Kentuckians' Attitudes Toward Children with Epilepsy

Epilepsia, 36(10): 1003-1008 Raven Press, Ltd., New York 0 International League Against Epilepsy

Kentuckians’ Attitudes Toward Children with Epilepsy

*?Robert J. Baumann, $John F. Wilson, and $H. Jean Wiese

Departments of *Neurology, ?Pediatrics, and $Behavioral Science, College of Medicine, University of Kentucky, Lexington, Kentucky, U.S.A.

Summary: We explored Kentuckians’ attitudes toward children with epilepsy. Questions compared respondents’ attitudes about children with epilepsy, asthma, hyperactivity, and AIDS. Random digit dialing led to 617 completed interviews. The key questions asked concerned (a) how a pupil with each illness would alter the classroom environment, and (b) how the condition would affect the child’s quality of life (QOL) at age 21 years. Respondents used a 0-10 rating scale (0 = worst, 5 = normal, 10 = best). A dichotomous variable divided respondents into those who rated below and those who rated at or above the norm: 24% predicted a deterioration of the classroom environment with the addition of a pupil with epilepsy (similar to AIDS at 26%); 41% predicted a lessened QOL

Epileptologists will identify with Carlyle’s obser- vation: “[Wlhat is all Knowledge but . . . a product of History . . .” (1). They know that the history of epilepsy is tainted by erroneous beliefs avowing su- pernatural causation, uncleanliness, and evil conta- gion (2). Such “knowledge” has fostered prejudices directed against persons with epilepsy (3,4). Inves- tigators have measured the persistence of such prejudicial beliefs using a variety of techniques. Since 1947, a series of telephone surveys asking about respondents’ attitudes toward persons with epilepsy have shown a steady and pronounced de- crease in the percentage of respondents offering negative views (5) . These data have been inter- preted as suggesting that such prejudices are on the decline. Using the telephone interview technique but using comparative questions, we detect evidence that prejudicial attitudes about persons with epilepsy still persist and that they take varying forms in different communities.

Received May 9, 1994; revision accepted October 6, 1994. Address correspondence and reprint requests to Dr. R. J.

Baumann at Department of Neurology, University of Kentucky, Kentucky Clinic Room E-124, Lexington, KY 40536-0284, U.S.A.

at age 21 years (a worse rating than either asthma or hyperactivity). We created summary indexes, using difference scores between epilepsy ratings and ratings for the other conditions: Relative Educational Distance (RED) and Relative Quality of Life Distance (RQLD) measures. On the RED index, rural and Appalachian respondents showed the greatest prejudice toward pupils with epilepsy. Conversely, we noted the greatest prejudice in RQLD among urban residents (the best educated group). Our data do not support contentions that prejudices against persons with epilepsy are disappearing. Key Words: Epilepsy-Attitudes-Health Surveys-Psycho- social-Appalachia.

METHODS

In all, 617 interviews were completed by telephone in June 1992 as part of an annual health- related survey of Kentucky residents. The survey was conducted by the Survey Research Center at the University of Kentucky. The Survey Research Center, established in 1979, is a university-based center whose faculty and staff have expertise in survey design and administration and extensive expe- rience in state-wide, regional, and national surveys for university faculty, state government, and federal agencies such as the Centers for Disease Con- trol. The telephone survey protocol consisted of computer-assisted telephone interviewing. Quality- control procedures included telephone monitoring, supervisor follow-up verification, postinterview coding and editing, and consistency check analysis of all final data files. The questions specific to this project were included as part of the annual Ken- tucky Health Poll. Each residential telephone line in Kentucky had an equal probability of being called by the random-digit dialing procedure. In addition to use of standard screening approaches, every person called was specifically asked whether the line being called was a residential phone line. The

1003

1004 R . J . BAUMANN ET AL.

trained interviewers questioned the first respondent aged >18 years in the household. The 617 completed interviews required 1,230 calls; 409 respondents refused participation or did not complete the interview, and 204 respondents were ineligible (deaf, too ill, unavailable). Analyzable data were obtained from 50.2% of total calls and 60.1% of calls to eligible respondents.

