Determinants of Employment Status

Document Sample
Determinants of Employment Status Powered By Docstoc
					112                                                                                          RCB 47:2 pp. 112–120, 122 (2004)

           Determinants of Employment Status
          Among a Community-Based Sample of
                 People with Epilepsy:
                            Implications for Rehabilitation Interventions

Malachy Bishop                             People with epilepsy may face a number of complex and interacting problems and bar-
University of Kentucky                     riers related to employment. Understanding these barriers is an important step toward
                                           implementing effective rehabilitation interventions. This article explores the factors
                                           that predicted employment status among a community-based sample of adults with
                                           epilepsy. Data for the analysis were collected from 146 adults with epilepsy through
                                           Epilepsy Foundation chapters in Wisconsin, Kentucky, and Ohio. Several demographic
                                           and epilepsy-related variables were analyzed to explore the determinants of employment
                                           status. Results of a binary logistic regression analysis suggest that gender, number of
                                           antiepilepsy medications used, seizure frequency, and reported interference in daily func-
                                           tioning created by seizures were the best predictors of employment. Additional analyses
                                           suggested that the possession of a driver’s license was also associated with employment
                                           status. The results are discussed in terms of their implications for rehabilitation-based

       he employment situation for adults with epilepsy                  neuropsychological deficits, and these in turn
       has consistently been characterized by higher lev-                increase the risk for unemployment. It is also
       els of unemployment and underemployment than                      well-known that unemployment increases the
are seen among the general population (Cooper, 1995;                     risk for psychiatric disorders and psychosocial
Fisher, 2000). The potential problems faced by individu-                 problems, the latter being correlated with a
als with epilepsy in the area of employment are numerous                 bad seizure prognosis. (p. 2219)
and are typically experienced in an additive fashion, such
that achieving successful employment frequently involves                 It is to a great extent due to the complex and in-
overcoming multiple barriers.                                      teractive nature of these employment barriers that the
                                                                   development of effective vocational rehabilitation inter-
      There is broad agreement in the literature that              ventions remains a challenging goal. Understanding these
      the employment problems of people with                       barriers is an important step toward implementing ef-
      epilepsy cannot be reduced to one factor (i.e.,              fective rehabilitation interventions. The purpose of this
      seizure severity) but that they are rather the re-           article is to contribute to the understanding among reha-
      sult of a bundle of adversive factors interacting            bilitation professionals of the factors that affect the em-
      with each other in a complex fashion. (Thor-                 ployability and employment status of people with epilepsy.
      becke & Fraser, 1997, p. 2211)                               Understanding the factors that contribute to the risk of
These interrelated factors can, for some individuals, con-         unemployment, and developing a profile of such factors,
verge, complicating the employment situation exponen-              allows vocational rehabilitation counselors to prioritize
tially, as described by Thorbecke and Fraser:                      their services (Rumrill, 1996). Toward this end, this arti-
                                                                   cle describes an analysis of the factors that predict em-
      Frequent seizures may increase the likelihood                ployment status among adults with epilepsy. The results of
      of seizures occurring at work. . . . Frequent                the present study are discussed in terms of targeting reha-
      seizures at the same time increase the risk for              bilitation interventions toward those factors most amen-
Volume 47, No. 2 (Winter 2004)                                                                                        113

able to treatment, and toward those persons most needing       studies examining the relationship between employment
rehabilitation services.                                       and seizure frequency.

