Synthesis of the Research Literature on Job Accommodations for People with Psychiatric Disabilities
Disability Research Right to Know
| Boston University
Synthesis of the Research Literature on Job Accommodations for People with Psychiatric Disabilities
Conducted by the University of Maryland for the Center for Psychiatric Rehabilitation at Boston University
Table of Contents
Funded in part under a grant from the National Institute on Disability and Rehabilitation Research and the Center for Mental Health Services, Substance Abuse and Mental Health Services Administration, Grant # H133B090014.
Plain language summary
Job accommodations are those adjustments made by employers to a job or to the work environment that are mandated under the Americans with Disabilities Act of 1990 and the Rehabilitation Act of 1973.
They are designed to level the playing field so that people with disabilities can enjoy equal opportunities in employment and become contributing members of the workforce. Most of the literature on job accommodations in the 1980s and 1990s focused on people with a range of disabilities. Very little was written about accommodations for people with psychiatric disabilities.
Results of this synthesis of the research literature on job accommodations for people with psychiatric disabilities suggest that there are very few well-controlled studies of the effectiveness of job accommodations for people with psychiatric disabilities. There are, however, numerous studies that are descriptive or q use l describe the natu identify factors related to disclosure of psychiatric disabilities and job accommodations, or describe the process of disclosing and requesting accommodations.
Given how few well-controlled studies have been done in this area, the question of how effective job accommodations are for people with psychiatric disabilities remains open.
There is currently no conclusive evidence from well-controlled studies that job accommodations result in improved employment outcomes. At the same time it is important to note that neither are there well-controlled or rigorous studies indicating that job accommodations are ineffective. There is weak evidence from uncontrolled studies that job accommodations are positively associated with employment outcomes such as staying employed, job satisfaction, satisfactory job terminations, sense of mastery and well-being, and opportunity for promotion.
Promising findings from descriptive and qualitative studies suggest that most accommodations for people with psychiatric disabilities require little to nothing in direct costs, and most frequently involve flexible schedules, interpersonal supports (e.g., job assistance by vocational rehabilitation service providers, interpersonal or other support interactions provided by supervisors and coworkers), and changes in job tasks or the training process. The functional limitations requiring accommodation are most often cognitive or interpersonal in nature. Results also suggest that disclosure of psychiatric disabilities to employers, while necessary in accessing job accommodations, is a complex process for which individuals need guidance in order to weigh both the risks and the benefits of disclosing and to make decisions about what to say, when, and to whom. This issue is less complicated when employees with psychiatric disabilities are involved in supported employment (SE) services, since SE service providers often are actively involved in the disclosure and arranging of accommodations.
Contributors
Job Accommodations Study Group:
Kim MacDonald-Wilson,
Chia Huei Lin,
Marianne Farkas
Lead Reviewer:
Kim L. MacDonald-Wilson, Sc.D., CRC, CPRP
kmacdona@umd.edu
Assistant Professor, Rehabilitation Counseling
Counseling and Personnel Services Department
University of Maryland
3214 Benjamin Building
College Park, MD 20742
Additional Reviewers:
Chia Huei Lin, M.S., Doctoral Student, Research Assistant
University of Maryland
Introduction
Rationale for the Review.
Accommodations provided in the workplace are considered one of the essential components of the Americans with Disabilities Act of 1990 and the Rehabilitation Act of 1973, which level the playing field for people with disabilities and enable them to enjoy equal opportunities for employment. Mandated by law, employers are required to provide ‘reasonable accommodations’ to qualified individuals with disabilities unless doing so would create an undue hardship for the employer in terms of unreasonably high costs or significant disruption to the nature of the business. This protection under the law allows people with disabilities to perform the essential functions of the job, and to become contributing members of the workforce.
However, a limited amount is known about how the employer-provided supports such as job accommodations impact employment outcomes of people with psychiatric disabilities. While some empirical evidence exists that accommodations are associated with higher retention rates and satisfactory work outcomes for people with other disabilities (e.g., Campolieti, 2004; Fesko, 2001; Martin, Brooks, Ortiz, & Veniegas, 2003; Rumrill, Roessler, Battersby-Longden, & Schuyler, 1998; Schartz, Hendricks, & Blanck, 2006), none of these studies focused on people with psychiatric disabilities. There have been several theoretical models proposed and comprehensive literature reviews conducted regarding reasonable accommodations in the workplace for people with disabilities (Baldridge & Veiga, 2001; Cleveland, Barnes-Farrell, & Ratz, 1997; Colella, 2001; Florey & Harrison, 2000), but none of these literature reviews employed systematic review methodology. In fact, in 2006, an editorial in a special issue of the journal Work on accommodations in the workplace concluded that at that time, the existent knowledge base was insufficient to identify evidence-based practices in job accommodations for people with disabilities (Sanford & Milchus, 2006). What we do know is that people with psychiatric disabilities and employers have limited knowledge or understanding of the application of the Americans with Disabilities Act (ADA, 1990) regarding job accommodations for people with mental health conditions (Banks, Novak, Mank, & Grossi, 2007; Gioia & Brekke, 2003a; Granger, 2000). Employees or applicants with psychiatric disabilities are often hesitant to disclose disability to employers due to fears of discrimination, which may also interfere with requesting and using the accommodations that may allow them to enjoy employment opportunities often denied them (MacDonald-Wilson, Russinova, Rogers, et al., 2011).
Objectives of the Review
The primary goal of this review is to systematically analyze the job accommodation literature for individuals with psychiatric disabilities, not limited to randomized clinical trials (RCTs), using a grading
scheme for rating the rigor and meaning of disability research (Farkas, Rogers, & Anthony, 2008). The underlying assumption for this review is our belief that there is important and significant research on job accommodations that has accumulated which urgently needs to be synthesized for the mental health and rehabilitation field at large. In addition, the year of this review marks the 20th anniversary of the signing of the Americans with Disabilities Act of 1990, which authorized the implementation of reasonable accommodations for qualified people with disabilities in a wide range of employment settings under Title I of the Act.