The questions of interest for this report concerned the respondents’ attitudes toward children with one of four disorders: epilepsy, asthma, hyperactivity, and AIDS. For each of these four disorders, respondents were asked the following questions.

1. Have you ever personally known a child with one of these disorders?

2. Suppose you had an 8-year-old child in school. The learning environment of the classroom can be rated using a scale of 0-10, with 0 being the worst learning environment and 10 being the best possible learning environment. Sup- pose your child’s classroom is rated as 5. What rating would you give if your child’s class included a child who has asthma, (also called wheezing or smothering attacks); hyperactivity (attention deficit disorder: can’t sit still); epilepsy (also called convulsions or fits), or AIDS?

3. Using the same scale of 0-10, where 0 is the worst possible quality of life and 10 is the best possible quality of life (QOL), how would you rate what the QOL will be when they grow up (age 21 years)? Respondents were asked this question for a hypothetical 8-year-old child of their own and for 8-year-old children with each of the following conditions: asthma, AIDS, hyperactivity, and epilepsy. Routine demographic data were also obtained.

Question sets 2 and 3 yielded three variables as- sessing attitudes toward children with epilepsy:

1. The 0-10 rating of the learning environment of a classroom in which 1 child had epilepsy was used as a continuous variable.

2. Because a stated norm (5) was given in the question, a dichotomous attitude variable was formed by dividing the distribution into those who rated the environment below the norm (4) and those who rated it at or above the norm ( 2 5 ) .

3. The 0-10 rating of the QOL at age 21 years of a child with epilepsy was used as a continuous variable.

Because reported attitudes about children with epilepsy may simply reflect general biases about all persons with chronic illnesses, we created two summary indexes that reflected the extent to which an individual’s ratings about epilepsy were more negative or positive than ratings of the other chronic conditions. These indexes were formed by comput- ing difference scores between epilepsy ratings and the ratings of the other conditions, standardizing those difference scores, and then averaging the standardized difference scores. In this way, a Rel- ative Educational Distance measure (RED) was created from responses to question 2 and a Relative Quality of Life Distance measure (RQLD) was created from question 3. For both these measures, whose metric is that of a z-score, an individual with a greater distance score implies a relatively more positive view of epilepsy than the average of the other chronic conditions. A score of 0 on these indexes does not connote a neutral or unbiased attitude. A score of 0 merely indicates that the individual is no more negative (or positive) toward epilepsy than toward the other chronic conditions.

Statistical analysis of the survey results relied on simple chi-square tests dichotomous variables. Mc- Nemar tests for dichotomous variables involving paired data and multiple regression approaches de- rived from the General Linear Model (6). This form of analysis allows multivariate examination of both continuous and categorical predictor variables of our epilepsy attitude measure. Use of cross-product terms in the regression equation also allows examination of the interactive effects of multiple predictors, such as educational level and place of residence, on attitudes toward epilepsy. The Statistical Analysis System (SAS) was used for Regression analyses. The Statistical Package for the Social Sci- ences (SPSS) was used to obtain the McNemar test results.

Age, gender, educational level, rural residence, and residence in an Appalachian county and attitudes toward epilepsy were predictor variables examined to determine the extent to which demographic, social, and cultural factors influence attitudes toward epilepsy. Age was used as a continuous measure. EducatioMl level was dichotomized as less than a high-school degree or at least a high- school degree. The SAS was used for regression analyses. The SPSS was used to obtain the McNe- mar test results.

Rural residence was defined as living in a community with a population <2,500. Residence in an Appalachian community was defined by the counties formally identified as Appalachian counties by the federal government.

Epilepsia, Vol. 36, No. 10, 1995

KENTUCKIANS’ ATTITUDES TO WARD EPILEPSY 1005

RESULTS

Adults’ knowledge of children with chronic conditions and causes of chronic conditions

Less than one third (31.8%) of the adults respond- ing to the survey reported personally knowing a child with epilepsy. This percentage was significantly smaller (p < 0.01) than the percentage of adults knowing a child with asthma (58.8%) or a child with hyperactivity (63%). It was significantly greater (p < 0.01) than the percentage of adults who knew a child with AIDS (1.6%). These percentages are shown in Fig. 1 .