                                                               Seizure Impact and Employment
        E PILEPSY      AND       E MPLOYMENT
                                                               Although seizures are relatively objective events, individ-
In the United States the unemployment rate among peo-          uals vary in their subjective evaluation of the impact that
ple with epilepsy who are eligible workers or who are          seizures have on their lives. Recent research suggests that
maintaining an active job search ranges from 13% to 25%        the perceived impact of seizures may be an important
(Fisher, 2000; Thorbecke & Fraser, 1997). Higher unem-         predictor of employment status. For example, Schachter,
ployment rates (50% and up) have been reported among           Shafer, and Murphy (1993), in a study of 150 adults with
people in tertiary care centers, people who have one or        epilepsy, found employment status and reported impact of
more generalized tonic-clonic seizures or complex partial      seizures to be strongly correlated. Unemployed respon-
seizures a year (Hauser & Hesdorffer, 1990; Thorbecke &        dents in that study were more than twice as likely to re-
Fraser, 1997), and people who have been discouraged            port feeling greatly affected by seizures than employed
from seeking employment (Fisher, 2000). In a recent            respondents (p = 0.007). The relationship between the
community-based national survey of adults with epilepsy,       perceived impact of seizures and employment status has
25% of working-age respondents were unemployed                 not, however, been examined to a great extent.
(Fisher, 2000). Of those, 64% reported that the unem-
ployment was a direct result of their epilepsy. This study     Psychosocial Variables
was conducted at a time when the average unemployment          Various psychosocial factors have been implicated in the
rate in the United States was slightly over 5%.                high unemployment rate among persons with epilepsy, in-
      A number of epilepsy-related variables have been         cluding social isolation, lack of information, social skills
examined for their impact on employment status. The fol-       deficits, lack of family support, and fears about and expe-
lowing review outlines the findings of previous investiga-     riences with negative attitudes on the part of employers
tions.                                                         (Devinsky, 1994; Thorbecke & Fraser, 1997).
                                                                     The impact of epilepsy on social functioning has
Seizure Variables and Employment                               tremendous interindividual variability. Some people with
                                                               epilepsy have few, if any, disruptions of social interaction
The degree to which seizure type and frequency affect em-      and functioning, whereas others have severe problems
ployment status remains something of an open question.         that make it more difficult for them to lead productive
Some researchers have suggested that the evidence to           lives (Austin & deBoer, 1997). In general, however,
date does not conclusively support a direct relationship       adults with epilepsy have been found to have a higher
between seizure variables and employment (Hauser &             prevalence of social problems, including isolation and
Hesdorffer, 1990; Thorbecke & Fraser, 1997). In several        problems with adaptation, as compared with people in the
recent studies, however, unemployment was found to be          general population (Austin & deBoer, 1997).
higher among individuals with more frequent seizures                 Concerns about experiencing negative attitudes,
(Chaplin, Wester, & Tomson, 1998; Jacoby, Baker, Steen,        stigma, and discrimination by employers and co-workers
Potts, & Chadwick, 1996; Ratsepp, Oun, Haldre, &               has been identified in several studies as playing a signifi-
Kaasik, 2000; Yagi, 1998).                                     cant role in unemployment among persons with epilepsy.
      Yagi (1998) reported that in a sample of 312 outpa-      For example, in a large-scale study of adults recently diag-
tients in Japan, the frequency of their seizures was related   nosed with epilepsy, 70% of the participants expected to
to their employment. Among the patients who had six or         experience stigma and discrimination at work (Chaplin,
fewer seizures per year, 83% were employed full-time.          Yepez Lasso, Shorvon, & Floyd, 1992). In another large
Among the patients who experienced one or more                 study, conducted in the United Kingdom with people
seizures per month, only 31% were employed full-time.          whose epilepsy was in remission, 32% of the participants
Similar results were reported in an Estonia-based study, in    reported feeling that their epilepsy made it more difficult
which 90 people with epilepsy were surveyed and a signif-      for them than for others to get a job (Jacoby, 1994).
icant negative relationship between seizure frequency and
full-time employment was found (Ratsepp et al., 2000).
Jacoby et al. (1996), in a survey of an unselected popula-     Demographic Variables
tion (N = 696) in the United Kingdom, also found that          A number of demographic variables have been explored
seizure frequency was negatively associated with employ-       in previous investigations of the determinants of employ-
ment status. A review of the literature suggests that there    ment status, generally with the expected results. For
is a lack of recent community-based and large-scale U.S.       example, one would expect that people with previous ex-
114                                                                                       Rehabilitation Counseling Bulletin

