Public Knowledge, Private Grief: A Study of Public Attitudes to Epilepsy in the United Kingdom and Implications for Stigma

Authors


Address correspondence and reprint requests to Dr. A. Jacoby at Department of Public Health, Whelan Building, The Quadrangle, Brownlow Hill, Liverpool L69 3GB, U.K. E-mail: ajacoby@liv.ac.uk

Abstract

Summary: Purpose: For many people with epilepsy, the continuing social reality of their condition is as a stigma, thus representing a source of much private grief. To understand fully the nature of epilepsy stigma, it is important to examine attitudes and beliefs of not just the “targets” but also of the “perceivers” of stigma. Perceivers may hold erroneous beliefs and stereotypes that lead them to have negative expectations of people affected by epilepsy. This study examined levels of knowledge and attitudes among perceivers of epilepsy stigma in the U.K.

Methods: A random sample of >1,600 members of the general public was interviewed as part of the bimonthly UK Omnibus Survey. The response rate was 62%.

Results: One fourth of informants knew someone with epilepsy and one half had witnessed a seizure. One half agreed that people with epilepsy are treated differently by others, and themes of exclusion, restriction, and nonnormality were commonly cited as reasons for this. Epilepsy ranked second in a range of health conditions, in terms of which would cause them greatest concern if informants had to work with someone so affected. Responses to a series of attitude statements indicated that most held highly favorable attitudes; but one fifth agreed with the statement that people with epilepsy have more personality problems than do others. Responses were influenced by informants' sociodemographic characteristics.

Conclusions: The survey revealed attitudes and knowledge gaps that have the potential for discriminatory behavior. There are implications for how public education campaigns should be implemented, and who should be targeted.

For many people with epilepsy, the continuing social reality of their condition is as a stigma. Both historically and cross-culturally, and with only rare exceptions, epilepsy has been represented as such, and so has been the source of much private grief to those affected by it. Recent research has shown that the stigma people with epilepsy feel contributes to: increased rates of psychopathology, reduced social interactions, reduced social capital, and reduced quality of life (QoL) in both developed and developing countries (1–10). Despite a rapidly changing clinical reality, at least for those in the developed world, it has been argued that the legacy of the old myths about epilepsy linger on, so that negative attitudes toward it prevail, and its stigma remains “real and serious” (11). Goffman (12) described stigma as involving the definition by a social group of some attributes as ordinary and natural and others as somehow “deeply discrediting.” Those possessing only the former are the so-called “normals”; those possessing the latter become the “stigmatized.” Other important elements of Goffman's theory of stigma include the view of those who are stigmatized as passively accepting societal definitions of difference; the degree of visibility of this difference as critical to its management; and the ineradicable and irreversible nature of stigma. In relation to epilepsy specifically, Scambler (13,14) drew one further important theoretical distinction, that between “felt” and “enacted” stigma. In Scambler's taxonomy, felt stigma concerns the shame associated with being epileptic and the fear of encountering stigma, whereas enacted stigma refers to actual episodes of discrimination against people with epilepsy solely on the grounds that they have epilepsy. Scambler noted that the former is not necessarily dependent on the latter: indeed, almost everyone in his study reported suffering from felt stigma, even though few were able to cite examples of discrimination enacted against them personally. Nonetheless, Trostle (15) noted, “to have epilepsy is to open oneself to the full force of past and contemporary social prejudice and misunderstanding.”

This emphasis on the issue of prejudice and misunderstanding links to the work of recent stigma theorists (16,17) who have reframed the concept of stigma to focus less on the person who is stigmatized and more on those who do the stigmatizing. Thus for example, Link and Phelan (18) argued that stigma is the product of five interrelated components, the first of which is the definition of particular human differences as “socially relevant” by those holding the social, political, or economic power to do so. These relevant differences are then linked to negative stereotypes, leading to social distancing and separation and, subsequently, to status loss and discrimination. This widening focus on not just the “targets” but also on the “perceivers” of stigma is an important one, because research from these later theorists has highlighted that perceivers often hold erroneous beliefs; and the stereotypes they draw on lead them to have particular (and generally negative) expectations of the group they see as stigmatized. Stereotyped assumptions about people with conditions such as epilepsy are inherent in many cultures, based in old superstitions and beliefs and, it has been argued, often reinforced through disabling imagery in literature and the media (19).

Much of the research on epilepsy stigma has addressed the experiences of the “targeted” and the issue of “felt” stigma (5,6,8,13, 20–23); somewhat less has been directed at deconstructing the beliefs of the “perceivers” and the issue of “enacted” stigma. However, some notable reports exist on public attitudes to epilepsy from both the developed and developing countries, indicating that both temporal and sociocultural factors continue to be important predictors. In relation to the former, the most substantial evidence in the developed world comes from a series of studies (24) conducted in the United States at 5-yearly intervals from 1949 to 1979. Based on responses of the lay public to a series of questions (many replicated in subsequent studies, including the one we report here), the results suggest that although levels of knowledge about epilepsy in the United States have stayed about the same, attitudes toward it have improved steadily and significantly. So, for example, in response to a question asking whether they would object to their child playing with a child with epilepsy, 24% of U.S. respondents in 1949 answered “yes” compared with only 6% in 1979. Small but significant temporal improvements in public attitudes also were noted by Mirnics et al. (25), who compared public attitudes to epilepsy in Hungary across a 6-year period during which national initiatives to promote the integration of people with epilepsy were instigated; and by Novotna and Rektor (26), who examined changes in public attitudes to epilepsy in the Czech Republic over a 16-year period.