Adults’ knowledge of an epileptic child was un- related to ratings of the effect of that condition on the classroom environment or ratings of the QOL of that child at age 21 years or to the RED or RQLD scores.

Knowing a child with epilepsy was associated (p < 0.05) with being female (60% of females vs. 40% of males), and was marginally associated with greater education (p = 0.06) and with residence in an Appalachian county (37% in Appalachia, 30% in non-Appalachia, p = 0.08). It was not associated with residence in a rural county.

Forty-six percent of adults could list a cause for epilepsy. Twenty-one percent attributed it to hered- ity, and 14% suggested that the cause was a brain disorder. Seven percent listed injury, and 4% of- fered miscellaneous other responses. An adults’ ability to give a cause for epilepsy was not associated with ratings of epilepsy’s impact on QOL or educational environment. 1ndivid.uals who were un- able to give a cause for epilepsy were older (46.5 vs. 44.0 years, p < 0.05), had received less education (11.7 vs. 13.7 years, p < O.Ol), and were more likely to have a rural residence (39% rural, 51% nonrural,

.-- I 86 I

KNOWINQ A CHILD QlVlNQ AN ETIOLOQY

Eplleply Aethma 0 Hyperactlvlty

FIG. 1. Percentage of survey respondents who personally know a child with epilepsy, asthma, hyperactivity, or AIDS and the percentage of respondents who are able to give any etiology for epilepsy (solid columns) asthma (light hatched columns), hyperactivity (dotted columns), and AIDS (dark hatched columns).

ASTHMA 10.3

I I

Percent 0 10 20 30 40 50 60

FIG. 2. Percentage of survey respondents who rate the learning environment of a classroom more negatively than a standard normative classroom when a child with epilepsy, asthma, hyperactivity, or AIDS joins the class.

p < 0.01). Gender and residence in an Appalachian county were not associated with ability to give a cause for epilepsy.

Effect of chronic conditions on educational environment

We asked respondents how the presence of a child with a chronic condition would affect the learning environment of their own child’s elemen- tary school classroom. Our intent was to assess the degree to which attitudes toward epilepsy differed from attitudes toward other chronic conditions. The percentage of adults who rated the presence of a child with asthma, epilepsy, hyperactivity, or AIDS as a negative influence on the classroom (<5 on the visual analogue scale) is shown in Fig. 2. Epilepsy was rated as a negative effect on the classroom significantly (p < 0.01) more often (24.1%) than asthma (10.3%), was rated approximately equal to AIDS, and was rated as having a significantly less (p < 0.01) negative effect than hyperactivity (47.8%).

Effect of chronic conditions on QOL at age 21 years We asked respondents about the QOL at age 21

years of their own 8-year-old child (real or hypothetical) and the QOL at age 21 years of a current 8-year-old children with epilepsy, AIDS, asthma, or hyperactivity. Our intent again was to assess the degree to which attitudes toward epilepsy differed from attitudes toward other chronic conditions. Mean ratings of QOL scores for the four chronic conditions at age 21 years are shown in Fig. 3 . Ex- amination of differences among the chronic illnesses versus ratings for children without chronic illness showed epilepsy to be rated significantly worse than asthma (p < 0.01), marginally worse than hyperactivity (p = 0.054), and significantly better than AIDS (p < 0.001).

Another way of examining QOL assessments is



HYPERACTIVITY - AIDS 4

I I 2.5 3.5 4.5 5.5 6.5

BETTER QOL

FIG. 3. Mean quality of life ratings (QOL) on reaching age 21 years for one's own child and for a child with epilepsy, asthma, hyperactivity, or AIDS.