periences with successful employment would be more                  2. focus on the relationship between employ-
likely to be employed. And indeed, the amount of time                  ment status and seizure type, socioeconomic
employed since entering the workforce has been posi-                   status, and other predictors; and
tively correlated with current employment in several stud-          3. target predictors of unemployment for fur-
ies (Chaplin et al., 1998; Hauser & Hesdorffer, 1990). In              ther study and development.
addition, positive associations between a person’s educa-
tion level and employment status have been reported             The goal of the present population-based study was to
(Ratsepp et al., 2000; Scambler & Hopkins, 1980).               identify the demographic and epilepsy-related variables
      A relationship between comorbid physical or psychi-       associated with employment status and increase the cur-
atric disability and lower rates of employment has also         rent understanding by including variables previously
been found in several investigations (Jacoby, 1992; Mar-        unexplored or inadequately evaluated. The following re-
tin & White, 1988; Yagi, 1989). For example, Yagi re-           search hypotheses were evaluated:
ported lower employment rates among people with
physical disabilities and epilepsy, psychiatric disorders and        1. Age is negatively related to employment
epilepsy, and epilepsy and intellectual impairment, with                status.
employment rates being lower as severity of the disability           2. Gender is associated with employment
increased.                                                              status.
      The relationship between age and employment sta-               3. Graduation from college is positively re-
tus has been infrequently examined and with conflicting                 lated to employment status.
results. According to Hauser and Hesdorffer (1990), age              4. Seizure frequency is negatively related to
per se is not a predictor of employment status, but age at              employment status.
onset of epilepsy may be. That is, earlier onset seems more          5. Level of reported interference in daily life
likely to negatively affect employment status. These au-                by seizures is negatively related to employ-
thors continued to say, however, that the latter observa-               ment status.
tion is likely confounded by the presence of concurrent              6. Number of antiepilepsy medications used
underlying neurological disabilities, such as cerebral palsy            is negatively related to employment status.
and mental retardation.                                              7. Years since diagnosis is positively related to
      Despite numerous national and international studies               employment status.
concerning the employment situation of persons with                  8. Having tonic-clonic seizures or experienc-
epilepsy, the relationship between gender and employ-                   ing a combination of seizure types is nega-
ment status has not been studied. DeBoer (1991) reported                tively related to employment status
that “no literature seems to be available on this topic”             9. Having a valid driver’s license is positively
(p. 33). My review of the more recent literature similarly              related to employment status.
failed to find that this variable has received research at-         10. Increased size of town or city is positively
tention to date.                                                        associated with employment status.
      Also underexplored in the epilepsy and employment             11. Having a comorbid disability is negatively
research is the relationship between employment status                  associated with employment status.
and (a) access to transportation and (b) rural versus urban
dwelling. In the present study the relationships between
employment status, possession of a driver’s license, and                              M ETHOD
town size (urban vs. rural dwelling) were explored. Be-
cause living in an urban area implies access to a broader       Data Collection
range of employment opportunities than is found in rural
areas, this variable, like the possession of a valid driver’s   Data for this study were collected from people who were
license, was hypothesized to be positively associated with      on the mailing list of Epilepsy Foundation (EF) chapters
employment status.                                              in three midwestern states. Participants were adults with
      To develop comprehensive understanding of the ef-         epilepsy who were recruited through several chapters of
fects of epilepsy on employment, multiple variables and         the Epilepsy Foundation throughout Wisconsin, and in
their potential interrelationships must be considered.          one chapter each in Ohio and Kentucky.
Hauser and Hesdorffer (1990), in their comprehensive re-
view of employment among people with epilepsy, sug-                  Wisconsin Sample. In Wisconsin, data were ob-
gested that future research in this area should                 tained by way of a 110-item questionnaire. During the
                                                                winter of 1999–2000, 560 surveys were mailed to members
      1. use population-based, as opposed to clinical,          on the mailing lists of several Wisconsin Epilepsy Foun-
         samples;                                               dation chapters. One hundred seven usable responses
Volume 47, No. 2 (Winter 2004)                                                                                              115