Consumers, families, employers, advocates, and service providers are searching for useful information in this critical area. More than a decade has passed since the comprehensive literature reviews on accommodations for people with other disabilities with significant advancement in the literature on outcomes of job accommodations across disability groups. It is essential to validate the research findings specific to people with psychiatric disabilities and provide the innovative technical assistance necessary to ensure the use of these research findings among employers, advocates, and people with psychiatric disabilities.
Methods and procedures
Literature Search
The research team for the review was guided by the definition of “reasonable accommodations” specified in the Americans with Disabilities Act (ADA) of 1990 to be any adjustments that allow people with disabilities to enjoy equal employment opportunities (US Equal Employment Opportunity Commission [EEOC], 2000) as long as the required modifications do not result in “undue hardship for the employer” (ADA, 1990; Section 101). Thus job accommodations are “any change in the work environment or in the way things are customarily done (EEOC, 2002).” Job accommodations typically include modifications to the job application process, the work environment or job duties, or to the benefits and privileges of employment. Accommodations are intended to remove the workplace barriers for individuals with disabilities (EEOC, 2002).
The goal of the review was to analyze the evidence on accommodations for people with psychiatric disabilities in the workplace. Two research questions were specified to conduct this
What is the evidence on the outcomes of receiving job accommodations for people with psychiatric disabilities?
What is the nature and process of job accommodations for people with psychiatric disabilities?
We conducted a comprehensive search of the published and the unpublished ‘grey’ literature related to the questions above using the following search terms paired with the terms psychiatric disability, mental illness, or psychiatric illness:
Reasonable accommodations
Job Accommodations
Disability Disclosure
Accommodations (in searching through the business literature, this term generated more meaningful results)
Work or Workplace Supports
The terms ‘work supports’ or ‘workplace supports’ have been used in more recent employment literature for people with disabilities to refer to reasonable accommodations or job accommodations, among other types of supports, that people with disabilities use to be successful on the job. While these terms generally encompass any resources or supports that may be provided by practitioners, families, peers, or systems in addition to those provided by employers, we included these terms in our search since such publications may also refer to employer-provided supports such as job accommodations. We hoped, by including these additional terms, to maximize the chances of locating relevant literature on job accommodations for people with psychiatric disabilities.
The search of the published literature was conducted using a number of electronic databases, including Pubmed, PsychInfo, Google Scholar, and Business Source Complete (an upgraded version of Business Source Premier, replacing ABI/Inform). Further, we also accessed the database of cross-disability reasonable accommodations literature developed under a subcontract to a NIDRR funded project, the Mid-Atlantic ADA Center (Region III Disability and Business Technical Assistance Center) (Grant # H133A060085).
In addition to published literature, hand searches of relevant rehabilitation journals and citations contained in each located publication were also examined to identify additional relevant studies.
Grey literature, including dissertations, government reports, and other unpublished studies were located through relevant websites of rehabilitation and employment-related research, training, and technical assistance centers, including such sources as:
The Job Accommodation Network ()
The Virginia Commonwealth University RRTC on Workplace Supports and Job Retention ()
Boston University Center for Psychiatric Rehabilitation ()
National Research and Training Center (NRTC) at University of Illinois in Chicago ()
The National Institute on Disability and Rehabilitation Research (NIDRR) ()
The Institute for Community Inclusion ()
Dissertation Abstracts International
A research assistant was responsible for querying the databases, locating the documents, and conducting hand searches of the relevant journals. The Senior Researcher conducted a search of the grey literature, consulted her project accommodation database, and consulted with other experts about potential sources of other studies.
These search activities resulted in 100 documents for potential inclusion. An initial screening of document titles and abstracts was conducted for relevance to the topic of job accommodations for people with psychiatric disabilities – each document had to include job accommodations and refer to people with psychiatric disabilities, resulting in 60 documents for further review. After initial screening, previously developed inclusion and exclusion criteria were used to determine which studies to include in the systematic review, using a screening form. For consideration for inclusion in the review, the document had to describe an empirical study, with the sample consisting of exclusively people with psychiatric disabilities or a significant proportion of the sample consisting of people with psychiatric disabilities with analysis of results focusing specifically on people with psychiatric disabilities. Acceptable research designs included: experimental, quasi-experimental, correlational, pre-post evaluation, observational cohort, and survey research. We excluded qualitative studies, process evaluations, legal guidance and case law reports related to the topic, or descriptive, educational, or policy documents that did not analyze data collected about people with psychiatric disabilities. We also excluded studies that primarily focused on variables and analyses unrelated to workplace accommodations for people with psychiatric disabilities. Many of these excluded documents were related to discrimination in the hiring or employment process, or to people with disabilities other than psychiatric disabilities, or to non-workplace accommodations (e.g., academic or residential accommodations). Our initial screening for the systematic review yielded 15 quantitative studies, which included four outcome-related studies and 11 process-related studies.
Ratings of the Quality of the Research
The review was conducted using a system developed at the Center for Psychiatric Rehabilitation at Boston University for rating the quality of disability research, especially those that are non-experimental in nature (e.g., evaluation, policy, survey, pre-post designs, correlational). An instrument developed for rating research quality was used to rate the rigor of the outcomes-related research studies we located (Rogers, Farkas, Anthony, & Kash, 2008). The quality standards rating scale for each of the items was based on a 4-point scale, with 4=Definitely (Above Adequate), 3= Somewhat (Adequate), 2=Minimally (Minimally Adequate), and 1=Not at all (Not at All Adequate).