Differences in public attitudes to epilepsy cross-culturally are highlighted by studies conducted from the early 1980s on in a number of European and other developed countries (25–30). The authors of these studies have, for example, reported widely varying levels of public affirmation for the question about children with epilepsy cited earlier, ranging from 27% for Italy in the 1980s (28) to 6% for the Czech Republic in the early 2000s (26). Further evidence about the “undesirable differentness” of epilepsy in the developed world comes from studies such as the street surveys carried out in the United Kingdom in the late 1970s and early 1980s (31,32), in which commonly articulated views were of people with epilepsy as excitable and aggressive, and even potentially violent, retarded, antisocial, and physically unattractive. Epilepsy has been found to engender much more hostility than comparator chronic health conditions in the studies reported by Bagley (33) and Baumann et al. (34) and despite the attitudinal improvements noted by Caveness and Gallup (24), Gouvier et al. (35) documented persisting misconceptions about epilepsy, particularly among their younger and less-educated respondents living in the United States.

It was to examine the nature and extent of continuing misperceptions and negative attitudes toward epilepsy in the developed world that the project reported here was directed.

The broad objectives of the study were:

  • • To explore public attitudes to and knowledge about epilepsy in the U.K.;
  • • To compare findings to those from earlier studies of lay attitudes carried out here and elsewhere in the developed world; and
  • • To identify foci for future public-attitude campaigns.

STUDY DESIGN AND METHODS

The study involved three separate arms. In the first arm, for which results are reported here, members of the U.K. general public were surveyed by using face-to-face interviews. In the second and third arms, results of which are reported elsewhere, questionnaires were mailed to samples of U.K. employers and to friends and/or relatives of people with epilepsy.

The survey of the general public was carried out as part of the U.K. Omnibus Survey, a nationwide survey undertaken bimonthly by the U.K. Office of National Statistics (ONS). The authors (A.J., J.G., G.B.) worked with ONS staff to develop a question module relating to knowledge of and attitudes toward epilepsy, which involved 16 precoded and two open questions (see Appendix). This module was incorporated into the November 2001 survey round.

The sampling frame for the Omnibus Survey is the U.K. Postcode Address File, which includes all private household addresses in Great Britain. At each survey round, 30 addresses are selected randomly from each of 100 postal sectors (in turn selected with probability proportional to size and stratified by region, housing type, and socioeconomic status of household reference persons). Face-to-face interviews are then conducted with one person aged 16 and older at each address. Because only one person in each sampled household is interviewed, those in households containing few adults have a better chance of selection than do those in larger households. ONS applies a weighting factor to correct for this unequal probability (calculated as [no. adults in sampled household/average no. adults per household]), the base being adjusted back to the number of individuals interviewed. Our analyses were conducted by using the survey commands in STATA version 8. Analyses incorporate weights to adjust for unequal probabilities of selection and primary sampling units to account for clusters.

All interviews are carried out by a trained field force. Advance letters are sent to all addresses before interview, giving a brief account of the survey, and interviewers make a minimum of three calls at each address before categorizing it as a noncontact. Interviews for the Omnibus Survey are computer-assisted (CAPI), with responses recorded on a laptop computer. The interviewers record verbatim all open question responses.

RESULTS

In November 2001, of the 3,000 selected addresses, 256 proved ineligible. Of a potential 2,744 individuals at the remaining addresses, 815 refused, and 235 could not be contacted. Data collection extended over a 3-week period from late November to early December. The response rate of 62%, although lower than normally achieved in the Omnibus Survey, is similar to that achieved in previous survey rounds conducted at that time of year. Forty-five percent of the 1,694 informants were men; 35% were aged between 25 and 44 years. Fifty-seven percent had left school at or below the present-day U.K. minimum school-leaving age (16 years), and a similar percentage (65%) had no or only the lowest level of formal educational qualifications. Fifty-seven percent had been in paid employment in the 7-day period preceding the interview, of whom 33% were classified as working in semiroutine or routine occupations, and a further 32%, in lower managerial and professional or supervisory technical occupations. Eighty-seven percent of informants gave their ethnic origin as White British, and a further 4.6% as other White. The characteristics of informants were similar to those of the U.K. general population (see Table 1), although men and those in the lower age bands were slightly underrepresented and those currently in paid employment slightly overrepresented. Two percent of the sample declared themselves as having epilepsy, and a further 24% reported having a relative or close friend with epilepsy. Although the remainder were unfamiliar with the condition, 57% of informants reported having seen someone have a seizure.

Table 1. Sociodemographic characteristics of informants in the November 2001 UK Omnibus Survey (n = 1694)
 Sample
population
UK
population
  1. Data source: UK Office for National Statistics (UK Census 2001; UK General Household Survey 2002: UK Labour Force Survey); no directly comparable figures available for educational status.