were returned. The purpose of this earlier study was to es-          Epilepsy-related information was also collected from
tablish relationships between epilepsy and quality-of-life      the participants. Specifically, participants were asked
ratings. However, data were collected regarding several
variables related to the hypotheses in the present study.            1. how many years it had been since they were
The response rate for the Wisconsin sample was 19%.                     diagnosed with epilepsy,
                                                                     2. how many antiepilepsy medications they
      Ohio and Kentucky Samples. In the fall of                         currently used,
2001, two hundred 37-item surveys were mailed to people              3. how frequently they experienced seizures,
on the mailing lists of the Ohio and Kentucky Epilepsy               4. what type of seizure they experienced, and
Foundation chapters (100 surveys mailed by each chap-                5. to what extent seizures interfered in their
ter). The survey included questions about the respon-                   daily function.
dents’ epilepsy and its effect on their quality of life, work
satisfaction, and employment status. The survey questions       Seizure frequency was categorized as daily, weekly,
related to the hypotheses in the present study were             monthly, yearly, or no seizure in more than 2 years. Extent of
phrased in exactly the same way as they were in the in-         reported interference by seizures in daily life was catego-
strument used for the Wisconsin sample.                         rized as not at all, somewhat, moderately, quite a bit, or ex-
      The Epilepsy Foundation mailing lists for Ohio and        tremely.
Kentucky were not stratified by age or diagnosis. In other
words, the mailing lists included both children and adults
with epilepsy who were currently receiving services, as
well as people who had previously requested information         A total of 153 completed surveys were included in the
from the Epilepsy Foundation chapters but may or may            initial sample. For the purpose of this employment-
not have been diagnosed with epilepsy. A total of 80 sur-       status–based analysis, participants age 60 or above were
veys were returned to the investigator with notes stating       removed from the sample, leaving a total of 146 partici-
that the recipients did not meet the criteria for participa-    pants. The ages of the remaining participants ranged from
tion (adults with diagnosed epilepsy). Forty-six completed      18 to 59 years. Ninety-two (63%) of the participants were
surveys were returned from Ohio and Kentucky. Assuming          employed either full- or part-time, and 54 (37%) were
that all the recipients who did not respond were eligible       unemployed. Approximately 97% were White (non-
to do so, this suggests a response rate of approximately        Hispanic). The sample was 66% female. Approximately
38%. It is more likely, however, that some additional per-      31% of the respondents were college graduates. Forty-five
centage of the nonrespondents were not eligible, and in         percent of the participants reported a monthly household
that case, the response rate among adults with diagnosed        income of $2,000 or less, and 55% reported a monthly
epilepsy would be higher.                                       household income of above $2,000.
      Due to financial limitations associated with this               Just under half (49.3%) of the sample had been diag-
study, a second mailing of the survey was not conducted.        nosed with epilepsy more than 20 years ago. The remain-
In an effort to ensure a good response rate, the EF chap-       ing half of the sample was fairly evenly distributed
ters in each state were asked to include in their newslet-      between 1 and 20 years since diagnosis. Approximately
ters (which were sent to the entire mailing list) and on        94% of the sample were taking medication for seizure con-
their Web pages a message describing the survey, encour-        trol, with more than half of the sample (57.7%) taking
aging the recipients to complete and return the survey,         more than one antiepilepsy medication. More than half of
and giving the contact information for the researcher.          the respondents (56%) reported having seizures at least
                                                                monthly. At the same time, approximately 70% reported
                                                                that seizures interfered in their daily life only a little bit or
                                                                not at all. The majority of the respondents (approximately
Data were collected from the two samples using essen-           62%) reported that they experienced a single type of
tially the same survey instrument, with the following ex-       seizure, such as simple-partial (11%), complex-partial
ceptions. In the Ohio and Kentucky survey, participants         (30%), or tonic-clonic (formerly grand mal; 12%). The
were asked whether they had a valid driver’s license,           remaining participants reported experiencing mixed
whether they had a comorbid disability, and the size of the     seizure types.
town or city in which they lived. This information was                Information regarding possession of a driver’s license,
not collected on the instrument used for the Wisconsin          existence of a comorbid disability in addition to the diag-
sample. Data on demographic variables were collected for        nosis of epilepsy, and size of participant’s town or city was
all three samples. These included gender, age, employ-          collected only from the Kentucky and Ohio samples.
ment status, ethnicity, level of education, and monthly         Twenty-six (58%) of the people in these samples reported
household income.                                               having valid driver’s licenses. Twenty-one (45%) reported
116                                                                                          Rehabilitation Counseling Bulletin

