Content uploaded by Marsha Langer Ellison
Author content
All content in this area was uploaded by Marsha Langer Ellison on Jul 03, 2022
Content may be subject to copyright.
ORIGINAL ARTICLE
Professionals and Managers With Severe Mental Illnesses
Findings From a National Survey
Marsha Langer Ellison, PhD,* Zlatka Russinova, PhD,† Asya Lyass, PhD (Cand.),†
and E. Sally Rogers, ScD†
Abstract: This study explores the capacity of individuals with
severe mental illness to be employed in managerial or professional
jobs and the correlates of their vocational success. Using purposive
sampling techniques, we identified a national sample of 347 indi-
viduals for a mail survey who had succeeded in obtaining and
retaining mid to upper level managerial or professional positions.
The majority worked full-time and held their job for more than 2
years. Their vocational success was operationalized based on 4
employment outcomes: employment status (full-time vs. part-time),
job tenure, occupational rank, and annual income. Key factors that
contributed to respondents’ vocational success were lesser severity
of the illness as indicated by lack of lifetime receipt of disability
benefits, capacity to manage one’s own psychiatric condition, and
higher education. Study findings point to the role of supported
education and self-efficacy in promoting the employment outcomes
among individuals with severe mental illnesses.
Key Words: Employment, careers, severe mental illness, mental
health, managers, professionals, vocational rehabilitation,
psychiatric rehabilitation.
(J Nerv Ment Dis 2008;196: 179 –189)
The prognosis for steady employment with adequate in-
come among individuals with severe mental illness has
been historically pessimistic. Despite the efforts of rehabili-
tation programs to advance opportunities for this population
through a range of vocational programs (Bond et al., 2001),
overall employment rates remain appallingly low (Cook,
2006; Mueser et al., 2001; President’s New Freedom Com-
mission on Mental Health, 2003). Employment services, even
in programs using evidence-based interventions, have tended
to result in low-wage and entry level positions typically in the
janitorial, clerical, and food services occupations (Baron and
Salzer, 2000; Cook, et al., 2005; Fabian, 1999). In fact, findings
from the largest employment study conducted to date in the
United States (Cook et al., 2005) found that only approximately
35% of the participants who received enhanced vocational
rehabilitation services worked for pay at any given point in time
during the study and most worked in “low-skilled” jobs. Clearly,
the picture painted by these data suggests significant upper
boundaries in the achievement of advanced levels of employ-
ment for individuals with severe mental illness.
Given the low rates of employment and underemploy-
ment for individuals with severe mental illness, researchers
have attempted to examine whether and how robustly demo-
graphic and clinical variables can predict vocational success
among individuals with severe mental illness (Cook et al.,
2005; Rogers et al., 1995; Russinova et al., 2002). Although
comparisons across employment studies have been chal-
lenged by the use of different metrics for vocational out-
comes, certain robust predictors have been found. In terms of
clinical correlates, recent studies have found a modest rela-
tionship between psychiatric symptoms and work outcomes
(Mueser et al., 2001; Razzano et al., 2005; Rogers et al.,
1997). Further, studies have demonstrated fairly consistently
that a diagnosis of schizophrenia or other psychotic disorders
is related to poorer vocational outcomes (Cook et al., 2005;
Jacobs et al., 1992; McGurk et al., 2003; Wewiorski and
Fabian, 2004). Yet other studies confirm that higher levels of
social skills, greater social supports and better premorbid
psychiatric functioning predict better work outcomes (Cook
et al., 2005; Rogers et al., 2004; Tsang et al., 2000). The role
of substance abuse in predicting employment success has
been equivocal (Drebing et al., 2002; Razzano et al., 2005;
Rogers et al., 2006). Studies also suggest that work history
positively predicts vocational outcomes and that being white,
male, younger, having a higher education, being married, and
not receiving disability benefits, are predictors of positive
vocational outcomes (Burke-Miller et al., 2006; Cook et al.,
2005; Pluta and Accordino, 2006; Rogers, et al., 2006;
Salkever et al., 2007; Tsang et al., 2000). The role of
cognitive functioning in employment as been the more recent
*Center for Health Policy and Research, University of Massachusetts Med-
ical School, Shrewsbury, Massachusetts; and †Center for Psychiatric
Rehabilitation, Sargent College of Health and Rehabilitation Sciences,
Boston University, Boston, Massachusetts.
Supported by a grant (H133B40024) from the National Institute on Disability
and Rehabilitation Research (NIDRR), US Department of Education and
the Center for Mental Health Services (CMHS), Substance Abuse and
Mental Health Services Administration.
The findings and interpretation of the data expressed in this article do not
necessarily represent the views of NIDRR or the CMHS but are the sole
responsibility of the authors.
Send reprint requests to E. Sally Rogers, ScD, Center for Psychiatric
Rehabilitation, Sargent College of Health and Rehabilitation Sciences,
Boston University, 940 Commonwealth Avenue West, Boston, MA
02215. E-mail: erogers@bu.edu.
Copyright © 2008 by Lippincott Williams & Wilkins
ISSN: 0022-3018/08/19603-0179
DOI: 10.1097/NMD.0b013e318166303c
The Journal of Nervous and Mental Disease • Volume 196, Number 3, March 2008 179
object of investigation. Research suggests that cognitive im-
pairment may present barriers to vocational achievement (cf.,
McGurk et al., 2003, 2005).
Despite the almost wholesale focus on entry-level em-
ployment for individuals with severe mental illness, a handful
of studies in recent years have suggested that higher level
careers are possible. For example, Harding and Zahnizer
(1994) describe as a “myth” the belief that those with severe
mental illness can succeed only in low-level jobs, and refer to
the wide heterogeneity in employment outcomes and income
found in their longitudinal study (Harding et al., 1987).
Recently, Pluta and Accordino (2006) reported that predis-
ability occupational experience and younger age predicted the
return to work among disability insurance claimants with
mental illness. These authors retrospectively identified 300
people from an insurance database who before their disability
were competitively employed in professional, technical, or
managerial occupations and had a median annual salary of
$58,519. McCrohan et al. (1994) reported that among 274
individuals with severe mental illness 1.8% had a current or
previous professional, technical or managerial employment.
A few other studies have also provided descriptive information
about professionals with a severe mental illness (Baron and
Salzer, 2002; O’Day et al., 2006; Salyers, et al., 2004). However,
the small sample sizes and qualitative nature of these studies
precludes a fuller understanding of the correlates their success or
testing inferences about job tenure. In summary, previous stud-
ies which intimate that higher levels of vocational achievement
are possible among people with severe mental illness have
provided very little information about the clinical, demographic,
or job characteristics of those individuals who are able to sustain
a professional or managerial position.