  2. aFigures for age bands 16–44 and 45–64 combined in UK ONS sources

  3. bFigures for higher professional/managerial combined in UK ONS sources

  4. cFigures for white British/other white combined in UK ONS sources

% Males44.648.6 
Age (% per age band; yr)
 16–2412.2 51.2a
 25–4435.0 
 45–5415.9 30.0a
 55–6415.0 
 65–7412.410.0 
 75 and older 9.6 8.7 
Age left full-time education (%; yr)
 Up to 1413.4 
 15–1859.9 
 19–2518.0 
 25 and older 3.8 
 Still in full-time education 4.5 
 No education 0.3 
% in paid employment in last 7 days57.047.1 
Socioeconomic grouping (%)
 Large employer/higher managerial 3.610b  
 Higher professional 5.8 
 Lower managerial/professional23.122  
 Intermediate13.313  
 Small employer/own account worker 7.1 8  
 Lower supervisory and technical 9.110  
 Semiroutine19.018  
 Routine13.615  
 Not classified 5.4 5  
Ethnic origin (%)
 White British87.492.1c
 Other white 4.6 
 Mixed white/black Caribbean/African/Asian 1.1 1.2 
 Asian subcontinent 3.6 4.0 
 (Indian/Pakistani/Bangladeshi)
 Black/black British 1.8 2.0 
 Chinese 0.6 0.4 
 Any other 1.0 0.4 

Informants were asked to estimate how many people of 1,000 would have epilepsy. Estimates of prevalence ranged from 0 to 800 per 1,000, with 15% correctly estimating it as between six and 10 per 1,000; and a further 22% as between 0 and five per 1000. Epilepsy was described as “a physical health problem” by 31%, “a mental health problem” by 19%, and as “both a physical and a mental health problem” by 34% of informants (a further 4% described it as “neither; ” and 11% said they did not know). Informants appeared reasonably well informed about the causes of epilepsy, although 16% attributed it to stress (see Table 2). Their descriptions of the features of an epileptic seizure suggested that the majority thought of them in the stereotypical terms of the sudden and dramatic tonic–clonic seizure (see Table 2). Seventy-nine percent of informants agreed with the statement that “doctors can treat epilepsy successfully with drugs in most cases.”

Table 2. Causes of epilepsy and signs of seizures

Causes of epilepsy (a)
%
saying

Signs of seizures (b)
%
saying
  1. Bases on which the percentages were calculated were (a) n = 1,549, (b) n = 1,538; and exclude small numbers of individuals who did not answer the questions.

Infection 2Loss of consciousness20
Alcohol/drug abuse14Convulsions/shaking79
Old age 2Blank spell17
Other illness/disease22Repeated movements11
Congenital/birth problem57Dropping to ground50
Stress/pressure16Rapid blinking 9
Mental illness13Choking13
Nervous system problem50Vomiting10
Brain disorder57Fainting 7
Accident44Other signs26
None of these 1None of these 3

In answer to a question asking informants to consider whether people with epilepsy are treated differently by society from those without epilepsy, 51% agreed that they were. When asked to say in what ways, informants suggested that the difference was manifested as exclusion or restriction from areas of life such as driving, employment, and insurance; avoidance by others, and, if not avoidance, a certain level of wariness; and as though people with epilepsy were “disabled” by their condition (see Box 1). Informants attributed these responses to epilepsy to fear, ignorance, and misunderstanding, concerns around issues of safety and, echoing Goffman, the outcome of societal ideas about normality and difference.

Table BOX 1. . Informants' views about how and why people with epilepsy are treated differently
How:
“Epileptics are excluded from some employment, I suppose, because of the fear they might be injured or put other people in danger in the course of their work.”
“People would avoid them and not want to ask them to do anything or go anywhere, in case something went wrong and they wouldn't be able to do it properly.”
“They are classed as disabled.”
“ I think they [other people] would be fearful of a time when they would have a seizure, so would not socialize or befriend them.”
“They are identified as being out of the normal, labelled as being abnormal.”
“They're known as freaks and people are wary of them.”
Why:
“They're not like us; it's embarrassing; because it's a ‘brain thing’ and people are nervous of that; associate it with mental illness.”
“Because of them not being in full control of themselves at all times, so they are different in a sense—they can't help it—it is just something they suffer from which stops them being the same as the rest of us.”
“Fear and ignorance—it can be hard to tell if a person in a seizure is drunk or on drugs—and people are afraid to get involved as a result.”
“They are a health risk to the public.”

Informants were asked to say from which of a number of specified occupations people with epilepsy should be barred. Not surprisingly, large percentages thought they should not be employed as drivers of heavy goods vehicles or in the uniformed services. Somewhat more surprisingly, significant numbers also thought people with epilepsy should be barred from professions such as teaching and nursing, and a small minority thought they should not work as shop assistants and solicitors (Table 3). When asked to rank epilepsy alongside a range of other health conditions, in terms of which would cause greatest concern if they themselves had to work with someone so affected, informants ranked epilepsy second, after stress/depression, with heart attack ranking third (Table 4). Overall, epilepsy was ranked as the condition causing greatest concern by one fifth (21%) of informants, and as the condition causing least concern by only 1%. Last, informants were asked to complete a series of six attitude statements, some replicating statements used in earlier studies of public attitudes toward epilepsy. As shown in Table 5, most expressed highly favorable attitudes; but 22% agreed with the statement that people with epilepsy have more personality problems than do those without; and 90% were of the opinion that “it is frightening to see someone having an epileptic seizure.”