having a comorbid disability. There was a range of comor-         uted to the explanatory power of the model, χ2(5, N =
bid disabilities reported, including traumatic brain injury,      146) = 41.760, p < .01. Together, these variables ac-
depression, learning disability, arthritis, and diabetes.         counted for 35% of the variance (Nagelkerke R2 = .35).
Among these samples, 32.5% reported living in a town of           The results of the final stage of the analysis are summa-
fewer than 10,000 people, 14% lived in towns of between           rized in Table 2. This model correctly predicted group
10,000 and 50,000 people, 7% lived in cities of between           membership 78% of the time. Specifically, the model ac-
50,000 and 100,000 people, 20.9% reported living in               curately classified 91% of those who were employed and
cities of between 100,000 and 500,000 people, and over a          55.8% of those who were unemployed (see Table 3).
quarter (25.6%) lived in cities of over 500,000 people.                 One of the demographic variables and three of the
Summary information about the combined sample is pre-             epilepsy-related variables were found to contribute to the
sented in Table 1, with descriptive statistics related to the     explanatory power of the model. Gender was found to be
independent variables presented according to employ-              significantly related to employment status (Wald = 4.888,
ment status category.                                             p = .027). The Exp (B) for gender (2.786) is interpreted as
                                                                  meaning that compared to men, women in this sample
                                                                  were almost 3 times more likely to be employed than
Data Analysis
Binary logistic regression analysis and cross-tabulation                The number of antiepilepsy medications used was
analysis were used to test the hypotheses. A backward bi-         found to be positively associated with employment status.
nary logistic regression analysis was conducted to deter-         This categorical variable was coded using the “simple”
mine the relationship between employment status and the           contrast method (Norusis, 1999), such that the people
selected demographic variables (age, education level, and         taking no medication were contrasted (a) with people
gender) and epilepsy-related variables (seizure frequency,        taking one medication and (b) people taking more than
level of reported interference in daily function by seizures,     one medication. Interestingly, people who were taking
number of antiepilepsy medications used, seizure type, and        antiepilepsy medication were much more likely to be em-
years since diagnosis). Logistic regression analysis is ap-       ployed than people who were not. People taking one med-
propriate when the dependent variable (in this case, em-          ication were 20 times more likely to be employed than
ployment status) is dichotomous (Wright, 1995). The               people taking no medication, Exp (B) = 20.038. People
predictor variables in a logistic regression analysis can be      taking two or more medications were almost 30 times
continuous or noncontinuous. Like linear regression, the          more likely to be employed than people taking no med-
logistic regression model relates one or more predictor           ication, Exp (B) = 29.746.
variables to a dependent variable and yields regression co-             Seizure frequency was found to be significantly re-
efficients, predicted values, and residuals. In this type of      lated to employment status (Wald = 5.428, p = .020). As
analysis, one attempts to predict the probability that an         hypothesized, as seizure frequency increased, the likeli-
observation belongs to each of two groups. In other words,        hood of employment decreased, Exp (B) = 1.725. Re-
if a dependent variable is coded as 0 (e.g., unemployed)          ported level of interference in daily life by seizures was
and 1 (e.g., employed), the logistic regression analysis pre-     also found to be significantly related to employment sta-
dicts a probability that an observation belongs to the            tus (Wald = 10.018, p = .002). As the reported level of in-
group designated as 0 and a separate probability value that       terference increased, the likelihood of being employed
the observation belongs to the group designated as 1. The         decreased, Exp (B) = .439.
observation is assigned to the group having the higher                  Cross-tabulation analysis was used to determine
predicted probability.                                            whether employment status was associated with the pos-
      In addition to the logistic regression analysis, a cross-   session of a valid driver’s license, existence of a comorbid
tabulation analysis was used to evaluate the relationship         disability, and rural versus urban dwelling. People who re-
between employment status and the data from only the              ported having a driver’s license were significantly more
Ohio and Kentucky samples (i.e., possession of a valid            likely to be employed than those who did not, χ2(1, N =
driver’s license, city or town size, and existence of a co-       45) = 9.012, p = .003. Neither city size, χ2(1, N =
morbid disability). All tests were conducted at the .05           43) = 9.012, p = .362, nor comorbid disability, χ2(1,
level of significance.                                            N = 43) = 5.372, p = .372, was associated with employ-
                                                                  ment status.

                        R ESULTS
Four variables—gender, seizure frequency, number of med-
ications used to control epilepsy, and reported level of          The results of this study add to our understanding of the
seizure interference in daily life—significantly contrib-         effects of epilepsy on employment, and of both (a) which

TABLE 1. Characteristics of Respondents for the Independent Variables According to Employment Status
                                                         Employed                         Unemployed
Characteristic                                    n                 %                 n                %

   Women                                          57                62               39                42
   Men                                            35                38               14                58
      18–29                                       17                19               11                21
      30–39                                       28                30               13                25
      40–49                                       35                38               16                29
      50–59                                       12                13               13                25
Educational attainment
    College degree or above                       32                35               14                26
    Less than a college degree                    60                65               38                74
Number of years diagnosed with epilepsy
   5 years or less                                20                22               10                19
   6 to 10 years                                   8                 9                6                12
   11 to 15 years                                  8                 9                5                10
   16 to 20 years                                 12                13                4                 8
   More than 20 years                             43                47               27                51
Number of antiepilepsy mdications
   None                                            3                 3                4                 8
   One                                            39                43               15                29
   More than one                                  49                54               33                63
Seizure frequency
     Daily                                         2                 2                6                13
     Weekly                                       12                13               22                42
     Monthly                                      28                32                9                17
     Yearly                                       18                20                8                15
     More than 2 years since seizure              30                33                7                13
Seizure type
     Single                                       68                75               34                64
     Mixed                                        23                25               19                36
Experience tonic-clonic seizures
    Yes                                           81                89               46                87
    No                                            10                11                7                13
Seizures interfere in daily life
     Not at all                                   36                40               12                23
     Somewhat                                     41                45               11                22
     Moderately                                    8                 9               12                23
     Quite a bit                                   5                 6               15                28
     Extremely                                     0                 0                2                 4
City size
     10,000 or less                                7                26                7                41
     10,000 to 50,000                              3                12                3                17
     50,000 to 100,000                             1                 4                2                12
     100,000 to 500,000                            8                31                1                 6
     More than 500,000                             7                27                4                24
Possession of driver’s license
     Yes                                          21                75                5                29
     No                                            7                25               12                71
Comorbid disability
   Yes                                            14                52                6                38
   No                                             13                48               10                62
118                                                                                          Rehabilitation Counseling Bulletin