POORER QOL----- -----__--______

to describe the percentage of adults who rated the QOL at age 21 years of a child with various chronic conditions as worse than the QOL at age 21 years of their own child (Fig. 4). Again, statistically significant differences (p < 0.05) were evident, with epilepsy being rated as having more effects on QOL than asthma and less effects than AIDS.

Moderator variables influencing attitudes toward epilepsy

We examined relationships among age, gender, educational level, rural residence, and residence in an Appalachian county and attitudes toward epilepsy to determine the extent to which demographic, social, and cultural factors influenced attitudes toward epilepsy. Bivariate relationships were examined between these variables and four indicators of attitudes toward epilepsy: the percentage who believed an epileptic child would damage the educational environment, the rating of the QOL at age 21 years of an epileptic child, and the two summary distance measures that reflect the relative position of epilepsy among the four chronic conditions

EPILEPSY 41.2

I I 0 20 40 60 80

% Reporting QOL Decrement at 21 FIG. 4. Percentage of survey respondents who rate the quality of life (QOL) at age 21 years of a child with epilepsy, asthma, hyperactivity, or AIDS as worse than the QOL of their own normal child at age 21 years.

in terms of educational environment and QOL (RED and RQLD). All variables were entered as main-effect terms in a hierarchical multiple regression, followed by two-way interaction terms.

Age was significantly associated only with the RQLD. Older respondents were more likely to place epilepsy in a better QOL position relative to the other chronic diseases than were younger co- horts (p < 0.01). Gender was associated only with the direct ratings of QOL of a person with epilepsy at age 21 years. Women estimated higher QOL ratings for epileptic persons (mean 6.0, SD 1.5) than did males (mean 5.0, SD 1.4) (p < 0.001).

Influence of urban, rural, and Appalachian residence

The size of the respondent's community and whether or not that community was Appalachian had a strong association with all of our attitudinal measures. The effects of these variables on the absolute rating of the effects of epilepsy on the classroom environment is shown in Fig. 5. Metropolitan- area residents rated the classroom effects of epilepsy as significantly less negative than did those in rural or Appalachia counties. Both rural residence (community population <2,500) and Appalachian residence were associated with more negative attitudes toward the impact of a child with epilepsy on the educational environment (p < 0.05). Although Appalachian counties are predominantly rural, these two effects are independent predictors, suggesting that rural and Appalachian contexts operate differently in terms of attitudes toward epilepsy.

For rural and Appalachian responders, educational level was a significant predictor of attitudes toward epilepsy. The interaction between educational level and community was statistically significant for both the direct (p < 0.01) and the summary measure (RED) (p < 0.01) of the educational impact of epilepsy. Figure 6 shows the percentage of re-

METROPOLITAN

NON-RURAL NON-APPAL.

RURAL NON-APPAL.

NON-RURAL APPALACHIA

RURAL APPALACHIA

4 4.5 5 5.5 6 POORER CLASSROOM--------BETTER CLASS

FIG. 5. Mean ratings of the learning environment of a classroom with a child with epilepsy by survey respondents from metropolitan, rural, and Appalachian areas of Kentucky.


KENTUCKIANS’ ATTITUDES TOWARD EPILEPSY 1007

spondents at each educational level in each community category who displayed a negative bias (scores 4) on the educational environment of a classroom with an epileptic child. At high educational levels, the percentage of respondents with a negative bias was equal across rural, urban, and Appalachia lo- cations. At low educational levels, both rural and Appalachia respondents showed greater negative bias toward children with epilepsy.

When we examined the effect of community size and location on the rating of QOL at age 21 years, the negative rural and Appalachian bias disap- peared. Absolute ratings for QOL at age 21 years for a child with epilepsy did not differ significantly by education or community type, (mean levels ranged from 5.67 to 5.98). In a striking reversal, urban residents placed epilepsy lower relative to the other chronic illnesses than did individuals from rural and Appalachian settings (p < 0.05) on the RQLD (Fig. 7).