Table 3. Views about employment of people with epilepsy

Occupation
% saying “PWE should
not be employed”
  1. Base on which percentages were calculated excludes 75 individuals who did not answer the question.

  2. PWE, people with epilepsy.

Solicitor 5
Shop assistant 6
Teacher20
Nurse28
Factory worker33
Police officer46
Armed forces57
Fire fighter72
Lorry driver87
None of these 8
Table 4. Ranking of level of concern associated with health conditions
Condition causing greatest concern:% selecting
  1. aSpecified in the question as involving a work absence of 3 months in last year.

  2. bSpecified in the question as associated with having two seizures in the last year.

  3. cSpecified in the question as occurring a year ago.

  4. Base on which percentages calculated excludes 161 individuals who did not answer the question.

Stress or depressiona34
Epilepsyb19
Heart attackc17
Condition necessitating wheelchair use13
Facial disfigurement 9
Age older than 50 years 1
None cause concern 6
Table 5. Expressed attitudes toward people with epilepsy
StatementResponse (%)a
Strongly
agree
Tend to
agree
Neither agree
nor disagree
Tend to
disagree
Strongly
disagree
Don't
know
  1. aFor each item, percentages were calculated excluding a small number of informants who did not provide any answer. Base n for each item was 1,640.

  2. bChildren with epilepsy.

a) PWE are as intelligent as everyone else712421-2
b) PWE have more personality problems than others 31920 28 228
c) CWEb should be allowed to play with other children73222111
d) PWE can be as successful as other people in their chosen career61333102
e) PWE can lead a normal life48396502
f) It is frightening to see someone having an epileptic seizure36545212

Tables 6 and 7 present results from the subgroup analysis, which showed that both levels of knowledge about and attitudes toward epilepsy were influenced by the characteristics of informants. In summary, knowledge was greater, and more favorable attitudes tended to be expressed by those familiar with epilepsy, women, those in the middle-age bands, and those from higher social classes. For example, agreement that people with epilepsy can be successfully treated was higher for familiars (86% cf. 77%), women (85% cf. 71%), those in the middle-age bands (87% aged 55–64, cf. 63% aged 16–24 and 75% aged 75 and older); and those of higher social class (83% SCI cf. 77% SCV, and 64% unclassifiable). Conversely, those unfamiliar with epilepsy, men, those at the extremes of age, and those from lower social classes tended to express a more negative position: for example, older people were more likely to agree that people with epilepsy are not as intelligent (14% older than 75 years, cf. 5% or fewer aged 16–64), cannot be successful in their careers (16% older than 75 cf. 3–6% aged 16–64), and cannot lead a normal life (26% older than 75, cf. 8–12% aged 16–64). The pattern for level of education was not a consistent one, although there was a suggestion in the data that more negative attitudes tended to be expressed by those without any formal educational qualifications (a finding that may be confounded by age effects).

Table 6. Responses to knowledge items by informant characteristics
 
Prevalence>5%
Epilepsy is a
mental problem
Seizures are
frightening to see
E cannot be
treated successfully
PWE are treated
differently
  1. Values expressed as percentages and 95% confidence intervals.

  2. aDenotes p < 0.05 calculated by using the Pearson χ2 statistic corrected for the survey design using the second order correction of Rao and Scott (47).

  3. bPeople are grouped into six socio-economic classes where 1 = managerial and professional groups; 2 = intermediate occupations; 3 = small employers and own account workers; 4 = lower supervisory and technical occupations; 5 = semi-routine and routine occupations; and 6 = not classifiable (i.e. people who have never worked, are long-term unemployed, full-time students, or where their occupations are not stated or inadequately described).