TABLE 2. Final Model for the Logistic Regression Analysis
Predictor                                      B                SE                Wald                 df                ExpB

Gender                                        1.025              .463              4.888                1                2.786
Number of medications
   No meds. vs. 1 med.                        2.998              .948              9.989                1               20.038
   No meds. vs. +1 med.                       3.393             1.012             11.238                1               29.746
Seizure interference                          −.823              .260             10.018                1                  .439
Seizure frequency                              .545              .234              5.428                1                1.725

TABLE 3. Classification Table for the Final Model of            though rehabilitation counselors are not involved in the
the Logistic Regression Analysis                                treatment of seizures, gathering information about the
                                  Predictor                     client’s seizures can allow the client and counselor to plan
                                                                ahead and minimize the impact of seizures on the job
Observed            Unemployed    Employed         % Correct    through appropriate job matching, job accommodations,
                                                                and client awareness. Specifically, counselors should ask
Unemployed              29             23             55.8      the following (Fraser & Clemmons, 1983; Thorbecke &
Employed                 8             81             91.0      Fraser, 1997): What is the frequency of the seizures? Is
                                                                there an aura, or warning, that allows the individual to
% Correct                                             78.0      prepare for a seizure? What is the state of consciousness
                                                                during a seizure? What is the typical behavior during a
                                                                seizure? How does the person behave after the seizure
people may be at greater risk of unemployment and               (e.g., confusion, speech disturbance, sleep)? How long
(b) which rehabilitation interventions may be most effec-       does it take the person to resume usual activity after a
tive in preventing unemployment. The results suggest            seizure? Do seizures occur at a certain time (e.g., during
that unemployment among persons with epilepsy is more           sleep, after awakening), or are they unpredictable? Do any
likely among men who have more frequent seizures, who           factors precipitate seizure activity (e.g., stress, alcohol, fa-
are not on antiepilepsy medications, who perceive their         tigue, menses, flickering lights)?
seizures to be more intrusive on daily function, and who              In addition, it is important that counselors are aware
do not have their driver’s license. The following discus-       of the complexities of antiepilepsy medications and their
sion describes the rehabilitation counseling implications       relationship to employment. It is clear from this and the
of these results, specifically with regard to gender, seizure   other studies described in the literature review that seizure
frequency, antiepilepsy medications, seizure interference,      frequency is an important determinant of employment
and possession of a driver’s license.                           status. In the present study, seizure frequency was strongly
                                                                and negatively correlated with the number of medications
                                                                used (r = −.46, p = .005). Thus, counselors should be
Gender                                                          aware of, and emphasize the importance of, compliance
Given the lack of previous research examining the rela-         with antiepilepsy medication.
tionship between gender and employment, it is difficult at            At the same time, however, counselors must be
this point to identify the implications of this finding for     aware that the use of antiepilepsy medication is frequently
the rehabilitation counseling relationship. In light of the     accompanied by significant adverse effects with tremen-
finding that the women in this study were almost 3 times        dous potential to disrupt or impair daily functioning. Side
more likely than the men to be employed, it will be im-         effects of antiepilepsy medications can include double vi-
portant in the future to further evaluate the gender issue.     sion, fatigue, nervousness or agitation, difficulty concen-
In the present study, the correlations between gender and       trating, nausea, dizziness, gum overgrowth, facial hair
the other predictor variables were examined. There were         growth, sexual dysfunction, and, occasionally, even more
no significant correlations, suggesting that additional va-     severe side effects (Baker, Jacoby, Buck, Stalgis, & Mon-
riables need to be explored.                                    net, 1997; Fraser & Clemmons, 1983). Generally, at least
                                                                one side effect will be experienced (Baker et al., 1997),
                                                                and Fisher et al. (2000), in a large, community-based U.S.
Seizure Frequency                                               study, reported that almost one out of five people reported
The results of this study underscore the importance of          that the side effects and cost of antiepilepsy medication
seizure control in the maintenance of employment. Al-           were the worst things about having epilepsy.
Volume 47, No. 2 (Winter 2004)                                                                                          119