DISCUSSION Most population-based surveys of attitudes to-

ward persons with epilepsy have used a set of questions similar to those developed by 3 senior neurol- ogists in 1949 and first administered by George Gal- lup (5). These questions were designed to be answered either yes or no. Since 1949, many of them have been used repeatedly in the United States and adapted for use elsewhere (7-9). The questions inquire directly about misconceptions and prejudices regarding persons with epilepsy. Over the years, fewer and fewer respondents have admitted to a belief in these stereotypes. For example, between 1949 and 1987, the percentage of respondents who agreed that epilepsy is a form of insanity decreased from 12 to 3% and the percentage who would allow their child to play with a child with epilepsy increased from 57 to 89%; between

METROPOLITAN

NON-RURAL NON-APPAL.

RURAL NON-APPAL.

NON-RURAL APPALACHIA I I I 0 10 20 30 40 50 Percent Reporting Negative Effect

= High Education Low Education

FIG. 6. Percentage of respondents at different educational levels from metropolitan, rural, and Appalachian areas who rate the learning environment of a classroom with a child with epilepsy as poorer than that of a standard classroom.

METROPOLITAN -

NON-RURAL NON-APPAL. -

RURAL. NON-APPAL. -

NON-RURAL APPALACHIA -

RURAL APPALACHIA

-0.1 0 0.1 0.2 0.3 0.4 0.5 Relative QOL Distance Score (RQLD)

FIG. 7. Mean Relative Quality of Life Distance ratings (RQLD) comparing the quality of life at age 21 years of individuals with epilepsy with ratings for other chronic illnesses by survey respondents from metropolitan, rural, and Appala- chian regions of Kentucky.

1949 and 1979, the percentage who believed that persons with epilepsy should not be employed had decreased from 33 to 9%.

Our survey invited interviewees to give a graded response (from 0 to 10). We asked them to envision the learning environment in their child’s classroom as modified by the addition of various students, in- cluding one with epilepsy. We then asked them to rate the QOL for these students as young adults. Deviations from the outcome predicted for their own (normal) child are indirect indicators of their beliefs. We did not inquire about the existence of specific misconceptions and prejudices.

Less than 33% of our respondents had personal knowledge of a child with epilepsy, in contrast to 49-63% of national respondents who were asked whether they had ever known “anyone with epilepsy.” Our respondents were more likely to have personal knowledge of a child with hyperactivity (63%) or asthma (59%). The respondents’ knowledge of an affected child did not correlate with the ratings on the effect of that condition on the classroom environment or on eventual QOL. Our respondents were similar to national survey respondents in that 46.5% were willing to offer a cause for epilepsy (51% in the national sample).

Twenty-four percent of our respondents envi- sioned a deterioration in ‘the classroom environment if a pupil with epilepsy joined the class. This percentage is similar to the percentage (26%) who predicted a deterioration of the environment if a pupil with AIDS was enrolled. An even greater number of respondents (41%) predicted a lessened QOL at age 21 years for the child with epilepsy. This rate of adverse response appears to be greater than that recorded in the more recent Gallup surveys, which, for example, recorded a negative response rate of



9% toward employment for adults with epilepsy. Furthermore, the QOL for the young adult with epilepsy was scored significantly worse (p < 0.05) than that of the young adult with asthma and marginally (p < 0.10) worse than that of the young adult with hyperactivity.

Like other large geographic entities, Kentucky is a heterogeneous state. When we analyzed our data by the size and location of the respondent’s community, the relationship between these factors and the reported ratings was striking. When evaluating the classroom environment, urban interviewees viewed the effects of a pupil with epilepsy as significantly more positive than did rural non- Appalachian or Appalachian respondents. Even though Kentucky’s Appalachian counties are predominantly rural, the two locales were independent predictors. On the other hand, when predicting adult outcomes, metropolitan residents were the most negative, and Appalachian residents the most positive, with rural non-Appalachian residents again in an intermediate position. These data strongly suggest the presence of different cultural communities with differing aggregate attitudes toward persons with epilepsy. In the Appalachian mi- lieu, a child with epilepsy apparently is viewed as more threatening to the classroom environment than is a child with epilepsy in the urban setting. However, in the urban community, the presence of epilepsy is rated as more threatening to the child’s future than ratings from individuals living in rural and/or Appalachian communities. Given the infre- quent occurrence of seizures and the high rate of success of modern anticonvulsant medications, both attitudes appear to reflect stereotypical beliefs or prejudice.