Familiarity with epilepsy
 No33.0 (29.5, 36.7) 52.5 (49.1, 55.8) 89.6 (87.4, 91.5) 23.3 (20.6, 26.2)a48.0 (44.8, 51.3)a
 Yes38.6 (32.2, 45.5) 55.7 (50.3, 61.0) 91.0 (86.8, 93.9) 14.4 (10.6, 19.4) 56.5 (50.8, 61.9) 
Sex
 M24.8 (20.7, 29.4)a58.5 (54.6, 62.3)a89.6 (86.6, 92.0) 28.7 (25.1, 32.6)a43.8 (39.7, 48.0)a
 F42.5 (38.0, 47.2) 49.0 (45.6, 52.5) 90.2 (87.9, 92.2) 15.1 (12.5, 18.1) 55.1 (51.3, 58.9) 
Age (yr) 
 16–2442.0 (31.9, 52.8)a66.5 (57.7, 74.2)a94.2 (88.8, 97.1) 36.6 (27.8, 46.3)a31.9 (23.9, 41.0)a
 25–4437.9 (33.7, 42.3) 56.5 (52.0, 60.8) 87.8 (84.7, 90.4) 22.6 (18.4, 27.4) 50.0 (45.8, 54.3) 
 45–5430.1 (23.8, 37.3) 45.7 (38.7, 52.8) 92.0 (87.6, 94.9) 15.1 (10.8, 20.7) 59.1 (52.2, 65.7) 
 55–6432.7 (26.6, 39.5) 48.9 (42.0, 55.9) 90.9 (86.5, 93.9) 12.9 (9.4, 17.3) 60.5 (53.4, 67.2) 
 65–7425.8 (19.2, 33.7) 50.1 (43.0, 57.3) 89.8 (84.9, 93.3) 17.8 (13.0, 23.9) 52.2 (44.5, 59.7) 
 >7531.5 (23.0, 41.6) 48.8 (41.7,56.0) 87.9 (81.8, 92.1) 24.6 (18.9, 31.3) 39.5 (32.6, 46.9) 
Socio-economic classb
 123.2 (19.4, 27.6)a49.8 (45.3, 54.3) 92.7 (89.5, 95.0)a17.2 (13.8, 21.2)a59.1 (54.5, 63.5)a
 236.9 (28.8, 45.9) 49.6 (42.1, 57.2) 91.6 (87.4, 94.5) 18.1 (13.0, 24.7) 49.3 (41.0, 57.6) 
 327.2 (18.8, 37.5) 45.8 (36.2, 55.8) 87.9 (79.7, 93.1) 19.1 (12.7, 27.6) 37.5 (28.7, 47.2) 
 429.3 (21.5, 38.6) 39.5 (32.4, 47.0) 87.7 (81.1, 92.2) 26.8 (20.4, 34.3) 55.7 (47.5, 63.7) 
 549.0 (43.5, 54.6) 45.4 (40.6, 50.4) 88.8 (85.4, 91.5) 23.0 (18.8, 27.7) 46.0 (40.8, 51.2) 
 Not classified35.6 (23.5, 49.9) 41.8 (30.1, 54.4) 81.5 (69.6, 89.4) 36.4 (25.0, 49.6) 30.0 (20.0, 42.4) 
Higher Education
 Degree +18.4 (13.4, 24.7)a47.3 (41.3, 53.3)a92.7 (87.8, 95.7)a21.2 (15.8, 27.9)a59.5 (53.1, 65.6)a
 Higher ed24.5 (16.9, 34.0) 44.9 (36.7,53.4) 91.6 (84.5, 95.7) 11.5 (6.8, 18.6) 67.5 (58.4,75.5) 
 A levels22.1 (15.1, 31.1) 63.2 (52.7, 72.6) 86.8 (78.5, 92.1) 32.0 (22.7, 43.0) 44.5 (34.3, 55.2) 
 ONC/BTEC31.9 (21.9, 43.9) 55.9 (44.5, 66.6) 89.0 (79.4, 94.4) 33.1 (21.8, 46.8) 51.0 (37.8, 64.1) 
 O level or equiv44.3 (38.5, 50.3) 58.9 (52.7, 64.8) 93.4 (89.2, 96.1) 17.7 (13.2, 23.4) 46.9 (40.8, 53.1) 
 GCSE(D-G) 50.0 (38.4, 61.6) 59.6 (47.7, 70.5) 92.1 (84.4, 96.2) 24.1 (15.9, 34.9) 48.1 (37.5, 58.9) 
 Other34.6 (26.5, 43.8) 51.6 (40.5, 62.7) 93.0 (87.0, 96.4) 19.5 (12.9, 28.4) 51.3 (40.2, 62.4) 
 No formal q40.2 (33.8, 47.0) 51.2 (46.1, 56.4) 86.1 (82.5, 88.9) 20.6 (16.8, 25.1) 44.6 (40.0, 49.4) 
Table 7. Responses to attitude statements by informant characteristics
 PWE are not
as intelligent
as others
PWE have more
problems personality
problems than others
CWE should not
be allowed to
play with others
PWE cannot be as
successful in their
chosen career
PWE cannot
lead a normal
life
  1. Values are expressed as percentages and 95% confidence intervals.

  2. aDenotes p < 0.05 calculated by using the Pearson χ2 statistic corrected for the survey design by using the second-order correction of Rao and Scott (47).

  3. bPeople are grouped into six socio-economic classes where 1 = managerial and professional groups; 2 = intermediate occupations; 3 = small employers and own account workers; 4 = lower supervisory and technical occupations; 5 = semi-routine and routine occupations; and 6 = not classifiable (i.e. people who have never worked, are long-term unemployed, full-time students, or where their occupations are not stated or inadequately described).