      Thus, although compliance with a medication regi-        and subsequent related legislation, as well as employer
men is an important aspect of seizure control, counselors      drug-testing policies. Decisions about disclosure should al-
should be aware that compliance is a significant problem       ways be the client’s. The role of the counselor is to make
area among persons with epilepsy, primarily because of the     sure that such decisions are informed and founded on
adverse side effects just mentioned. More antiepilepsy         accurate information. With regard to deciding about dis-
medications are available today than ever before (Brodie       closure, the Epilepsy Foundation’s 1991 manual on job
& Kwan, 2002), and new treatments are rapidly being de-        finding is an excellent resource in the form of a decision-
veloped. Rehabilitation counselors should encourage clients    making matrix through which individuals are guided in
to consult their neurologists as to whether current regi-      developing a disclosure strategy (Epilepsy Foundation,
mens and treatments remain the most effective and the          1991).
least problematic in terms of adverse effects.
                                                               Possession of a Driver’s License
Antiepilepsy Medications                                       Having a driver’s license was positively associated with
In the present study, as the number of antiepilepsy med-       employment in the present study. This may well be due to
ications used increased, the likelihood of employment          the relationship between seizure frequency and the pos-
increased significantly. This finding is particularly in-      session of a driver’s license, a relationship strongly sup-
teresting given the established cognitive and behavioral       ported in this sample (r = −.63, p = .005). However, this
effects of antiepilepsy medications and their potential im-    can be a complicated issue for the rehabilitation coun-
pact on employment, as well as in light of the above dis-      selor, requiring, among other things, knowledge of his or
cussion of the adverse side effects often associated with      her state’s laws related to epilepsy and driving. Each state
these medications. Because antiepilepsy medication has         has its own regulations regarding licensing and epilepsy.
frequently been associated with cognitive or behavioral        The Epilepsy Foundation reviews legal issues and laws
impairment (Kanner, Pellock, & Meador, 2001), I pre-           concerning people with epilepsy, including laws related to
dicted a negative effect of medication on employment.          driving (see
The finding in this study that the likelihood of employ-             The results of the present study highlight the impor-
ment increased with the number of medications used             tance of transportation to the successful maintenance of
underscores the importance of seizure control in the at-       employment. Because acquiring or maintaining a driver’s
tainment and maintenance of employment.                        license can frequently be unpredictable for people who
                                                               experience seizures, counselors need to be cognizant of the
                                                               importance of ensuring that alternate means of trans-
Seizure Interference
                                                               portation to employment are available and accessible.
As has been the case in previous studies, the reported im-
pact of seizures was found to be related to employment
status. Because the perception of seizure interference is                           LIMITATIONS
subjective, it may be amenable to change through coun-
seling. Understanding the client’s perspective is therefore    Limitations to this study included the demographic re-
an important step in vocational counseling. The coun-          strictedness of the sample, the small proportion of respon-
selor may, for example, explore specifically which areas of    dents from racial or ethnic minority backgrounds (97% of
the client’s life are affected and in what way. Regarding      the sample was White [non-Hispanic]), and the reliance
employment, for example, fear of having a seizure at work      on self-report data. The lack of studies among U.S. popu-
has been identified in several studies as a significant con-   lations of the determinants of employment status among
cern among people with epilepsy (Bishop, 2002; Fisher          people with epilepsy was noted. Future research is needed
et al., 2000), and may contribute to their unwillingness to    and should replicate this study among a more geographi-
enter employment. Preparing the client, through educa-         cally and culturally diverse population.
tion about appropriate disclosure, and perhaps through               Another important limitation to be considered in in-
role-play, to communicate with employers and co-workers        terpreting the results of this study is the relatively low re-
that a seizure may occur, what might happen during a           sponse rate. Given that the response rates for the two
seizure, and what the co-workers should and should not do      samples (Wisconsin and Ohio/Kentucky) were estimated
may increase the confidence and comfort of both client         to be 19% and 38%, respectively, there exists some ques-
and co-workers.                                                tion as to the representativeness of the combined sample.
      Similarly, uncertainty about disclosing seizure status   The goal of analyzing a community-based, rather than a
in the application process is a significant concern for        clinical, sample by utilizing the undifferentiated mailing
many people with epilepsy (Bishop, 2002; Bishop &              list of the Epilepsy Foundation accounts for my inability
Allen, 2001). Counselors should develop client awareness       to accurately assess the actual response rate. Also, the fail-
of the guidelines of the Americans with Disabilities Act       ure to conduct a second mailing is certainly a contribut-
120                                                                                              Rehabilitation Counseling Bulletin