The four outcome-related studies were all correlational no experimental, quasi-experimental, or randomized controlled trial (RCT) studies, considered more rigorous in design, were evident. Rigor ratings of the four outcome studies yielded overall ratings of methodology that were slightly less than adequate, with other components of the studies rated as minimally or not adequate. Specifically, analysis of the rigor ratings of the four studies indicates that the average scores of the introduction and rationale for the studies were 3.5 out of 4, which means that the authors were able to sufficiently describe the current related literature in the field and build the rationale for their studies. All four articles were rated for the primary methodology item (“Study/research uses rigorous or sound research methods that allow the questions of interest to be addressed”). With 3.0 on this item as cutoff, the four articles were qualified to be included in the review. The average rating of all rigor ratings in the methodology and procedures sections was 3.4, which is above Somewhat Adequate. In particular, the methodology items on the rationale of the design and the definitions of variables to be measured were often missing or inadequate. The lowest rigor ratings were of the discussion and conclusion sections. The average rating was 2.4, minimally adequate. Many authors did not sufficiently describe the limitations of their studies (including threats to validity) or provide alternative explanations of their results.
Within the grading process developed by the Center for Psychiatric Rehabilitation, the threshold for the possibility of a systematic review of the literature is five outcome studies. Although we were able to conduct rigor ratings for the four outcomes-related articles, the small number of articles available is insufficient to draw valid conclusions about the impact of accommodations on employment outcomes for people with psychiatric disabilities. Another issue is that none of these studies used experimental or quasi-experimental designs, which also limited the level of evidence needed to support the validity of their findings. The sample size of these studies was small: three out of the four studies had sample sizes below 70.
Since the studies fell below the minimum in terms of overall rigor, ratings of meaningfulness were not conducted on these four studies.
Given that few high quality studies exist on the effectiveness of job accommodations for people with psychiatric disabilities, the researchers proceeded to focus on a structured synthesis of the findings of the research literature to provide a comprehensive picture of what is known about job accommodations for people with psychiatric disabilities. In addition to the 15 quantitative studies, the researchers identified nine qualitative studies that described the nature or process of requesting or accessing accommodations in the workplace. Therefore, this synthesis of the research literature includes outcomes-related and process-related quantitative studies and qualitative studies of job accommodations in order to address this important research area in which stakeholders are seeking guidance.
Time Period of the Studies and Research Covered
The Americans with Disabilities Act (ADA) was signed into law in 1990. The year 1990 was used as the starting point for the review of the literature because Title I of the ADA specifically focused on the issue of accommodations. Under Title I, employers were required to provide reasonable accommodations to qualified individuals with disabilities as long as the accommodation did not cause undue hardship on the employer. While reasonable accommodation was first mandated with the Rehabilitation Act of 1973, it covered few employers, primarily federal government employees or employers receiving federal dollars. With the signing of the ADA, the majority of employers in the United States were required to follow the law, as long as the employer had a minimum of 15 employees (as of 1994). Most of the research on accommodations for people with psychiatric disabilities was conducted after 1990. Therefore, the search was conducted on all research articles in the 20 years between 1990 and 2010.
Training of Reviewers
Two raters were used for the synthesis. One rater was an assistant professor in Rehabilitation Counseling in the Department of Counseling and Personnel Services at the University of Maryland. The other was a senior doctoral student in Counselor Education – Rehabilitation Counseling specialization at the University of Maryland. Both raters had participated in a project at UMd on a comprehensive review of the literature on job accommodations for people with disabilities and were knowledgeable about the literature base. Each rater was also knowledgeable about research methods.
Training of raters was conducted by experienced researchers affiliated with the Center for Psychiatric Rehabilitation at Boston University who had developed the rating scales and used them to conduct other systematic reviews. The training focused on rating the quality of the research and identifying acceptable indicators of quality for each of the 4 points on the rigor and meaning rating scales. Each item in the scales was reviewed and the meaning of the item and the evidence that could be considered for each indicator was discussed in detail. Sample research articles were used to test inter-rater reliability with each rater independently rating each article, and then comparing their ratings until agreement was achieved. Formal tests of inter-rater reliability yielded an 80% agreement between raters.
Results and conclusions
Background
In the process of searching for studies to include in this review, we found a number of conceptual and educational publications that provided useful background information, laying the context for the current study. What follows is a description of the history of job accommodations.
History of Job Accommodations for People with Psychiatric Disabilities.
Job accommodations in the workplace were first mandated with the implementation of the Rehabilitation Act of 1973. This was the first law to prohibit discrimination against people with disabilities and required employers who were federal agencies, federal contractors, or organizations receiving federal dollars to provide “reasonable accommodations” to individuals with disabilities so that they could enjoy equal opportunities for employment, education, and access to public services.
During the 1970’s and 1980’s, since there was limited experience with accommodating people with disabilities, the Handbook of Reasonable Accommodations (US Office of Personnel Management, 1984) was developed by the federal government to provide information about reasonable accommodations and how to implement them.
In an early study, funded by the US Department of Labor (Berkeley Planning Associates, 1982), accommodations were found to be relatively inexpensive, provided by large and small employers, to people identified as having primarily physical or sensory disabilities. Accommodations were reported to be effective in improving the productivity of workers provided them, as well as benefiting other employees in the organization.
In 1984, the Job Accommodation Network (JAN) was implemented at West Virginia University as a service of the President’s Committee on the Employment of People with Disabilities (Rochlin, 1985). By the early 1990s, information on accommodations for people with psychiatric disabilities was available at JAN. JAN is a national, confidential, and free expert technical assistance and consultation service on workplace accommodations and disability employment issues (JAN, undated). Today, JAN is funded as a service of the US Department of Labor, Office of Disability Employment Policy (ODEP) available through the internet (askJAN.org) and by phone with a toll-free number (1-800-526-7234).
In 1990, the Americans with Disabilities Act (ADA) was signed into law (ADA, 1990). Under Title I of this law, all employers with 15 or more employees (as of July, 1994) are required to provide “reasonable accommodations” to qualified individuals with disabilities unless doing so would impose an undue hardship to the employer. The U.S. Equal Employment Opportunity Commission (EEOC), which was responsible for handling discrimination complaints under Title I of the ADA, developed guidance materials about how the ADA applies to people with psychiatric disabilities (EEOC, 1997) and on reasonable accommodations (EEOC, 2002). Since the implementation of the ADA in 1990 and with the development and implementation in the 1980s and 90s of supported employment services nationwide, there has been a noticeable increase in the research on job accommodations for people with disabilities, including a focus on people with psychiatric disabilities.
In summary, while job accommodations have been in use for more than 35 years since the implementation of the Rehabilitation Act of 1973, most of what was known applied to people with physical and sensory disabilities, and very little was known about accommodations for people with psychiatric disabilities until the implementation of the Americans with Disabilities Act of 1990. Early educational and conceptual publications described the types of job accommodations that would meet the needs of people with psychiatric disabilities.
Few rigorous studies on job accommodations have been conducted, especially with a focus on people with psychiatric disabilities. Indeed, we found no experimental or quasi-experimental designs related to the effectiveness of accommodations, a point that was made by Sanford and Milchus (2006) in their editorial on the difficulty in defining evidence-based practices in accommodations in the workplace. In conclusion, research is challenged by the rigor of the experimental design in investigating the effectiveness of job accommodations, when it would be unethical to withhold a legally sanctioned right to accommodation under the ADA.
This is the context for a synthesis of the literature on job accommodations for people with psychiatric disabilities.
Measurement of Job Accommodation Outcomes.
Finally, the job accommodations literature dealing with people with a variety of disabilities use a number of different variables related to process or outcome findings of the provision of accommodations that may be examined. The typical process measures noted in studies of job accommodations involve attitudes toward accommodations, receipt of requested accommodations, number and types of job accommodations provided, and costs of accommodations. Typical outcomes-related measures include satisfaction with accommodations provided, improvement in job performance, maintaining the job, advancing to another position through promotion, return to work from a disability leave, benefits to the employer or organization, and benefits to other workers in the job environment. Not all of these outcome variables may be considered measures of effectiveness of the intervention involving providing accommodations, which involves primarily the impact of the accommodation on the employee’s satisfaction with the job, job performance, or job retention.
Results of the Synthesis of the Job Accommodations Literature – Outcomes Studies
Before discussing outcomes or effectiveness studies, it is important to note that the number of studies that focus on the impact or outcomes of accommodations is limited. We found no experimental or quasi-experimental studies on the effectiveness of job accommodations for people with psychiatric disabilities.
Random assignment is the key design factor that defines an experimental study.
The lack of experimental studies may be due to the fact that the legal right to accommodations makes it impossible to randomly assign the provision of accommodations
to an experimental group and withhold it from a control group. We found four studies that examined relationships between accommodations and outcomes such as return to work, job tenure, and job terminations. These were primarily correlational studies or follow-up analyses of data in which an experimental design was used to examine a supported employment intervention. However, correlational studies only permit a description of the associations between these variables, and do not permit the inference of cause and effect relationships between accommodations and employment outcomes.
Outcomes of Accommodations – Correlational Studies.
An early correlational study exploring the effectiveness of job accommodations among individuals diagnosed with mental health conditions participating in supported employment, was conducted by Fabian and colleagues using retrospective records review and staff interviews. The study investigated the types and frequency of job accommodations and work outcomes (Fabian, Waterworth, & Ripke, 1993). The authors identified a total of 231 job modifications for the 47 jobs of the 30 participants, for an average of 5.1 accommodations per job. The most frequently identified accommodation was orientation and training of supervisors to provide necessary assistance (38.1%), followed by modifications of the non-physical work environment (16.4%), and modifications of work hours and schedules (15.6%). Using survival analysis techniques, the authors examined the relationship between job duration and number of accommodations. With the average number of accommodations (5) as the cutoff point, the authors divided the sample into two groups: accommodations of less than five (n = 19) and accommodations of five or greater (n =27). They found that five or more accommodations on a job resulted in twice the average job tenure, which indicated that job tenure was significantly associated with number of job accommodations. The major limitations of this study were that the sample size was small (n=30), and the data was collected retrospectively from a single supported employment program in a mid-Atlantic state, which limited the ability to generalize the results to other programs, or to people with psychiatric disabilities who are not involved in supported employment. In addition, the study design only indicates an association between the number of accommodations and job tenure, and cannot be used to show a causal relationship between these variables. Nonetheless, this study was important as the first study to show a relationship between accommodations and employment outcomes for people with psychiatric disabilities.
Akabas and Gates (2000) investigated the relationship between provision of job accommodations and employment outcomes among New York City workers with a diagnosis of a mental health condition. This study consisted of two phases: during the first phase, 181 insurance claimants who received reimbursement for criterion psychotropic medications for at least eleven continuous months, (89% who were employed full-time), were randomly selected to be interviewed face-to-face to determine factors related to sustaining employment. During the second phase, 43 individuals out of work and on short term disability leave with a primary or secondary psychiatric diagnoses were interviewed regarding factors needed to assist in the return to work.
The results indicated that two-thirds of Phase I participants experienced “serious and persistent” mental illness (categorized by long-term use of psychotropic medications, hospitalization history, and use of sick and disability leave in the past due to psychiatric reasons), although 79% rated themselves as slightly or not at all disabled. Regarding job accommodations, 61% of the Phase I workers reported receiving at least one accommodation, while 60% of Phase II participants on disability leave reported receiving no job accommodations prior to leave.
Another important outcome variable of this study is “accommodation adequacy,” which was created by combining both the concepts of condition stability (the extent to which functional limitations interfered with work) and workplace responsiveness (whether or not the worker was accommodated). About half of the Phase I participants (53%) who worked reported that they were inadequately accommodated. Inadequate accommodation was associated with lower job satisfaction, lower sense of mastery and well-being, and a lack of opportunity for promotion. Nearly all of the Phase II participants on disability leave (95%) reported that their conditions interfered with doing their jobs, which indicated that their accommodations were inadequate prior to leave.
Of all the types of accommodations provided to Phase 1 workers (task, routine, relationship), the need for relationship accommodations was often unmet. Seventy-six percent of the participants identified the need for relationship accommodations but did not receive it. The authors suggested from this and the finding that supervisory and coworker support was significantly related to adequate accommodation that “accommodation is a social process as much as a technical method (p. 177).” Thus, the authors concluded that work group supports were essential to developing effective accommodations for employees diagnosed with serious and persistent mental health conditions.
This study has several limitations that affect the strength of the conclusions about the effectiveness of accommodations in the workplace. First, the study design is correlational in nature, so that causal relationships between accommodations and employment outcomes cannot be reliably inferred. Accommodations were not independently manipulated. In addition, there is no control group identified, although there are comparisons between individuals who were employed and individuals who were on short term disability leave. The study is based on two samples from a population of individuals insured in one state, and it is not clear that results can be generalized to all workers with psychiatric disabilities. Evidence suggests that having adequate accommodation is associated with maintaining employment. However this conclusion is based on comparing the adequacy of accommodations given to workers who are maintaining employment to those who are out on disability leave. Variables other than accommodations could be contributing to the outcome.
Despite this limitation, overall the findings suggest that the adequacy of accommodations are related to maintaining employment, and deserve further investigation.
Becker, Drake, Bond, Xie, Dain, and Harrison (1998) investigated the job terminations of 63 individuals in supported employment programs in New Hampshire who experienced a job termination within 18 months of participation in a randomized controlled trial study of supported employment. In a correlational follow-up study using the Job Termination Interview (JTI) developed for this study, data was collected from clients, employment specialists, and case managers on reasons for job terminations and accommodations that were desired. The authors classified job terminations as either satisfactory (left for another job, laid off due to elimination of the job, or left because the job was time-limited) or unsatisfactory (fired or left without another job). Desired accommodations were identified by asking raters what accommodations would have improved the job.
Approximately half (52%) of the terminations were rated as “unsatisfactory.” The most common accommodation desired by individuals who experienced unsatisfactory job terminations was flexible hours (30%), followed by idiosyncratic items (27%) such as wanting more training, more support from the employer, more feedback, or different working conditions. It is not known how many of the unsatisfactory job termination situations (out of approximately 33 individuals) had accommodations identified that were believed to have improved the job, nor do we know how many types of accommodations were identified. What is reported is that the flexible hours and miscellaneous accommodations significantly differentiated the unsatisfactory from the satisfactory termination group, although the direction of the relationship is unclear. Evidence suggests these types of accommodations are somehow related to the type of job termination.
There are a number of limitations to this study related to the effectiveness of accommodations. First, the study focused on a relatively small and select sample of Caucasian individuals with psychiatric disabilities in supported employment programs in New Hampshire during the early 1990’s, limiting the generalizability of the results to those who are employed, including those not involved in supported employment programs.
In addition, this sample was a subgroup of individuals participating in a larger, randomized controlled trial of supported employment, where the employment intervention was independently manipulated but accommodation was not an independently manipulated variable. Detailed information is not provided on the accommodations data, limiting the usefulness of the findings. In addition, the data on reasons for job terminations and accommodations desired was gathered retrospectively, and therefore may be biased by recollections, although attempts were made to triangulate the data with multiple perspectives (about half of the data provided was rated from all 3 perspectives). No analysis was done however, regarding the influence of the perspectives on the data provided. Direct evidence on the effectiveness of accommodations on retaining the job is only suggested, indirectly, by asking raters about what would have improved the job.
In a similar study, Mak, Tsang, and Cheung (2006) examined job terminations among 60 individuals with mental health conditions in a supported employment program in Hong Kong. They translated and validated the Job Termination Interview (JTI; Becker et al., 1998) into Chinese. The results indicated that about half of the job terminations (53%) were unsatisfactory, which was similar to Becker and colleagues’ findings (52%). In terms of job accommodations desired to improve the job, 13 out of 60 participants reported that the accommodations that would have helped included changes in work schedule (30% or 4 individuals) and a reasonable and stable salary and better working environment (23% or 3 individuals). Other miscellaneous job changes were suggested, such as supplying spacious and air–conditioned environment, helping participants understand more about job demands, and providing sufficient supervision at work.
The limitations of this study are similar to those of the Becker et al. (1998) study. In addition, this study was conducted using a convenience sample of 60 individuals participating in supported employment in Hong Kong, and a small number of individuals (n=13) identified accommodations, further limiting generalizability of these results. Translation of the JTI interview instrument may also have affected the reliability and validity of the findings, particularly in reference to accommodations.
In contrast to the US definition of accommodations as changes to an actual job to support an individual with disability, some of the reported accommodations (e.g., finding a job that matched employee’s interests and understanding the job demands better) are not typically defined as accommodations. Because of the small sample of accommodations, analyses related to type of job termination could not be determined.
Results of Synthesis of Job Accommodations Literature – Process Studies
The majority of the research literature on job accommodations for people with psychiatric disabilities involves descriptions of accommodations and other information related to providing accommodations in the workplace. While these studies do not focus on the effectiveness or outcomes of accommodations, they do provide descriptive information about job accommodations for people with psychiatric disabilities. These studies focus on attitudes about job accommodations, the nature of accommodations, and the process of requesting and accessing accommodations. Studies on disclosure of psychiatric disabilities in the workplace are included when they relate to requesting and accessing accommodations. The following synthesizes the literature on the process related to job accommodations for people with psychiatric disabilities.
Attitudes about Job Accommodations.
There were three studies examining attitudes about job accommodations provided to people with psychiatric disabilities using survey design methods to analyze relationships between variables.
The first two studies surveyed businesses about policies and practices, although accommodations were not the primary focus of the studies. The third study is the most focused and most rigorous with a large sample randomly selected regarding accommodations and coworker perspectives about job accommodations.
In the first study, Jones, Gallagher, Kelley, and Massari (1991) conducted a national survey of Fortune 500 companies about their policies concerning people with psychiatric disabilities (they used the term “psychiatrically handicapped”). A total of 127 firms out of 488 included on the Fortune 500 list responded to the survey, resulting in a 26% response rate. While this study was published after the signing of the Americans with Disabilities Act, data was collected in the mid-1980’s. Only eight businesses (6.3%) provided “extra monitoring and/or support from supervisors” after hiring, although specific accommodations are not mentioned. While this survey did focus on employees with psychiatric disabilities, the information is very dated, and little detail is provided about policies related to specific accommodations, other that extra support or monitoring from supervisors. Fortune 500 employers tend to be large corporations with multiple offices and/or branches, with significant resources. Due to these factors, the generalizability of these results to the range of employers in the U.S. is limited.
Scheid (1999) conducted a telephone survey to investigate the response of employers to the Americans with Disabilities Act (ADA), with specific focus on the employment of people with psychiatric disabilities (they used the term ‘mental disabilities’). A total of 190 businesses in a southern metropolitan area were randomly selected for this survey and 117 businesses responded to the survey (a 62% response rate). While the majority of the survey focused on awareness of and compliance with the ADA, 61% of these companies reported providing specific accommodations to people with psychiatric disabilities, with the majority adjusting work hours, creating part time jobs, or restructuring jobs. Regarding the cost of the accommodations, 79.2% reported that these accommodations were not expensive at all. Only 16.7% found them somewhat expensive and 4.2% found them very expensive. While this survey involved a more rigorous survey design and indicates more awareness of and more positive attitudes about accommodations for people with psychiatric disabilities than the Gallagher et al. (1991) study, it is limited by the narrow geographic area sampled.
Peters and Brown (2009) conducted a survey to investigate the reactions of employees toward coworkers with mental health conditions and toward job accommodations. A sample of 600 employees was randomly selected from a unionized Canadian healthcare organization with approximately 6700 employees. A total of 305 employees completed the survey (51% response rate). Respondents rated each of seven accommodations regarding the extent to which they felt that the accommodation was appropriate for coworkers with mental health conditions using a 5-point scale (1 = Inappropriate to 5 = Appropriate). Job accommodations included were flexible work hours, providing a job coach, self-paced work, banking of overtime hours for use in case of illness, longer or more frequent breaks, time off for counseling, and increased supervision. Employees were less likely to view longer or more frequent work breaks as appropriate accommodations for coworkers compared to flexible hours, banking of overtime hours, and counseling. Overall, the extent of employee contact with individuals with mental health conditions had no significant effect on support for job accommodations. However, occupation type was related: employees working in occupations with limited contact with individuals with mental health conditions were more supportive of accommodations than those in occupations with daily contact. Authors interpret this finding that health care workers with more frequent contact may be in contact with people who are in more acute phases of the illness, resulting in more negative perceptions. Peters and Brown also found that employee perceptions of the likelihood that their employer would treat coworkers with mental health conditions fairly predicted the likelihood that they would self-disclose a mental health condition and seek treatment for it.
The study is limited to a sample of healthcare workers in Canada and may not be able to generalize to workers in other professions or in the U.S. Insufficient detail was provided on the types and severity of mental health conditions and contact with people with mental health conditions, particularly regarding whether the contact was with a coworker or service participant, which also makes conclusions about the relationship between positive contact and positive attitudes more complex. It is also difficult to draw any conclusions about the relationship between attitudes and behavior, although a relationship between attitudes and intention to act was found.
Nature of Accommodation.
As mentioned, much of the empirical literature on job accommodations for people with psychiatric disabilities is descriptive in nature. These studies examine the nature of accommodations such as number, type, and costs, and limitations of people with psychiatric disabilities that are associated with accommodations, as well as some information describing the process of accommodation – how accommodations are implemented. This section describes four studies focusing on the nature of accommodations: one is a secondary analysis of a large national health survey dataset that contains questions related to workplace accommodations, two studies are quasi-experimental using a time-series design, and one study is correlational in design. The quasi-experimental studies were in fact two different analyses from the same research project.
Zwerling, Whitten, Sprince, Davis, Wallace, Blanck, et al. (2003) reported a secondary analysis of data gathered from the National Health Interview Survey Disability Supplement , describing the nature of workplace accommodations in the American workforce and factors associated with the provision of such accommodations. Using a large, nationally representative sample of workers aged 18 to 69 years with a wide range of impairments and conditions (n=4937), 12% reported receiving workplace accommodations. The study found that people with mental health disabilities were 50% less likely to receive work accommodations compared to people with other types of disabilities. Specifically, those workers with diagnosed major depression or substance abuse disorder were significantly less likely to report accommodations.
While this study’s strengths are in its large, nationally representative sample, this study has several limitations. Types of accommodations were listed for respondents to identify whether they used them at work. Only three of those accommodations were likely to be associated with mental health conditions, so accommodations may be underreported. Information is not provided on accommodations or on employment characteristics by type of impairment or health condition, so it is not clear which accommodations were more often used by people with mental health conditions. Employer perspectives are not represented, as data is gathered by self-report of the participants.
In the first prospective empirical study of job accommodations for people with psychiatric disabilities, MacDonald-Wilson, Rogers, Massaro, Lyass, and Crean (2002) collected data about accommodations and employment on 191 individuals with psychiatric disabilities who were involved in supported employment programs in several eastern states. Researchers gathered data from employment specialists at the time of job entry and quarterly for up to one year (including through multiple jobs) using a quasi-experimental time series design.
The 191 participants in the study received a total of 322 accommodations during the year of their participation in the study. The jobs obtained by participants were primarily entry-level, unskilled jobs in the service and retail industries that participants maintained for an average of six months. The accommodations were primarily provided in one job, but 9% of the study participants had two or more jobs during the year of the study. The majority of accommodations (60%) involved the presence of a job coach, either on the job or in the hiring process. Flexible scheduling (21%) occurred frequently as well. Less common accommodations included changes in training and supervision and modified job duties. Only one of the 322 accommodations involved a direct expenditure of money amounting to $25 for a job performance test. However, employment specialists were also asked about indirect costs of accommodations, especially as they involved extra supervisor or coworker time. Over one third of the accommodations involved a reallocation of coworker or supervisor time, hours, or job duties and other indirect expenditures. Twelve percent of the accommodations required extra hours to be expended by coworkers, on average about nine hours per month. Extra supervisory hours were also reported to be involved in 28% of the accommodations with an average of five hours per month. Many of these extra time accommodations involved additional training early in the job and the supports were often reduced or eliminated as time passed.
In this study, MacDonald-Wilson et al. (2002) also reported the types of functional limitations that employment specialists identified as the basis of the need for the accommodations. The types of limitations of workers with psychiatric disabilities were primarily in interacting with others (23%), learning the job tasks (21%), and maintaining work stamina (11%). Other less frequently reported limitations were in managing symptoms or tolerating stress, working independently, attending work, and adjusting to changes in work.
Initial information about the process of accommodation was also identified in this study. Most of the time, accommodations were identified during the hiring process (63%) or once the employee was on the job (47%). Of those identified while the employee was on the job, 53% were identified during the first two months of employment. In terms of what was accommodated, accommodations were made to the hiring process 23% of the time (e.g., allowing the job coach to attend the interview, providing assistance with the application or testing process) and to the working on-the-job activities (e.g., additional training, extra supervision) 77% of the time.
While this study contains a large sample and uses a quasi-experimental design, it is limited by its focus on employees involved in supported employment programs in the Eastern U.S., limiting generalizability of results to people with psychiatric disabilities not involved with vocational services. In addition, accommodations were only reported from the service provider’s perspective, and not from employers’ or employees’ perspectives which may bias the types of accommodations reported.
In a further analysis of relationships of factors to accommodations, MacDonald-Wilson, Rogers, and Massaro (2003) conducted a multi-site, longitudinal study to investigate the relationships between functional limitations, accommodations, and demographic factors using a quasi-experimental time series design described previously. Their sample consisted of 191 workers in 22 supported employment programs from three eastern states. Data on accommodations and the functional limitations associated with the need for those accommodations were gathered prospectively from employment specialists in a structured, narrative form without preconceived categories.
Analysis indicated that the most frequent functional limitations of employed persons with psychiatric disabilities leading to the need for accommodation were cognitive limitations, followed by social, physical, and emotional/other. Accommodations were grouped into five categories: human assistance (of a job coach), flexible scheduling, extra or modified training and/or supervision, modified job duties, and other.
Demographic factors (gender, race, education level, diagnosis, substance abuse, or months of previous employment) were not associated with the number or type of reasonable accommodations. However, people diagnosed with psychotic disorders had significantly fewer cognitive limitations than people with other disorders. The strongest relationships were between limitations and accommodations: the number of functional limitations was associated with the number and type of accommodations received: the fewer limitations that were identified, the fewer number of accommodations were provided. In particular, employees with one limitation had significantly lower average total extra hours provided by coworkers and supervisors than did employees with more limitations. The types of functional limitations were also significantly related to the types of accommodations. Cognitive limitations were the strongest and most consistent predictor of accommodations provided when the accommodation type was the provision of extra supervisor and coworker hours. Individuals with functional limitations in the cognitive and social areas were more likely to receive accommodations involving human assistance (62% and 92%, respectively) while those with physical limitations were more likely to receive flexible scheduling (65%).
The limitations of this study are similar to those identified for the study by MacDonald-Wilson et al. (2002). In addition, the categorization of limitations and accommodations were coded by group consensus, and may not represent current understanding of the functional limitations that may be attributed to specific mental health conditions.
In a study with a similar focus, Symanski-Tondora (2003) conducted a dissertation using a correlational design to examine how individual characteristics (e.g., cognitive impairment, baseline work functioning, clinical symptomatology) and work site characteristics (i.e., site complexity) impact the need for reasonable accommodations among a sample of 87 participants with schizophrenia or schizoaffective disorder who were participating in a work rehabilitation study in the Veterans Administration investigating employment outcomes. Accommodations information was collected and then data from the larger study was accessed retrospectively to analyze relationships.
The most frequently used specific accommodations were allowing the worker to contact his or her therapist during the workday, accepting a longer learning period, creating an initial job match based on worker strengths and needs, establishing a flexible return to work schedule after hospitalization, and allowing the worker more time to learn job tasks. Using a factor analysis approach, the author developed an instrument to collect data on specific accommodations. The five categories of accommodations were a) supportive supervision and scheduling policies such as flexible leave and proactively b) use of on-site physical accommodations including physical space arrangements and allowing on- c) accommodations that support organization and task structure such as to-do lists, reminders, an d) interpersonal accommodations such as coworker support, educating employers, and contac e) modified expectation of task completion including extended learning periods and allowing for mistakes without penalty. These categories were then used for other analyses of the relationships between accommodations and individual and worksite characteristics.
Cognitive impairments, specific social domains of baseline work behavior, and site complexity all positively predicted the use of interpersonal accommodations. People with cognitive impairments were more likely than people with no cognitive impairments to use on-site physical accommodations such as job coaching although there were no differences between groups in the total number of accommodations.
People with lower baseline work behavior were more likely to use onsite physical accommodations and modified expectations of task completion.
Symanski-Tondora (2003) suggested that rehabilitation professionals should consider an individual’ s personal characteristics, such as cognitive impairments, social interpersonal skills, and baseline work behavior, as well as characteristics of the job site in identifying job accommodations. Limitations of this study however focus on the generalizability of the results to people other than veterans working in rehabilitation program jobs. Participants in this study were primarily Caucasian men who were veterans participating in a work rehabilitation program in the northeast. Characteristics of the jobs may also have differed significantly from competitive jobs in the community, which may have affected the types of accommodations provided in the workplace. Accommodations information was collected at one point in time, not over a period of time on the job.
Process of Requesting and Accessing Accommodations.
This section describes four studies focusing on the process of accommodations: all four studies are survey or correlational in design. Studies involving disclosure of psychiatric disabilities in the workplace related to requesting and accessing accommodations are included.
Granger, Baron, and Robinson (1997) conducted a national mail survey of employment support practitioners working in community-based agencies in 10 states that provided employment support services to people with psychiatric disabilities. The total sample is a non-probability convenience sample of 194 respondents geographically spread throughout the U.S., with a return rate of 35%. Primary analyses are descriptive in nature.
Job coaches and job developers report that the types of job accommodations used most frequently include use of the job coach in facilitating communications, phone access to the job coach while on the job, use of positive feedback by the supervisor, the option for part-time work hours, and gradual introduction of tasks while learning the job. The costs of job accommodations for people with psychiatric disabilities wer approximately 58% of the respondents reported no cost and 32% reported costs of less than $100. Costs were similar for initial accommodations provided on the job, as well as for accommodations maintained over time on the job.
In the process of accommodation, employment support practitioners describe themselves as very involved in facilitating communications between the employee and employer related to arranging job accommodations. Over one-third of the time, the employer was aware of the nature of the agency and types of disabilities of the individuals served or otherwise knew the individual had a mental health condition. One-fourth of the time the individual themselves disclosed to the employer, and one-fifth of the time, the individual chose not to disclose. In terms of the timing of disclosure of psychiatric disability to the employer, two-thirds of the time this occurs before the job offer (46% prior to interviews, 18% during interviews). In terms of how the accommodations are arranged, over half of the respondents indicated that they never or seldom referred to the ADA when arranging job accommodations. In addition, disclosure was a somewhat complex process, initially just revealing the fact of psychiatric disability, and later revealing more information regarding how the disability affected the job and what accommodations are needed.
While this study provided empirical evidence about the use of accommodations by people with psychiatric disabilities in supported employment programs from the perspective of employment support practitioners, the findings are limited due to the non-representative, convenience sample which makes it difficult to generalize findings nationally or to people not involved in such programs. Questions on accommodations involved pre-developed categories and specific accommodation checklists with space for practitioners to add accommodations not listed, which may have limited the range of accommodations reported. The perspectives of employers and employees with psychiatric disabilities are not included in the study, also limiting the usefulness of the findings to people not involved with service providers in the workplace.
Banks, Novak, Mank, and Grossi (2007) conducted a survey on the role of disclosing a psychiatric disability and the employment experiences of 162 people participating in ten supported employment programs across eight states (CA, IN, KS, NY, OR, PA, RI, VA) in the United States. Data for the study included demographic, disability, disclosure, employment, and support information. The disclosure study was a supplement to research on job performance and workplace supports.
They found that 82% of the employees in supported employment programs had their disability disclosed in the workplace, however disclosure was typically made by the employment agency. In general, employees had little role in disclosure once they chose to be involved in supported employment services. For those who had disability disclosed, details about the process of disclosure revealed that specifically 54% had the SE staff disclose (in 45% of the disclosures the employer was aware of the nature of the agency), and 29% of the time the employee themselves disclosed. In terms of who disclosed, women, people with a mood disorder, and people who displayed no symptoms at work were less likely to disclose. Those who did disclose most often did so during the job development and hiring process.
Most often the identified purpose of disclosure was to enlist the support of workplace personnel (80%) and negotiate accommodations (68%). Supported employees who disclosed their disability were significantly more likely to have an accommodation at the workplace than supported employees who did not disclose their disability. Fifty-two percent (n=133) of those who disclosed had workplace accommodations while only 31 percent (n= 29) of those who did not disclose had workplace accommodations. In particular, company personnel of employees who had disability disclosed were more likely to have training and were more likely to provide support compared to personnel of employees who had not disclosed. This study is one of the few studies that directly links disclosure to receiving accommodations and other employment outcomes. However, since it is a correlational study, we are unable to conclude that disclosure conclusively resulted in these outcomes. In addition, results may only be generalized to people involved in supported employment programs, and provides little insight into the role of individuals with psychiatric disabilities in the disclosure and accommodation process.
Ellison, Russinova, MacDonald-Wilson, and Lyass (2003) conducted a national survey with a large, non-representative, purposive sample of professionals and managers (n=495) who identified themselves as having psychiatric conditions.
Response rate was calculated at 66% using the proportion of respondents to total number of surveys sent out, less ineligible surveys. To be eligible for the study, participants had to be working in a professional or managerial position for at least 6 months during the previous five years, and have experienced a serious mental health condition. The majority of respondents were Caucasian, female, over age 40, and well educated, with the majority (73%) reporting a mood disorder diagnosis, and most never receiving federal disability benefits, although most did report at least one hospitalization (78%) and were taking medications to treat the mental health condition (83%). This sample is clearly a different group from individuals in other studies who primarily participating in vocational rehabilitation or supported employment services.
An unexpectedly large proportion (87%) of study participants reported having disclosed their mental health condition to someone in the workplace. Most frequently, respondents disclosed to supervisors. In terms of what was disclosed, the majority (64%) tended to describe their diagnosis or that they had a psychiatric disability or mental health condition, and 31% described modifications that they needed (e.g., accommodations) or problems in keeping the job. In addition, likelihood of disclosure was positively related to confidence in the job and “having learned how to manage their psychiatric condition and have a satisfying life.”
In terms of when disclosure happened, one third made their disability known when applying for the job, another 16 percent within one year of working on the job. Investigation into the patterns of disclosure revealed that about half of the disclosers reported unfavorable circumstances leading to disclosure while one third disclosed when they felt comfortable. There was a significant relationship between when someone disclosed and under what circumstances – individuals who disclosed under unfavorable circumstances (such as a hospitalization or displaying symptoms on the job) tended to disclose at a later time (after an average of six months on the job) compared to those who disclosed under favorable circumstances (when they felt comfortable or when they felt secure in the job) who tended to disclose sooner (after about one month on the job). People with schizophrenia diagnoses, people with part time employment, and former disability benefit recipients tended to disclose earlier in their job process, suggesting that a need for accommodations may play a role in earlier disclosure patterns.
Limitations of this study include the possibility that the sampling approach may have produced a self-selection bias, positively skewing the results toward higher reported disclosure ra