Familiar with epilepsy
 No6.4 (4.9, 8.4)a21.2 (18.3, 24.5) 5.5 (4.1, 7.3) 7.3 (5.6, 9.3)a13.9 (11.9, 16.3) 
 Yes3.5 (2.0, 6.1) 24.4 (20.2, 29.2) 3.1 (1.6, 6.0) 4.1 (2.3, 7.0) 10.7 (7.2, 15.6) 
Sex
 M6.7 (4.8, 9.4) 22.2 (18.9, 25.9) 6.0 (4.1, 8.7) 7.6 (5.6, 10.2) 15.3 (12.6,18.5)a
 F4.9 (3.6, 6.8) 21.9 (18.8, 25.3) 4.0 (2.7, 6.0) 5.6 (4.1, 7.5) 11.4 (9.2, 14.0) 
Age (yr) 
 16–243.1 (1.1, 8.1)a12.7 (7.4, 21.2)a 6.4 (2.9, 13.2) 3.6 (1.3, 9.5)a11.6 (7.3, 18.1)a
 25–444.4 (2.8, 6.9) 11.5 (8.9, 14.9) 3.4 (1.8, 6.3) 4.5 (2.6, 7.6) 12.5 (9.7, 16.1) 
 45–544.1 (2.1, 8.0) 24.6 (18.5, 31.9) 4.8 (2.4, 9.1) 5.8 (3.2, 10.1)  8.7 (5.3, 13.9) 
 55–645.0 (2.9, 8.6) 30.0 (24.3, 36.1) 3.3 (1.5, 7.1) 4.4 (2.6, 7.3) 11.3 (7.6, 16.5) 
 65–74 8.4 (5.0, 13.7) 34.2 (27.8, 41.2)  6.3 (3.8, 10.2) 11.2 (7.5,16.4) 14.1 (10.0, 19.5) 
 >7514.2 (9.9,20.0) 39.5 (32.6, 46.7)  9.7 (6.1, 15.0) 15.6 (10.7,22.1) 26.3 (20.4, 33.2) 
Socio-economic classb
 12.9 (1.8, 4.6)a21.2 (17.7, 25.2) 2.6 (1.5, 4.4)a4.1 (2.7, 6.1) 12.5 (9.9, 15.7) 
 21.5 (0.5, 4.2) 18.9 (13.1, 26.3) 2.7 (1.0, 7.4)  5.0 (2.1, 11.5) 13.7 (9.4, 19.4) 
 3 7.4 (3.9, 13.8) 21.9 (15.1, 30.5)  4.2 (1.6, 10.4)  7.0 (3.4, 13.8) 14.9 (9.0, 23.6) 
 4 6.9 (3.9, 11.9) 20.7 (14.8, 28.0) 5.1 (2.6, 9.8)  6.5 (3.4, 12.0) 13.4 (8.5, 20.5) 
 5 8.4 (5.5, 12.4) 24.7 (20.5, 29.4)  6.9 (4.8, 10.0)  8.7 (6.2, 12.1) 13.3 (10.1, 17.2) 
 Not classified13.9 (7.1, 25.5) 21.2 (11.1, 36.6) 13.2 (6.9, 23.9) 11.3 (5.4, 22.1) 11.9 (6.0, 22.4) 
Higher Education
 Degree2.9 (1.4, 5.9)a18.1 (13.5, 24.0)a 4.3 (1.9, 9.3)a3.9 (2.2, 6.8)a13.4 (9.8, 18.2) 
 Higher ed2.6 (0.9, 7.3) 18.5 (12.0, 27.5)  0.9 (0.1, 6.2) 1.3 (0.3, 5.5) 7.9 (4.3, 14.2) 
 A'levels 4.3 (1.5, 11.7) 16.2 (9.8, 25.8)  3.0 (0.8, 10.4)  6.0 (2.4, 14.1) 10.3 (5.2, 19.1) 
 ONC/BTEC1.4 (0.4, 5.3) 22.1 (12.8, 35.3)  2.8 (0.7, 10.6)  4.8 (1.7, 12.8) 13.8 (7.0, 25.4) 
 O'level or equiv4.3 (2.4, 7.4) 13.7 (9.0, 20.2)  3.9 (1.9, 7.7) 4.1 (2.0, 8.0) 9.0 (6.0, 13.5) 
 GCSE(D-G)1.5 (0.3, 6.3) 20.2 (12.4, 31.1)  1.5 (0.3, 6.2)  5.4 (2.0, 14.1) 15.3 (8.9, 25.1) 
 Other 6.1 (2.6, 13.5) 25.1 (18.0, 33.9)  3.3 (1.2, 8.6)  7.4 (3.6, 14.7) 17.2 (11.4, 25.1) 
 No formal q10.5 (7.4, 14.6) 30.8 (26.4, 35.6)  8.5 (6.0, 11.9) 10.7 (8.0, 14.2) 15.8 (12.4, 19.9) 

Levels of knowledge and attitudes toward epilepsy were not dissimilar in this U.K. sample to those reported in other recent studies in the developed world (see Fig. 1), although direct comparison is hindered by some differences in question wording.

Figure 1.

Public attitudes to epilepsy: findings from studies in developed countries. (1) In earlier studies, informants were asked whether they had heard/read about the disease called epilepsy; in the present study, they were asked whether they or a close family member or friend had ever had epilepsy, and 1% answered that they had never heard of it. (2) In the studies in Denmark (29), Czech Republic (26), and Hungary (25), question wording was, “Do you think epilepsy is a form of mental illness?” In the studies in Italy (28) and New Zealand (30), wording was, “Do you think epilepsy is a form of insanity?” In the present study, wording was, “Do you think epilepsy is a physical problem, a mental problem, both, or neither?” (3) In earlier studies, question wording was, “Would you object to a child of yours associating with one who had seizures?” In the present study, item wording was, “Children with epilepsy should be allowed to play with other children.” (4) In earlier studies, question wording was, “Do you think people with epilepsy should be employed in jobs like other people?” In the present study, item wording was, “People with epilepsy can be as successful as others in their chosen career.”

DISCUSSION

Our study investigated levels of knowledge and expressed attitudes to epilepsy in a random sample of the U.K. general public. Overall, this group appeared generally well informed about this chronic health condition, including in relation to its potential causes, the way in which seizures are manifested, and the role of antiepileptic drugs (AEDs) in maintaining good seizure control. The group was, perhaps understandably, less well informed about the prevalence of epilepsy, with estimates varying very widely; and significant percentages categorized epilepsy as a mental, rather than a physical, health problem. Expressed attitudes to epilepsy in this U.K. sample were also generally highly favorable. Around 90% either agreed strongly or tended to agree with the statements that: people with epilepsy are as intelligent as everyone else; children with epilepsy should be allowed to play with other children; people with epilepsy can be as successful as anyone else in their chosen careers; and people with epilepsy can lead a normal life. One less positive finding, in light of the preceding, was that a quarter of the sample either agreed strongly or tended to agree that people with epilepsy have more personality problems than others, and that this was a view no less commonly expressed by the quarter of respondents familiar with the condition through having an affected family member or close friend (see later). Another was that despite their positive responses to the items cited earlier, more than half of our informants agreed that people with epilepsy are treated differently by society. This suggests that they recognized that those affected had the potential to lead a normal life and be successful in the same life domains as others, but were limited from doing so by society. Their answers to the two open questions about how and why this differential treatment occurred are congruent with Goffman's (12) argument about societal definitions of nonnormality and differentness. They also lend support to Scambler's (13) proposal that people with epilepsy are perceived as creating ambiguity in their social interactions and so threatening the social order, by being unpredictable and out of control.

Based on previous research showing that contact is an important stigma-reducing strategy (36), we hypothesized that familiarity with epilepsy would be an important determinant of knowledge levels and expressed attitudes toward it, and our analysis supported this to some extent. Those familiar with epilepsy were more likely than others to cite as features of a seizure signs such as loss of consciousness, blank spells, and rapid blinking, and to identify particular causes of epilepsy as being accidents and other illness (data not presented); they were more likely to say that epilepsy can be successfully treated. They were also more likely to agree that compared with others, people with epilepsy are as intelligent and can be as successful in career terms, and were less likely to express concern about having to work with a person affected by epilepsy (data not presented). Interestingly (and somewhat surprisingly to us), they were no less likely than those unfamiliar with epilepsy to agree that people with epilepsy have more personality problems than others. We can only speculate as to the reasons for this finding: it may reflect familiars' particular view of the person they know as not the person she or he was before developing epilepsy; or their generalized knowledge of the research evidence indicating that those developing a chronic illness such as epilepsy are also at increased risk of psychological problems. Further qualitative work is needed to elicit the reasons for their and the nonfamiliars' affirmations of this statement and to establish whether these are based in lived experience or myth. Nonetheless, it is interesting to note that in the earlier study by Gouvier et al. (35), from which the statement was derived, a similar percentage of respondents as in the present study concurred with it.

Factors other than familiarity that were predictive of good levels of knowledge and positive attitudes toward epilepsy were being female, being in a middle-age band, and being of higher social class. The data suggest that the relation with age is a U-shaped one, with the youngest and oldest generally having a poorer knowledge base and more negative attitudes, but the pattern for this variable was less consistent. Our analysis also showed (37) that gender, age, class, and educational-level differences were more pronounced among those members of our sample who were unfamiliar with epilepsy. Positive effects of both familiarity and sociodemographic factors, including female gender and higher educational level, have been reported in previous attitude surveys (29,30,35,38).

Large percentages of informants in the present study expressed the view that people with epilepsy should not be allowed to work in a series of occupations from which they are currently barred by U.K. legislation: as a heavy goods vehicle driver, fire fighter, or member of the armed forces. Around one fourth also thought they should also be barred from working in nursing or teaching, both occupations in which, in the U.K. currently, people with epilepsy are subject to individual assessment of their suitability. Although these responses may have been indicative of relatively high levels of prejudice toward people with epilepsy in the area of employment, it is worth noting that significant numbers of informants also commented that whether a particular occupation was suitable depends on the specific tasks it involves, so the use of blanket descriptors such as “factory worker” or “armed forces” may have been unhelpful and, indeed, may have limited opportunities for informants to express an understanding of the subtlety of such issues and of the clinical complexity of the condition.

Field (39) characterised all health conditions as fitting one of four broad categories: acute illness, chronic illness that is nonstigmatizing, stigmatizing chronic illness, and, as a separate category, mental illness. The issue of visibility has been identified as a key component of the stigma associated with some chronic health conditions (40). Reflecting these taxonomies, the question set included a novel question aimed at eliciting the concerns informants would feel if asked to work with persons with a range of more or less stigmatizing chronic health conditions. The six conditions were selected to include one neutral one (older than 50 years); two visible, stigmatizing physical conditions (facial disfigurement and wheelchair use); one nonvisible, nonstigmatizing chronic health condition (heart attack); one nonvisible, stigmatizsing health condition (epilepsy); and one mental health problem (stress and depression). The condition causing our informants the greatest concern, by a considerable distance, was stress/depression. This may, in part, have been a response to specification in the question of stress/depression involving a lengthy work absence and recognition of the consequent workload implications for colleagues. Equally, it may reflect the relatively higher prevalence of this condition compared with the others cited, meaning that informants had more active experience of it. Epilepsy was ranked as the condition causing greatest concern by one fifth of informants and was also most often cited as the condition causing second or third greatest concern (data not presented). This finding seems at odds with informants' highly positive responses to the attitude statements, including the one relating to the likelihood of people with epilepsy achieving career success. This may be reflective of a process highlighted by Auslander and Gold (41), whereby attitude change occurs more readily at the abstract, intellectual level, “than in relation to real people in social or emotional proximity.” Again, further qualitative work would be of value here.