A possible explanation for these differences in attitudes toward a schoolchild with epilepsy and an adult with epilepsy might lie in different subcultur- ally patterned attitudes toward formal education. Evidence shows that in the Eastern Kentucky sub- culture, the activities which contribute to success in the classroom are not necessarily the same activities and qualities most valued in the adult population. Data from Abbott’s (10) work with adolescent subjects from Eastern Kentucky indicate that both males and females “. . . place greater emphasis on moral aspects of a person’s behavior, attitudes toward authority, and qualities of the person like having a good temperament, being loving and caring, wise and sociable . . .” than on formal education (10). An explanation for our findings consistent with this view would be that for our Appalachian respondents the stereotype of an individual with epilepsy holds that this person might well have difficulty in

the controlled and individually competitive environment of the classroom but that the condition would not necessarily affect that individual’s achievement of a high-quality adult life, where evidence of inner qualities is most valued.

Our findings regarding the patterns of attitudes in Kentucky communities of different size and cultural background are new and certainly require replication because they were not a priori hypotheses of the researchers. Because Kentucky, as compared with the rest of the United States, is a relatively homogeneous state with few recent immigrants and a small minority population, we believe it is of interest if replication shows that belief systems vary among Kentucky communities. Although our find- ing does not negate the hypothesis that education will work to eliminate untoward beliefs about persons with epilepsy, it does suggest that educational programs will need to be individualized for various communities. Moreover, the evidence that prejudicial views remain in the most highly educated (urban) communities-communities that show the low- est rate of prejudice on the Gallup-style surveys- also suggests that the continual increase in the educational level of Americans will not alone suf- fice to eliminate these types of prejudicial attitudes.

Acknowledgment: This work was supported in part by a grant from the Survey Research Center, University of Kentucky, and by the Vice Chancellor for Research, Uni- versity of Kentucky College of Medicine.

1 .

2.

3 .

4.

5.

6.

7.

8.

9.

10.

REFERENCES Carlyle T. Critical and miscellaneous essays. New York: Appleton, 1878:220. Heitmann B. Representations of epilepsy as the healing of the possessed. In: Liiders H, ed. Epilepsy surgery. New York: Raven Press, 1991:l-5. Engel, J Jr. Seizures and epilepsy. Philadelphia: F. A. Davis, 1989. Jonsson A, Kuhlman P, Dam M. Epilepsy: the effect of in- formation on knowledge and attitudes. In: Dam M, Gram L, Penry JK, eds. Advances in epileptology: the XIZ epilepsy international symposium. New York: Raven Press, 1981:

Caveness W, Gallup G Jr. A survey of public attitudes toward epilepsy in 1979 with an indication of trends over the past thirty years. Epilepsia 1980;21:509-18. Cohen J , Cohen P. Applied multiple regressionlcorrelation for the Behavioral Sciences. Hillsdale, NJ: Lawrence Earl- baum, 1983:162-5. Canger R, Cornaggia C. Public attitudes toward epilepsy in Italy: results of a survey and comparison with U.S.A. and West German data. Epilepsia 1985;26:221-6. Iivanainen M, Uutela A, Vikkumaa I . Public awareness and attitudes toward epilepsy in Finland. Epilepsia 1980;21:413-23. Jensen R , Dam M. Public attitudes toward epilepsy in Den- mark. Epilepsia 1992;33:459-63. Abbott S . Symptoms of anxiety and depression among East- em Kentucky adolescents: insights from comparative field- work. In: Health in Appalachia: proceedings from the 1988 Conference on Appalachia. Lexington, KY: University of Kentucky Press, 1989: 102-15.

183-8.

Epilepsia, Vol. 36, NO. 10, 1995

Documents

Kentuckians' Attitudes Toward Children with Epilepsy