ing factor regarding the limited response rate. However, I                        ABOUT THE AUTHOR
did make a number of other efforts to obtain a representa-
tive sample. Specifically, I asked the EF chapters that were   Malachy Bishop, PhD, CRC, is an assistant professor of re-
assisting with this research to include in their newsletters   habilitation counseling in the Department of Special Education
and Web pages a message to the survey recipients encour-       and Rehabilitation Counseling at the University of Kentucky.
aging them to complete and return the survey and giving        His current research interests include quality of life and adap-
my contact information so that I could address any ques-       tation to disability, particularly among persons with epilepsy
tions.                                                         and other chronic neurological conditions. Address: Malachy
      Regarding research design, due to the descriptive and    Bishop, 220 Taylor Education Building, University of Ken-
correlational nature of this study, no definitive causal at-   tucky, Lexington, KY 40506.
tributions can be made regarding the relationship be-
tween the independent variables and employment.
Internal validity was threatened by the selection of inde-                            AUTHOR’S NOTE
pendent variables. The variables in the present study ac-
counted for 35% of the variance. The variables affecting       I would like to express sincere thanks and appreciation to
employment status among adults with epilepsy clearly rep-      the Epilepsy Foundation affiliates who assisted with this
resent a complex and multidimensional set of factors, and      research, particularly Art Taggart, Jane Bolz, and Bert
they may be assessed using both other strategies and addi-     Doyle, of the Epilepsy Foundation of South-Central Wis-
tional variables.                                              consin; Debbie McGraff, of the Epilepsy Foundation of
                                                               Kentuckiana; and Frank Burden, of the Epilepsy Founda-
                                                               tion of Greater Cincinnati. I would also like to acknowl-
                   CONCLUSIONS                                 edge the assistance of Chase Allen, MRC, for his
                                                               significant contribution to the development and comple-
For many persons with epilepsy, success in obtaining and       tion of this research.
maintaining employment is predicated on overcoming a
complex and interacting set of barriers. Rehabilitation in-
terventions aimed at assisting persons with epilepsy “will                               REFERENCES
be effective only if they address joint factors and as early
as possible” (Thorbecke & Fraser, 1997, p. 2219). The          Austin, J. K., & deBoer, H. M. (1997). Disruptions in social function-
present study broadens current understanding of some of           ing and services facilitating adjustment for the child and adult. In
the important factors in employment status and employa-           J. Engel, Jr., & T. A. Pedley (Eds.), Epilepsy: A comprehensive textbook
                                                                  (pp. 2191–2201). Philadelphia: Lippincott-Raven.
bility. There are several implications of this study for re-   Baker, G. A., Jacoby, A., Buck, D., Stalgis, C., & Monnet, D. (1997).
habilitation counseling practice.                                 Quality of life of people with epilepsy: A European study. Epilepsia,
      Seizure control was a significant employment deter-         38, 353–362.
minant, and counselors can help reduce the impact of           Bishop, M. (2002). Barriers to employment among people with
seizures on employment by gaining information about the           epilepsy: Report of a focus group. Journal of Vocational Rehabilitation,
                                                                  17, 281–286
client’s seizure patterns. The relationship between em-        Bishop, M., & Allen, C. (2001). Employment concerns of people with
ployment and medication use highlights the importance             epilepsy and the question of disclosure: Report of a survey of the
of compliance with medication regimens, but counselors            Epilepsy Foundation. Epilepsy and Behavior, 2, 490–495.
must also understand that adverse medication effects can       Brodie, M. J., & Kwan, P. (2002). Staged approach to epilepsy manage-
be significant. These effects may, in some cases, be man-         ment. Neurology, 58(8, Suppl. 5), S2–S8.
                                                               Chaplin, J. E., Wester, A., & Tomson, T. (1998). Factors associated
aged or reduced through consultation with the treating            with the employment problems of people with established epilepsy.
neurologist. The perceived impact of seizures was found to        Seizure, 7, 299–303.
have a significant relationship with employment status,        Chaplin, J. E., Yepez Lasso, R., Shorvon, S. D., & Floyd, M. (1992). Na-
and counselors can, through education and counseling,             tional General Practice study of epilepsy: The social and psycholog-
help reduce this impact.                                          ical effects of a recent diagnosis of epilepsy. British Medical Journal,
                                                                  304, 1416–1418.
      To identify those individuals with epilepsy most at      Cooper, M. (1995). Epilepsy and employment: Employers’ attitudes.
risk for unemployment, and to continue to develop effec-          Seizure, 4, 193–199.
tive rehabilitation interventions, researchers must con-       deBoer, H. M. (1991). Counselling women towards independence. In
tinue studying the problems and barriers faced by persons         M. R. Trimble (Ed.), Women and epilepsy (pp. 19–31). New York:
with epilepsy. The present study represents a further step        Wiley.
                                                               Devinsky, O. (1994). A guide to understanding and living with epilepsy.
toward better understanding, but the continued low em-            Philadelphia: F. A. Davis.
ployment rate among people with epilepsy suggests that
much work remains to be done.                                                                                      (continued on p. 122)

Shared By: