion of a few adolescents that were mixed with an adult population within the employment study did not warrant the removal of these studies from our final pool. A second limitation was the lack of a quality assessment of studies that were included in this review. Although this would have enhanced the descriptive content included in the study table, the objective of this review was to identify all outcome measures utilized within the literature and be rather conservative about how employment outcomes are evaluated in the NDD literature. A third limitation was the challenge in identifying measures and their constructs within the identified papers in this review. Due to the lack of standard reporting methods, the identification of the measures and consistent themes was difficult to do objectively, thus relying on ongoing discussions within our investigative team to reach consensus.
Conclusion
With the growing number of people with NDDs living well into adulthood, we must target our efforts to mitigating the barriers to fair participation in work and employment (WHO, 2011). Through this review, a list was compiled of the wide array of employment outcome instruments available for research involving persons with disabilities. Many of these person- and environment-focused tools should prove useful in both job matching and training practices, as well as in follow-up evaluation and monitoring efforts to encourage best outcomes. When designing future studies, disability researchers should plan to report systematically about their study participant characteristic and clearly identify the standardized measures or structured procedures used for assessing employment outcomes. Evidence informed data are essential to inform good clinical practices and policies, and VR professionals are poised to play an important role in enabling good vocational outcomes. This repository can serve as a good starting point for researchers to utilize similar tools for better surveillance, as well as sharing outcomes of employment initiatives to begin to identify best practices to best increase the employability of people with disabilities.
Conflict of interest
The authors have no conflict of interest to report.