To the contrary, there is a vast body of literature on the
correlates and predictors of job tenure in the general popula-
tion (Aamodt, 2007; Kinicki et al., 2002; Landy and Conte,
2007) and more narrowly, among workers in mental health
professions (cf., Lum et al., 1998; Robinson et al., 2005;
Shader et al., 2001). Over the past few decades this research
has been sharpened and refined, resulting in an array of
empirically tested factors which have repeatedly demon-
strated their relationship to predicting job satisfaction and
promoting job tenure in the general population. Examples of
such factors include: organizational and job commitment;
role conflict and ambiguity; occupational stress; pay satisfac-
tion and reward structure; group cohesion; job involvement
and “embeddedness”; characteristics of the job such as au-
tonomy and variety, organizational structure, climate and
leader relations, communication; opportunities for advance-
ment; workload; decision latitude; and supportive work rela-
tionships (Aamodt, 2007; Griffeth et al., 2000; Kinicki et al.,
2002; Meyer and Allen, 1997). Results from studies and
relevant meta-analyses suggest a complex relationship among
these variables with multiple mediators, and that antecedent
and job-related variables including job satisfaction and orga-
nizational commitment can predict job tenure.
Despite a rather sophisticated understanding of the
correlates and predictors of job success and tenure in the
general population, this enormous body of literature has not
addressed the applicability of these findings to the vocational
experiences of individuals with severe mental illness, nor has
mental health research examined the broad occupational fac-
tors that have been studied in the general population. Studies
involving the general population do not have as their focus
the illness-related issues and concerns that can impede work
performance among persons with severe mental illness. At
the same time, mental health research has not yet examined in
depth the characteristics of individuals who are able to obtain
and sustain more than entry-level employment and the pre-
dictors of their vocational success. Our study was designed to
fill this gap and generate new information that can guide the
delivery of vocational rehabilitation and mental health ser-
vices to individuals with severe mental illnesses.
The purpose of this study was to explore the capacity of
individuals with severe mental illnesses to function in pro-
fessional and managerial jobs. The primary objective of the
study was to document that individuals with severe mental
illnesses who have succeeded in mid to upper level jobs do
exist. We hypothesized that there were persons with severe
mental illnesses who could obtain diverse professional and
managerial jobs and retain them for substantial periods of
time. Exploration of the factors that influence the employ-
ment success of such individuals constituted the second
objective of the study. More specifically, we examined the
correlates of 4 employment outcomes in this sample: (a)
employment status (full-time vs. part-time); (b) job tenure;
(c) occupational rank; and (d) annual income. We hypothe-
sized that employment success across these vocational out-
comes would be associated with a set of demographic and
clinical variables, internal resources (i.e., work motivation
and coping skills), supports, and occupational factors. Al-
though relevant mental health and occupational literature
guided our choice of predictors, we placed an emphasis on
variables highlighting the interplay between the person’s
clinical profile and work-related internal and external re-
sources that might account for vocational success.
METHODS
Study Participants
Eligibility criteria for the study included: (a) having a
severe mental illness, and (b) being currently employed in a
professional/managerial job for at least 6 months before data
collection. To be eligible due to a severe mental illness the
respondent had to self-report a psychiatric diagnosis com-
bined with a report of a lifetime receipt of Social Security
disability benefits due to mental illness or a history of at least
1 psychiatric hospitalization. Employment in a professional/
managerial job was determined in 2 steps. First, we screened
for endorsement of at least 1 out of 4 criteria for having a mid
to upper level job that were synthesized from the occupa-
tional literature (Pavalko, 1988; US Department of Labor,
1993). These included: (a) supervisory responsibilities; (b)
autonomy of daily decision making at work; (c) creative
nature of the job; and (d) control over work-related resources.
We implemented a second round of screening for mid to
upper level employment based on the occupational rank
Ellison et al. The Journal of Nervous and Mental Disease • Volume 196, Number 3, March 2008
© 2008 Lippincott Williams & Wilkins180
assigned to the respondent’s current job according to the job
coding procedures described below.
A total of 347 individuals were found eligible for this
study. Sixty-three percent (n⫽218) were women, and the
majority were aged 40 to 49 (n⫽166, 48%). Ninety-one
percent (n⫽315) were white, 42% were married or cohab-
iting (n⫽145), 30% (n⫽103) - single (never married), the
remainder 28% (n⫽97) were divorced, widowed, or sepa-
rated. Respondents were well educated: 48% (n⫽165) had
a master’s or doctoral degree and another 35% (n⫽122) had
a college degree.
Measures
An extensive mail survey was created in consultation
with mental health experts and a consumer advisory commit-
tee comprised of individuals with psychiatric disabilities who
had worked in professional jobs. The instrument consisted of
an initial self-screening eligibility form addressing the study
inclusion criteria, the informed consent form which was
reviewed and approved by the Boston University Institutional
Review Board, and the survey items.
The survey consisted largely of items developed for this
study and organized in 6 sections: current career status,
disclosure at work, professional challenges and workplace
accommodations, coping strategies, mental health history,
and demographic characteristics. We created items to address
key domains of respondents’ vocational and psychosocial
functioning. Items were primarily closed-ended, Likert scales
or checklists requiring either endorsement or ranking of all
relevant responses (specific metrics for each item are pre-
sented in the results section).
In addition to the items we created for the survey, we
included the ways of coping questionnaire (short version)
(Folkman et al., 1987) which is a 31-item scale with possible
subscale mean scores between 0 (not used) and 3 (used a
great deal). It is a widely used instrument measuring 8
different types of coping strategies used in a content-specific
stressful situation. Our survey instructed participants to de-
scribe the most stressful situation experienced at work in the
past month and then to respond to the questions coping with
that context in mind. The internal consistency coefficients for
the instrument’s 8 subscales were consistent with the origi-
nally reported values; they ranged between 0.41 and 0.78.
The survey instrument went through numerous itera-
tions of review, feedback, and revision to assure content
validity and appropriateness of questions and responses. It
was then pilot tested in 2 rounds with individuals with mental
illness employed in professional positions.
Recruitment and Data Collection
Participants were recruited using a nonrepresentative,
purposive sampling strategy due to the unknown scope of the
universe of professionals and managers with severe mental
illnesses. Recruitment methods relied on snowball sampling
techniques involving Internet and mail announcements,
newsletter postings, survey distribution by enrolled partici-
pants, and direct solicitation of professionals who had pub-
licly disclosed having a mental illness.
Recruitment was conducted from 1997 through 1999. A
total of 812 surveys were mailed to potential participants. A
prompt, including the survey, was sent out to 258 initial
nonresponders 2 to 6 months after the first mailing. By study
end, completed surveys were returned by 563 people. Of
these, 216 were not included because they did not meet the
study eligibility criteria. Ineligibility was determined either
by the respondents based on the self-screening form included
in the survey or by the authors based on consequent review of
reported data. Dividing the returned eligible surveys (n⫽
347) by the total sent out (n⫽812) minus the ineligibles
(n⫽216) resulted in a response rate of 58% using Dillman’s
formula (Dillman, 1978).
Coding Procedures
Information on participants’ occupation was gathered
through open-ended questions pertaining to job title and
duties that required subsequent coding by the investigators.
First, respondents’ jobs were classified by industry or type of
work using job categorizations available in the Occupational
Outlook Handbook and developed by the US Bureau of Labor
Statistics (U. S. Department of Labor, 2000). Next, we coded
for the job’s major occupational field using a coding scheme
that distinguished among jobs in mental health self-help and
advocacy, mental health services, other helping professions
and nonhelping professions. Finally, occupational rank was
assigned to each job Hollingshead nine-rank occupational
scale designed to rank occupations by complexity and sophis-
tication (Hollingshead, unpublished data, 1975).
Because occupational rank was one of the employment
outcomes examined in the study, we assessed the interrater
reliability of assigned codes. We established the interrater
reliability based on 70 surveys selected randomly from the
sample of 347 (20%). The first 2 authors coded each job
independently using Hollingshead’s coding scale. The inter-
rater correlation coefficient (ICC) for coding of occupational
rank was 0.88, indicating a high level of agreement.
Independent and Dependent Variables
In addition to systematizing the descriptive information
about vocational and mental health functioning in this study
sample, we explored the significant correlates of the follow-
ing employment outcomes: (a) employment status; (b) job
tenure; (c) occupational rank; and (d) annual income. Em-
ployment status was dichotomized (full-time vs. part-time)
after individuals with seasonal/temporary employment or
self-employment were excluded from the analyses. Job tenure
was dichotomized using 2-year retention as the cutoff point.
Occupational rank and annual income were treated as con-
tinuous variables: Hollingshead’s values ranged from 6 to 9
whereas annual income was reported in five $10,000 incre-
ments with the highest category accounting for income above
$50,000. Annual income was treated as a secondary outcome
and the first 3 outcome variables were added to the list of
independent variables. We used 6 sets of independent vari-
ables as possible correlates for each employment outcome:
(a) demographic variables (age, gender, marital status, race,
education); (b) clinical variables (diagnosis, symptomatol-
ogy, receipt of disability benefits, medical comorbidity, men-
The Journal of Nervous and Mental Disease • Volume 196, Number 3, March 2008 Professionals With Mental Illnesses
© 2008 Lippincott Williams & Wilkins 181
tal health services); (c) prior work history (before and after
illness onset); (d) internal factors (coping skills, work moti-
vation, level of stress at work); (e) occupational factors
(occupational setting and culture, work autonomy, benefits,
job accommodations and built-in flexibility, interpersonal
stressors); (f) supports (outside and at the job).
Statistical Analysis
All analyses were performed in SPSS 15.0 or SAS 9.
First, frequencies and descriptive statistics were obtained for
all variables. Next, the relationship between the independent
and dependent variables was tested in a univariate manner.
Linear regression or ANOVA were used to identify correlates
for occupational rank and annual income whereas logistic
regression was employed for analysis of job tenure and
employment status. Variables that were related to the out-
come at the
␣
level of 0.15 or below were used in the next
step of the analyses. Hierarchical multivariate analyses
(linear regression for continuous outcomes and logistic
regression for categorical outcomes) with a cutoff of 0.15
were performed for each separate set of independent vari-
ables. Independent variables from each of the tested 6
subsets that were significantly related to a given outcome
were combined and hierarchical multivariate analysis was
then performed to determine the final set of correlates.
This approach was repeated for each of the 4 employment
outcomes.
RESULTS
Vocational Outcomes
Table 1 presents descriptive information about study
participants’ occupational rank, employment status, job ten-
ure, annual income, and occupational field. All respondents
had a mid to upper level of employment with more than a third
(38%) having higher level managerial and professional jobs.
Three quarters of study participants (76%) had full-time perma-
nent jobs and two-thirds (64%) had held their job for more than
2 years. Those with higher level jobs were more likely to work
full-time (
2
⫽10.5, df ⫽3, p⫽0.015) and to be employed in
nonmental health or advocacy settings (
2
⫽80.2, df ⫽9, p⬍
0.0001). However, participants’ occupational rank was not as-
sociated with years on the job. At the same time, longer job
tenure was associated with full-time employment status (
2
⫽
5.4, df ⫽1, p⫽0.020). The type of respondents’ occupational
setting was not significantly associated with either job tenure or
full-time employment status.
Most study participants (83%, n⫽289) had worked
before the onset of their mental illness, with 46% (n⫽160)
reporting more than 5 years of prior work experience and 8%
(n⫽28) reporting less than a year of previous work. Lack of
employment prior to the onset of illness was not significantly
associated with psychiatric diagnosis, occupational rank, job
tenure, full-time employment, and the type of setting where
the person was currently employed. The majority of respon-
dents (78%, n⫽269) reported having at least 1 previous mid-
to upper-level job after the onset of their mental illness (that
is, a job before the one that made them eligible for study
participation). A very small portion of respondents got their
current job through a rehabilitation program (4%, n⫽13) or
with the help of their physician or therapist (2%, n⫽8). The
remaining respondents acquired their job through personal
contacts (29%, n⫽100), the newspaper (16%, n⫽56),
professional publications and bulletin boards (6%, n⫽20), or
they developed the job themselves (13%, n⫽44). Seventy-
one percent (n⫽245) reported an increase in job responsi-
bilities since starting the index job and 50% (n⫽175) stated
TABLE 1. Employment Outcomes Among Professionals and
Managers With Severe Mental Illnesses (N⫽347)
NPercent
Hollingshead occupational rank
Higher executives, major professionals (9) 64 18
Administrators, lesser professionals (8) 68 20
Managers, minor professionals (7) 133 38
Technicians, semiprofessionals and small business
owners (6)
82 24
Current employment status
Full-time, permanent 264 76
Part-time, permanent 48 14
Temporary (contractual or seasonal) 14 4
Self-employed 20 6
Job tenure
6–12 mo 78 23
13–24 mo 47 14
2–5 yr 124 36
Over 5 yr 97 28
Annual income
Under $10,000 26 8
$10,000–19,999 38 11
$20,000–29,999 93 27
$30,000–39,999 63 18
$40,000–49,999 48 14
Above $50,000 73 21
Occupational setting
Self-help and advocacy 85 25
Mental health 107 31
Health and human services (except mental health) 50 14
Nonhelping setting 105 30
Occupations
Top executives 27 8
Managers 126 36
Engineers 7 2
Scientists (math, computer, basic sciences) 11 3
Lawyers/paralegals 8 2
Social scientists 4 1
Researchers 12 4
Health professionals (physicians, nurses,
technologists)
18 5
Mental health professionals 74 21
Educators 27 8
Communications specialists (editors, writers, public
relations, marketing)
11 3
Arts 1 ⬍1
Advocacy 21 6
Ellison et al. The Journal of Nervous and Mental Disease • Volume 196, Number 3, March 2008
© 2008 Lippincott Williams & Wilkins182
that they were not concerned about losing the job because of
their mental illness.
Mental Health Characteristics and Challenges
at Work
Forty-five percent (n⫽155) of the sample reported a
diagnosis of bipolar disorder, 28% (n⫽98) a depressive
disorder, 15% (n⫽51) a schizophrenia spectrum disorder;
6% (n⫽19) reported other diagnoses; and 7% (n⫽24)
did not specify a diagnosis (note that individuals were
queried about their mental illness using 2 questions: the
first was a dichotomous question asking about the presence
of mental illness and the second requesting a diagnosis.
Thus, individuals could be eligible for the study without
providing a specific diagnosis).
Thirty-eight percent (n⫽131) reported receipt of Social
Security disability benefits at some time due to mental illness.
Those with a schizophrenia spectrum disorder were more likely
to have received disability benefits (
2
⫽19.7, df ⫽3, p⬍
0.0001). Almost all participants had a prior psychiatric hospital-
ization (97%, n⫽338), with 61% (n⫽211) having more than
3 hospitalizations over their lifetime. Age at first psychiatric
hospitalization for a large portion of the sample (46%; n⫽161)
occurred before the age of 25 whereas for 27% (n⫽95) it was
between the ages of 31 and 45. Several participants (39%; n⫽
137) had not had a hospitalization in the last 3 years.
Thirty-seven percent of the participants (n⫽129)
reported experiencing between 1 (25%, n⫽86) and 5 (1%,
n⫽2) co-occurring medical conditions at the time of the
survey. The most frequently reported medical conditions
were muscular-skeletal disorders (i.e., arthritis and back
problems) (12%, n⫽43), migraine/headaches (5%, n⫽17),
cardiovascular disorders (i.e., hypertension) (4%, n⫽15),
respiratory disorders (4%, n⫽15), and gastrointestinal dis-
orders (i.e., irritable bowel syndrome; 4%, n⫽13).
Almost all participants (98%, n⫽341) had taken
psychotropic medications at some point in their life; only 1
individual reported not using psychotropic medications de-
spite having 1 psychiatric hospitalization and 5 respondents
did not disclose their medication use. At the time of the
survey, 85% (n⫽296) were taking psychotropic medications
whereas 66% (n⫽229) reported being engaged in individual
psychotherapy. Individuals who considered the use of medi-
cations as contributing to their work outcomes were also more
likely to attribute primary importance to the support they have
been receiving from their psychiatrist or therapist (
2
⫽14.87,
df ⫽1, p⫽0.0001). Respondents with bipolar disorder were
more likely to consider the consistent use of medications as very
important for their vocational success (
2
⫽14.36, df ⫽3, p⫽
0.0024). Twenty-seven percent (n⫽92) were actively involved
with a self-help group, 35% (n⫽121) never engaged in
self-help and the remaining respondents used self-help sporad-
ically. Involvement with self-help did not affect the use of
psychotherapy; however, those who used self-help attributed
greater importance to the support of other consumers vis-a`-vis
working successfully (F⫽25.59, df ⫽1, p⬍0.0001).
Almost all study participants (98%, n⫽341) experi-
enced their jobs as moderately to very stressful and reported
that their ability to perform on the job was affected by
subjective experiences emanating from their psychiatric con-
dition as well as by a range of interpersonal challenges (Table
2). We also calculated 2 separate summary scores for the total
number of psychiatric symptoms and interpersonal challenges
experienced by participants. On average respondents reported
dealing with 5 symptoms that interfered with their work
performance and with 3 stressful interpersonal challenges.
We used summary scores as predictors.
Internal and External Factors Influencing
Employment Outcomes
To better understand the vocational success of the study
sample we inquired about a range of internal and external
factors that we expected to be related to work performance.
Internal factors focused primarily on coping with one’s own
psychiatric condition and on sources of work motivation
TABLE 2. Mental Health and Interpersonal Challenges
Experienced by Study Participants
NPercent
Level of stress at current job
Very high 117 33.7
High 133 38.3
Moderate 91 26.2
Low 2 0.6
Very low 3 0.9
Subjective experiences related to psychiatric
condition
Feeling tired. 244 70.3
Having fluctuations in my mood. 219 63.1
Not being able to concentrate on my work. 198 57.1
Getting too anxious regarding my performance. 187 53.9
Having difficulty remembering information. 176 50.7
Finding it hard to go to work every day. 166 47.8
Having trouble organizing/prioritizing tasks. 147 42.4
Having trouble organizing my time. 129 37.2
Having side effects from my medications. 126 36.3
Hearing voices or feeling disoriented at times. 63 18.2
Stressful interpersonal experiences
Feeling like I have to fit in or act like everyone
else who does not have psychiatric disabilities.
118 34.0
Asserting myself with my boss and colleagues
regarding taking on/not taking on additional
responsibilities.
113 32.6
Feeling socially isolated from my colleagues
(coworkers).
102 29.4
Interpersonal difficulties with coworkers. 99 28.5
Having colleagues that do not understand
psychiatric disability.
88 25.4
Feeling I have to hide my psychiatric condition
from others or that others might find out.
79 22.8
Feeling unsupported by my boss. 78 22.5
Feeling stigmatized on the job. 69 19.9
Having a supervisor that does not understand
psychiatric disability.
58 16.7
Having to negotiate work accommodations with
my boss.
55 15.9
Being treated differently than others on the job. 55 15.9
The Journal of Nervous and Mental Disease • Volume 196, Number 3, March 2008 Professionals With Mental Illnesses
© 2008 Lippincott Williams & Wilkins 183
while the external factors were limited to supports influencing
the person’s capacity to work and to key job characteristics
especially relevant to the study population (i.e., accommoda-
tions and inherent flexibility of the job). Table 3 describes
these independent variables.
Correlates of Employment Success
Table 4 displays the results from the final regression
models for the 4 employment outcomes examined in this
study. Below we report on the direction of the significant
associations at both the univariate and multivariate level of
analyses by employment outcome. To avoid repetition, we
first present the results from the multivariate analyses and
then describe the independent variables that were statistically
significant only at the univariate level but not the multivariate
level. (Note that some variables that are significant at the 0.15
level appear in the table because they were included in the
final multivariate model.)
Full-Time Employment
Respondents who never received Social Security dis-
ability benefits or who attributed their vocational success to
having learned how to manage their psychiatric condition
were more likely to work full-time. Males, those who were
more educated or those who experienced their job as more
stressful were also more likely to be employed full-time at the
univariate level of analysis.
TABLE 3. Internal and External Factors Influencing Participants’ Employment Outcomes
NPercent Mean ⴞSD
Internal factors
Ways of coping subscale (0 ⫽not used; 3 ⫽used a great deal). — — —
Planful problem solving — — 1.69 ⫾0.74
Seeking social support — — 1.45 ⫾0.74
Self-controlling — — 1.32 ⫾0.66
Confrontive coping — — 1.03 ⫾0.70
Escape avoidance — — 0.98 ⫾0.77
Accepting responsibility — — 0.95 ⫾0.76
Positive reappraisal — — 0.78 ⫾0.89
Distancing — — 0.76 ⫾0.62
Capacity to regulate work (1 ⫽hardly able; 10 ⫽fully able). — — 7.67 ⫾2.03
My interest and satisfaction with what I do.
a
258 74.4 —
My job-related skills and performance.
a
192 55.3 —
My own drive and will power.
a
185 53.3 —
I have learned how to manage my condition so it does not affect
my job too much.
a
177 51.0 —
I have learned how to adjust my medications so that I have fewer
problems.
a
82 23.6 —
Occupational factors —
Work flexibility.
c
203 58.5 —
Opportunities for growth and professional development.
a
149 42.9 —
Job accommodations.
b
142 40.9 —
The control I have over my work.
a
105 30.3 —
The organizational culture and values.
a
59 17.0 —
The benefits I get at the job.
a
51 14.7 —
Supports —
The consistent use of my psychiatric medications.
a
165 47.6 —
The support of my spouse (partner).
a
111 32.0 —
The support of my psychiatrist/therapist(s).
a
110 31.7 —
The support of my boss (supervisor).
a
99 28.5 —
The support of my colleagues on the job.
a
93 26.8 —
The support of my friends.
a
90 25.9 —
My spiritual/religious involvement.
a
69 19.9 —
The support of other consumers.
a
67 19.3 —
The support of a mentor or role model.
a
65 18.7 —
The recognition I get for the work I do.
a
63 18.2 —
The support of my parents or other family member.
a
49 14.1 —
The support of my children.
a
37 10.7 —
a
The N(%) represents the number of responders who ranked these items among the top 3 most important ones.
b
The N(%) represents the number of responders who received any job accommodations.
c
The N(%) represents the number of responders who received any job flexibility.
Ellison et al. The Journal of Nervous and Mental Disease • Volume 196, Number 3, March 2008
© 2008 Lippincott Williams & Wilkins184
Job Tenure
Participants who were older, never received Social Secu-
rity disability benefits, experienced their job as more stressful, or
did not attribute primary importance to the opportunities the job
provided for growth and development, to the control they had
over their work or to the organizational culture, had longer job
tenure. In addition, findings from the univariate analyses sug-
gested that those who were married and had jobs either in mental
health or nonhelping settings were more likely to have held their
job for a longer period of time.
Occupational Rank
Individuals who were male, more educated, had never
received Social Security disability benefits, more fre-
quently used problem-solving strategies to cope with
work-related stress, were employed in nonmental health
jobs, or those for whom the support of parents, children
and other consumers did not have a primary importance for
their vocational success, were more likely to hold a job
with a have higher occupational rank. Also, at the univar-
iate level, those who were older, who reported a greater
number of symptoms, attributed their job’s success to their
skills and performance, who relied on the support of their
psychiatrist or therapist but did not attribute primary im-
portance to the support of their boss or to their workplace
organizational culture, were more likely to hold a job with
higher occupational rank.
Annual Income
Participants with higher occupational rank, who were
employed full-time, had longer job tenure or were married,
had higher annual incomes. The univariate analyses revealed
significant associations of annual income with several other
independent variables: those who were male, more highly
educated, with a diagnosis of bipolar disorder (when com-
pared with a diagnosis of a schizophrenia spectrum disorder),
TABLE 4. Multivariate Analyses for Employment Status, Job Tenure, Occupational Rank,
and Annual Income

Wald
2
df p OR
Full-time employment
Disability benefits
a
1.26 13.12 1 0.0003 3.54
Learned how to manage my condition 1.07 8.37 1 0.0038 2.92
Level of stress at work
ab
⫺0.32 2.58 1 0.1081 0.73
Job tenure
Age 0.51 11.02 1 0.0009 1.67
Disability benefits
a
0.52 4.30 1 0.0381 1.68
Level of stress at work
a
⫺0.56 13.27 1 0.0003 0.57
Opportunities for professional growth ⫺0.55 4.68 1 0.0305 0.58
Control over work ⫺0.80 8.51 1 0.0035 0.45
Organizational culture and values ⫺0.75 5.57 1 0.0183 0.47

Fdf p Partial R
2
Occupational rank (R
2
⫽0.4107)
Education 0.27 64.53 1 ⬍0.0001 0.1998
Gender ⫺0.21 4.36 1 0.0376 0.0087
Disability benefits
a
0.28 7.95 1 0.0051 0.0111
Problem-solving strategies subscale 0.12 3.93 1 0.0483 0.0101
Occupational setting (mental health)
c
⫺0.27 4.49 1 0.0348 0.0071
Occupational setting (human services)
c
0.36 5.24 1 0.0228 0.0276
Occupational setting (nonhelping)
c
0.54 17.32 1 ⬍0.0001 0.0970
Organizational culture and values
b
⫺0.22 3.29 1 0.0709 0.0062
Support of other consumers ⫺0.25 4.23 1 0.0406 0.0062
Support of children ⫺0.40 6.80 1 0.0095 0.0170
Support of extended family ⫺0.30 5.02 1 0.0258 0.0200
Income (R
2
⫽0.5590)
Marital status 0.26 4.85 1 0.0284 0.0079
Gender
b
⫺0.19 2.46 1 0.1180 0.0036
Full-time status 1.22 55.97 1 ⬍0.0001 0.0999
Job tenure 0.32 6.95 1 0.0088 0.0112
Occupational rank 0.85 227.65 1 ⬍0.0001 0.4363
a
For these variables, low score corresponds to a positive outcome.
b
These variables are not significantly related to the outcome, but are adjusted for in the model because of the 0.15 cutoff
used in the hierarchical regression.
c
When compared with self-help and advocacy setting.
The Journal of Nervous and Mental Disease • Volume 196, Number 3, March 2008 Professionals With Mental Illnesses
© 2008 Lippincott Williams & Wilkins 185
with fewer psychiatric hospitalizations, who had never re-
ceived Social Security disability benefits, who reported a
greater number of symptoms, employed in jobs other than
mental health advocacy and services, who experienced their
job as more stressful, who attributed their vocational success
to their skills and performance, who relied on the support of
their spouse and their psychiatrist or therapist but did not
attribute primary importance to their workplace’s organiza-
tional culture or to the support of their children or who less
frequently used positive reappraisal strategies to cope at
work, were more likely to have higher annual income.
DISCUSSION
This study expands our current knowledge about the
capacity of individuals with severe mental illnesses to suc-
ceed in mid to upper level competitive jobs despite the
challenges and limitations arising from their psychiatric condi-
tion. Although the feasibility of career achievement in this
population has been outlined through personal or anecdotal
accounts by distinguished consumer advocates, high-profile
public figures and in a few recent studies conducted with either
very small (Hammen et al., 2000; O’Day et al., 2006; Salyers et
al., 2004) or selective samples (Elinson et al., 2004), our study
presents evidence based on a national sample selected specifi-
cally to investigate higher level employment outcomes and to
include more severe psychiatric disorders.
Study participants were employed in an impressively
rich spectrum of jobs cutting across all major professional
fields, including mental health, medicine, law, science, engi-
neering, business, education, and so forth. The majority of
respondents was working full-time and had a job tenure
consistent with the trends in the general population, with
more than a quarter of the sample having a tenure above the
national median (Copeland, 2006). These findings highlight
the potential of individuals with severe mental illnesses to
acquire and maintain jobs beyond entry-level positions that
correspond to their professional training and qualifications.
Thus, these findings suggest a different potential for the
vocational aspirations of individuals with psychiatric disabil-
ities and have major implications for vocational counselors
and mental health practitioners who traditionally tend to have
lower expectations and who direct clients toward low-level
jobs (O’Day and Killeen, 2002).
Respondents’ vocational achievements are remarkable
in light of their mental health histories and current mental
health status. All study participants had experienced a mental
illness of substantial severity as indicated by either a lifetime
receipt of Social Security benefits and/or a history of psychi-
atric hospitalizations. At the same time, the sample’s mental
health profile clearly indicated that most participants had not
reached the point of “cure” or remission and many continued
to struggle with psychiatric symptoms, cognitive deficits, and
medications’ side effects. Hence, these individuals were able
to get and keep mid to upper level jobs despite the challenges
presented by their psychiatric condition.
We operationalized the vocational success of study
participants based on 4 different employment outcomes (full-
vs. part-time employment status, job tenure, occupational
rank, and annual income) and explored the correlates associ-
ated with each of them. These outcomes were of particular
interest to us because they illustrate the individual’s ability to
work at full capacity and to hold a professional job for
extended periods of time. Although the sample’s composition
prevents us from identifying the factors that distinguish
between individuals with severe mental illnesses who work in
low-level jobs and those who succeed in mid to upper level
jobs, we explored the correlates of participants’ occupational
rank within the range included in the study. The correlates of
annual income were examined only as a secondary analysis
because when examining its correlates we controlled for the
first 3 employment outcomes given their likely impact on the
person’s annual earnings. We found different sets of variables
explaining each employment outcome, with only a small
number of correlates having significant associations with
more than 1 outcome. In addition, several correlates were
significant only at the univariate level of analysis while other
correlates were associated with some outcomes at the multi-
variate level but at the univariate level with others. These
findings suggest a complex interplay among the demo-
graphic, clinical, internal, and external factors that explain
study participants’ overall vocational success.
The factor that was consistently associated with all
primary employment outcomes at both the univariate and the
multivariate levels of analysis was respondents’ lifetime his-
tory of receiving Social Security disability benefits. Those
who have received disability benefits in the past or were still
receiving them were three-and-a-half times more likely to be
employed part-time (as opposed to full-time) and one-and-a-
half times more likely to have job tenure of less than 2 years.
Disability benefits also had a small impact on participant’s
occupational status with those having received benefits hold-
ing a lower-level job within the studied mid to upper spec-
trum of employment. These findings suggest that disability
benefits may play a dual role in the work lives of persons with
severe mental illnesses. On the one hand, they can serve as a
proxy measure for the severity of the person’s mental illness
whereas on the other hand, their receipt may bring into play
the disincentives inherent in such benefits. Although severity
of illness may have an overarching effect on employment
outcomes, disability-related policies and work disincentives
may exert a stronger influence on one’s ability to work
full-time and keep a job for prolonged period of times.
The impact of disability benefits have rarely been
investigated (Tsang et al., 2000), with only 1 study reporting
that receipt of Social Security benefits was negatively asso-
ciated with getting a job (Jacobs et al., 1992). However, the
disincentives to work entailed in disability benefits is well
understood (Stapleton and Burkhauser, 2003). O’Day and
Kileen (2002) reported that participants in their qualitative
study described limiting their work hours and seeking low-
level jobs or restricting themselves to volunteer work to
preserve their disability benefits. Although this might occur
frequently among individuals with severe mental illnesses
who receive disability benefits, our findings of full-time and
higher-level employment within this population suggests that
these disincentives may be overcome if other internal and
Ellison et al. The Journal of Nervous and Mental Disease • Volume 196, Number 3, March 2008
© 2008 Lippincott Williams & Wilkins186
external resources are available. It is also evident that receipt
of disability benefits is an important factor that needs to be
considered in future research on the employment outcomes in
this population.
Several demographic factors were associated with the
employment outcomes examined in our study. Consistent
with some previous research (Elinson et al., 2004; Salkever et
al., 2007) males were more likely to have a higher occupa-
tional rank, to earn more money and to work full-time.
However gender was not associated with job tenure. Unlike
most previous studies which associated younger age with
better employment outcomes (Tsang et al., 2000; Wewiorski
and Fabian, 2004), we found that being older was associated
with better job tenure and higher occupational rank. This
inconsistency might be due to the difference in employment
outcomes typically examined in other studies (i.e., working
vs. not working; getting a job; months of employment in 3
and 12 months postplacement; Wewiorski and Fabian, 2004)
and to the fact that these studies focused primarily on low-
level jobs. Being married contributed to having higher annual
earnings independent of occupational rank, full-time employ-
ment or job tenure and was also associated with longer job
tenure. These findings are consistent with reports about the
effects of marriage in the general population (Lillard and
Panis, 1996) and other studies of individuals with severe
mental illnesses (Cook et al., 2005). Finally, education was
the strongest predictor of higher occupational rank and was
also associated with annual income. Race was the only
demographic variable that was not associated with partici-
pants’ vocational functioning, possibly because of the very
limited representation of minorities in the study.
Contrary to previous research, clinical characteristics of
study participants were rarely significantly related to their em-
ployment outcomes. Psychiatric diagnosis was associated at the
univariate level only with annual income and not with other
outcomes. Individuals with a schizophrenia spectrum disorder
earned less than did persons with affective or other diagnoses.
However, individuals with a schizophrenia spectrum disorder
were more likely to have received Social Security disability
benefits. Because we examined disability benefits received as a
proxy measure of the illness severity, this finding suggests that
it maybe the severity of the illness that impacts employment
outcomes and not the diagnostic category per se. At the same
time, higher numbers of psychiatric hospitalizations, which are
often considered a proxy measure of illness severity, were
associated only with annual income.
Study findings about the associations between psychi-
atric symptomatology and employment outcomes were unan-
ticipated. Our summary score was significantly associated at
the univariate level only with occupational rank and annual
income. However, contrary to our expectations, those who
reported a greater number of symptoms tended to have both
higher occupational rank and annual income suggesting that
upper levels of work achievement may take a toll on symp-
toms. This unexpected finding prompted us to examine the
association of each reported symptom with employment out-
comes. We found that none of the individual symptom items
were significantly associated with employment outcomes. We
did not use a standardized symptom measure; rather we
calculated a summary score for 10 different symptoms and
cognitive deficits that might interfere with a person’s capacity
to work. Future studies should use a standardized symptom
measure to better understand the relationship between symp-
toms and work performance. This unexpected finding sug-
gests that our study participants had learned how to cope
more effectively with the impact of work related stress on
their emotional well being. Further support for these hypoth-
eses is provided by 2 other unexpected findings in the study:
first, individuals who experienced their jobs as more stressful
tended to have longer job tenure, work full-time and earn
higher income and second, medical comorbidity did not
affect employment outcomes.
The use of mental health services was not associated
with any of the employment outcomes, although overall
participants reported being heavy users of psychopharmaco-
logical and psychotherapy services and many of them stated
that these services played a primary role in their vocational
success. It is possible that the relatively pervasive use of such
services in the study sample did not generate sufficient
variance to allow for these variables to emerge as significant
predictors. Still the descriptive information provided by study
participants suggests that consistent use of psychotropic med-
ications and psychotherapy are essential for the professional
success for many persons with severe mental illnesses.
Contrary to our expectations, most of the internal fac-
tors examined in the study did not correlate with employment
outcomes despite the fact that many respondents described
their drive and will power, work satisfaction and ability to
manage their psychiatric condition as very important to their
vocational success. Only participants’ reported capacity to
manage their own psychiatric condition was associated with
longer job tenure. Very few of the coping strategies explored
in the study had an impact on employment outcomes. As
examples, individuals who relied more on problem-solving cop-
ing strategies were more likely to have a higher occupational
rank whereas those who used positive reappraisal strategies were
more likely to have longer job tenure, although they earned less
per year. However, it is important to note that overall, respon-
dents reported relying more frequently on problem-solving,
seeking social support, self-controlling, and confrontive coping
strategies when dealing with stressful situations at the work-
place. These types of coping strategies tend to be more adaptive
and characteristic of the general population when compared with
psychiatric populations (Folkman and Lazarus, 1986; Vitaliano
et al., 1987).
Among the occupational factors examined in the study,
type of occupational setting was the one with most significant
impact on employment outcomes. As expected, individuals
employed in nonmental health settings were more likely to
have a higher occupational rank and to earn higher income.
At the same time, at the univariate level of analysis employ-
ment in either a nonhelping or mental health setting was
associated with longer job tenure. Previous research suggests
that the culture and climate of the workplace and the fit
between the person’s and work environment values are im-
portant factors associated with continued employment among
The Journal of Nervous and Mental Disease • Volume 196, Number 3, March 2008 Professionals With Mental Illnesses
© 2008 Lippincott Williams & Wilkins 187
individuals with severe mental illness (Kirsh, 2000). How-
ever, contrary to our initial expectations, we found that study
participants who did not consider the organizational culture
as very important to their vocational success were more likely
to have a longer job tenure, higher occupational status or
annual income. Surprisingly, individuals who attributed less
importance to the opportunities their job presented for growth
and development and to the level of work autonomy had
longer job tenure. Although it is possible that individuals for
whom the above job characteristics are less important might
have more flexibility in adjusting to their work environment,
further research is necessary to examine in depth the impact
of the person – environment fit on the employment of persons
with severe mental illnesses.
Although social support has been identified as critical
to the recovery psychiatric disabilities (Rogers et al., 2004), it
seemed to have a very modest impact on the employment
outcomes of this study sample. Different types of support
were endorsed by approximately a quarter of the respondents
and not a single type of support was significantly associated
with job tenure or full-time employment. Only the support of
a spouse was associated at the univariate level with higher
occupational rank and annual income. Those who attributed
less importance to support from their family or children or
from other consumers were more likely to hold jobs with
higher occupational rank. Perhaps individuals in such jobs
rely less on the support of family members or peers.
Another unexpected finding from the study was the lack
of association between employment outcomes and previous
work history, which has been consistently found to be a
predictor of employment success in previous research (Tsang
et al., 2000; Wewiorski and Fabian, 2004). It was also
unexpected that reasonable accommodations and built-in
flexibility on the job were not associated with employment
outcomes in this study. It is possible that individuals in this
study have achieved a level of mental stability that, combined
with their professional expertise, sets them apart from indi-
viduals who, at the very early stages of their vocational
recovery, require more accommodations and greater support
both at and outside of the job.
Study Limitations
The nonrepresentative sampling techniques used in this
study may limit the generalizability of our conclusions. Cor-
relates of vocational success were established within the
scope of the study sample and may not account for individ-
uals with mental illnesses who are employed at lower-level
jobs. In addition, the survey relied on self-report and misrep-
resentation in responses cannot be ruled out; however, mul-
tiple checks on the data lend confidence to the data accuracy
(e.g., women and those employed in the human and social
services reported lower salaries than those employed in busi-
ness). Additionally, the survey methodology itself relies on
data gathering using constructed items rather than standard-
ized instruments. Consequently, all results must be consid-
ered tentative until they can be replicated in larger studies.
CONCLUSIONS
The national scope of this study highlights the existence
of a hidden population of individuals with severe mental
illnesses who have the potential to sustain professional and
managerial jobs. Participants succeeded in diverse occupa-
tional fields that corresponded to their training and expertise
despite significant mental health histories and current symp-
toms. Education, capacity to manage one’s own psychiatric
condition, and cope with stress emerged as the most promi-
nent malleable factors that contributed to this sample’s vo-
cational achievements. These findings point to the value of
supported education services and interventions to promote
self-management and self-efficacy as well as interventions to
improve coping strategies as key areas to be targeted by a
recovery-oriented mental health system.
REFERENCES
Aamodt MG (2007) In Industrial/Organizational Psychology: An Applied
Approach. Belmont (CA): Wadsworth/Thompson.
Baron RC, Salzer MS (2000) The career patterns of persons with serious
mental illness: generating a new vision of lifetime careers for those in
recovery. Psychiatr Rehabil Skills. 4:136–156.
Baron RC, Salzer MS (2002) Accounting for unemployment among people
with mental illness. Behav Sci Law. 20:585–599.
Bond GR, Becker DR, Drake RE, Rapp CA, Meisler N, Lehman AF, Bell
MD, Blyler CR (2001) Implementing supported employment as an evi-
dence-based practice. Psychiatr Serv. 52:313–322.
Burke-Miller JK, Cook JA, Grey DD, Razzano LA, Blyler CR, Leff HS,
Gold PB, Goldberg RW, Mueser KT, Cook WL, Hoppe SK, Stewart M,
Blankertz LE, Dudek K, Taylor AL, Carey MA (2006) Demographic
characteristics and employment among people with severe mental illness
in a multisite study. Community Ment Health J. 42:143–159.
Cook JA (2006) Employment barriers for persons with psychiatric disabili-
ties: update of a report for the president’s commission. Psychiatr Serv.
57:1391–1405.
Cook JA, Leff HS, Blyler CR, Gold PB, Goldberg RW, Mueser KT,
McFarlane WR, Shafer MS, Blankertz LE, Dudek K, Razzano LA, Grey
DD, Burke-Miller J (2005) Results of a multisite randomized trial of
supported employment interventions for individuals with severe mental
illness. Arch Gen Psychiatry. 62:505–512.
Copeland C (2006) Employee Tenure (Vol 28, pp 2–11). Washington, DC:
Employee Benefit Research Institute.
Dillman D (1978) Mail and Telephone Surveys: The Total Design Method.
New York (NY): John Wiley and Sons.
Drebing CE, Fleitas R, Moore A, Krebs C, Van Ormer A, Penk W, Seibyl C,
Rosenheck R (2002) Patterns in work functioning and vocational rehabil-
itation associated with coexisting psychiatric and substance use disorders.
Rehabil Couns Bull. 46:5–13.
Elinson L, Houck P, Marcus SC, Pincus HA (2004) Depression and the
ability to work. Psychiatr Serv. 55:29–34.
Fabian ES (1999) Rethinking work: The example of consumers with serious
mental health disorders. Rehabil Couns Bull. 42:302–316.
Folkman S, Lazarus RS (1986) Stress processes and depressive symptom-
atology. J Abnorm Psychol. 95:107–113.
Folkman S, Lazarus RS, Pimley S, Novacek J (1987) Age differences in
stress and coping processes. Psychol Aging. 2:171–184.
Griffeth RW, Hom PW, Gaertner S (2000) A meta-analysis of antecedents
and correlates of employee turnover: Update, moderator tests and research
implications for the next millennium. J Manage. 26:463–488.
Hammen C, Gitlin M, Altshuler L (2000) Predictors of work adjustment in
bipolar I patients: A naturalistic longitudinal follow-up. J Consult Clin
Psychol. 68:220–225.
Harding CM, Brooks GW, Ashikaga T, Strauss JS, Breier A (1987) The
Vermont longitudinal study of persons with severe mental illness. II.
Long-term outcome of subjects who retrospectively met DSM-III criteria
for schizophrenia. Am J Psychiatry. 144:727–735.
Harding CM, Zahniser JH (1994) Empirical correction of seven myths about
Ellison et al. The Journal of Nervous and Mental Disease • Volume 196, Number 3, March 2008
© 2008 Lippincott Williams & Wilkins188
schizophrenia with implications for treatment. Acta Psychiatr Scand
Suppl. 90:140–146.
Jacobs H, Wissusik D, Collier R, Stackman D, Burkeman D (1992) Corre-
lations between psychiatric disabilities and vocational outcome. Hosp
Commun Psychiatry. 43:365–369.
Kinicki AJ, McKee-Ryan FM, Schriesheim CA, Carson KP (2002) Assess-
ing the construct validity of the job descriptive index: A review and
meta-analysis. J Appl Psychol. 87:14–32.
Kirsh B (2000) Organizational culture, climate and person-environment fit:
Relationships with employment outcomes for mental health consumers.
Work. 14:109–122.
Landy FJ, Conte JM (2007) Work in the 21st Century: An Introduction to
Industrial Organizational Psychology (2nd ed). Malden (MA): Blackwell
Publishing.
Lillard LA, Panis CW (1996) Marital status and morality: The role of health.
Demography. 33:313–327.
Lum L, Kervin J, Clark K, Reid F, Sirola W (1998) Explaining nursing
turnover intent: Job satisfaction, pay satisfaction, or organizational com-
mitment. J Organ Bhav. 19:305–320.
McCrohan NM, Mowbray CT, Bybee D, Harris SN (1994) Employment
histories and expectations of persons with psychiatric disorders. Rehabil
Couns Bull. 38:59–71.
McGurk SR, Mueser KT, Harvey PD, LaPuglia R, Marder J (2003) Cogni-
tive and symptom predictors of work outcomes for clients with schizo-
phrenia in support employment. Psychiatr Serv. 54:1129–1135.
McGurk SR, Mueser KT, Pascaris A (2005) Cognitive training and supported
employment for persons with severe mental illness: One-year results from
a randomized controlled trial. Schizophr Bull. 31:898–909.
Meyer J, Allen N (1997) Commitment in the Workplace: Theory, Research
and Application. Thousand Oaks (CA): SAGE Publications.
Mueser KT, Salyers MP, Mueser PR (2001) A prospective analysis of work
in schizophrenia. Schizophr Bull. 27:281–296.
O’Day B, Killeen M (2002) Does U.S. federal policy support employment
and recovery for people with psychiatric disabilities. Behav Sci Law.
20:559 –583.
O’Day B, Killeen M, Goldberg S (2006) Not just any job: People with
psychiatric disabilities build careers. J Vocat Behav. 25:119–131.
Pavalko R (1988) Sociology of Occupations and Professions. Itaska: F. E.
Peacock.
Pluta DJ, Accordino MP (2006) Predictors of return to work for people with
psychiatric disabilities. A private sector perspective. Rehabil Couns Bull.
49:102–110.
President’s New Freedom Commission on Mental Health (2003) Achieving
the Promise: Transforming Mental Health Care in America. Final Report.
Rockville (MD): DHHS. Pub. No. SMA-03-3832.
Razzano LA, Cook JA, Burke-Miller JK, Mueser KT, Pickett-Schenk SA,
Grey DD, Goldberg RW, Blyler CR, Gold PB, Leff HS, Lehman AF,
Shafer MS, Blankertz LE, McFarlane WR, Toprac MG, Carey MA (2005)
Clinical factors associated with employment among people with severe
mental illness: Findings from the employment intervention demonstration
program. J Nerv Ment Dis. 193:705–713.
Robinson S, Murrells T, Smith EM (2005) Retaining the mental health
nursing workforce: Early indicators of retention and attrition. Int J Ment
Health Nurs. 14:230–242.
Rogers ES, Anthony W, Lyass A (2004) The nature and dimensions of social
support among individuals with severe mental illnesses. Community Ment
Health J. 40:437–450.
Rogers ES, Anthony WA, Cohen M, Davies RR (1997) Prediction of
vocational outcome based on clinical and demographic indicators among
vocationally ready clients. Community Ment Health J. 33:99–112.
Rogers ES, Anthony WA, Lyass A (2006) A randomized clinical trial of
psychiatric vocational rehabilitation. Rehabil Couns Bull. 49:143–156.
Rogers ES, Sciarappa K, MacDonald Wilson K, Danley K (1995) A benefit-
cost analysis of a supported employment model for persons with psychi-
atric disabilities. Eval Program Plann. 18:105–115.
Russinova Z, Wewiorski N, Lyass A, Rogers ES, Massaro JM (2002)
Correlates of vocational recovery for persons with schizophrenia. Int Rev
Psychiatr. 14:303–311.
Salkever DS, Karakus MC, Slade EP, Harding CM, Hough RL, Rosenheck
RA, Swartz MS, Barrio C, Yamada AM (2007) Measures and predictors
of community-based employment and earnings of persons with schizo-
phrenia in a multisite study. Psychiatr Serv. 28:315–324.
Salyers MP, Becker DR, Drake RE, Torrey WC, Wyzik PF (2004) A ten-year
follow-up of a supported employment program. Psychiatr Serv. 55:302–308.
Shader K, Broome ME, Broome CD, West ME, Nash M (2001) Factors
Influencing satisfaction and anticipated turnover for nurses in an academic
medical center. J Nurs Admin. 31:210–216.
Stapleton D, Burkhauser R (eds) (2003) The Decline in Employment of
People with Disabilities. Kalamazoo, MI: W.E. Upjohn Institute for
Employment Research.
Tsang H, Lam P, Ng B, Leung O (2000) Predictors of employment outcomes
for people with psychiatric disabilities: A review of literature since the mid
’80s. J Rehabil. 66:19–31.
U.S. Department of Labor (1993) Defining the Terms - Executive, Adminis-
trative, Professional and Outside Sales (Vol 1281). U.S. Department of
Labor Employment Standards Administration: Wage and Hour Division
WH Publication, Revised May 1993.
U.S. Department of Labor (2000) Occupational Outlook Handbook. India-
napolis: JIST works.
Vitaliano PP, Katon W, Russo J, Maiuro RD, Anderson K, Jones M (1987)
Coping as an index of illness behavior in panic disorder. J Nerv Ment Dis.
175:78 – 84.
Wewiorski NJ, Fabian ES (2004) Association between demographic and
diagnostic factors and employment outcomes for people with psychi-
atric disabilities: A synthesis of recent research. Ment Health Serv Res.
6:9 –21.
The Journal of Nervous and Mental Disease • Volume 196, Number 3, March 2008 Professionals With Mental Illnesses
© 2008 Lippincott Williams & Wilkins 189