How, then, did our findings compare with those from previous developed world studies of public attitudes to epilepsy? Before considering the answer to this question, two important caveats must be raised. First, as already noted, direct comparison is limited by the variations in question wording used across the various investigations. In the present study, development of the question set involved a detailed review of questions used in previous studies, with the intention to replicate them where possible. Where, however, they were judged culturally inappropriate or outdated conceptually, they were amended, or novel questions were developed. Second, as with all studies of expressed attitudes, a difficulty arises in the interpretation of our findings inasmuch as respondents in such studies may be subject to the phenomenon of social acquiescence (42), whereby they provide what they believe to be publically acceptable answers; and the extent of such acquiescence may be culturally determined. With these caveats, and based on comparisons that can reasonably be made, the results are encouraging, supporting the view that attitudes to epilepsy in the developed world have improved with time and are generally positive. With regard to the U.K. specifically, our data suggest that here, the general public has shifted considerably in its conceptualisations of epilepsy since the street surveys carried out in the 1980s and that, “the full force of past and contemporary social prejudice and misunderstanding” (15) may finally be diminishing. One explanation for this may be (43) that in the developed world at least, the trend now is to value rather than reject human differences and to redefine as normal conditions of being (including health conditions and disabilities) previously viewed as abnormal. This trend manifests itself in the changing language used to talk about human differences in health and the changing representations of health and illness, including of epilepsy, in the media (41,44). Further, although Hanson (43) suggested that historically the medical model has reinforced the classification of people with health problems and disabilities as abnormal, the reverse may be true for epilepsy: its transformation from the moral to the medical domain, from “badness” to “sickness” (45) has almost certainly contributed to its decreasing stigma. This is a positive message for people with epilepsy and may do much to lessen their private grief about their condition.

Strengths and limitations of the study

This was a large-scale study, involving a random sample of almost 1,700 members of the U.K. general public. The questionnaire was interviewer-administered (computer-assisted personal interview, CAPI), by using a highly trained field force. The achieved response rate of 62% is acceptably high (46) and was similar to that achieved in previous Omnibus Survey rounds conducted at that time of year. Data on nonresponders are not available for the Omnibus Survey, so the possibility of some selection bias exists. However, comparison of our informants' characteristics with those of the U.K. population suggests that any such bias is likely to be minimal. Replication, where possible, of questions from earlier studies of public attitudes elsewhere in Europe and in the North American continent allows some useful comparisons. However, the explanatory power of the study was limited by its quantitative nature: the use of structured questions does not allow detailed exploration of the reasons that the perceivers of stigma hold particular views about a condition such as the one under investigation here. A further limitation relates to the possibility of social acquiescence in the responses obtained: informants may have answered the questions in ways they perceived as socially desirable.

Conclusions and implications for practice

The general public in the U.K. appears reasonably well informed about aspects of epilepsy, but misunderstanding and misinformation also are apparent. Attitudes appear to be influenced by sociodemographic factors in the predicted directions, but some deviations from this general pattern should be further explored to clarify whether they are chance findings or represent some meaningful difference. The results from our parallel surveys of employers and carers will add to the overall picture and help to identify further requirements for intervention studies. Previous studies indicate that knowledge can be improved relatively easily by educational campaigns (38). However, changing attitudes is a more challenging task than increasing knowledge, and Young et al. highlighted the possible dissociation of the two. Future in-depth studies are needed to explore the reasons behind perceivers' negative attitudes, meaningfully to inform public attitude campaigns and ensure they are appropriately targeted, and hence to maximise the likelihood of their success.

Acknowledgments

Acknowledgment:  Our thanks, first and foremost, to the many people across the U.K. who took the time and trouble to answer our questions about epilepsy. We are highly appreciative of their interest in the study. We also thank the staff at the U.K. Office of National Statistics, in particular. Howard Meltzer and Amanda Wilmot for their help in refining our original questions; staff at Epilepsy Action, who assisted with the postal exercise through which we surveyed relatives and friends of people with epilepsy; and Kathleen McAdam in the Department of Public Health at the University of Liverpool for help in preparation of the manuscript. Finally, we are indebted to Sanofi Synthelabo for their financial support of the study and to Virginie Lavaud and Petra Sturm of Sanofi Synthelabo for their continuing commitment and interest in it.

This study formed the basis of an MPhil thesis by Jo Gorry, who was successfully awarded it in July 2003.

Appendix

Appendix: UK Omnibus Survey Module 301 Attitudes to Epilepsy

M301_1

Have you, or a close member of your family or a close friend ever had epilepsy?

M301_2

Of 1,000 members of the general public, how many do you think would have epilepsy?

M301_3

Have you ever seen anyone having an epileptic seizure?

M301_4

How would you know if someone was having an epileptic seizure? Please tell me all the things you would notice about someone who was having an epileptic seizure.

M301_5

Do you think that epilepsy is a physical problem, a mental problem, both a physical and a mental problem, or neither?

M301_6

Do you think that people with epilepsy are treated differently from others by society?

M301_7a

In what way do you think they are treated differently?

M301_8a

Why do you think they are treated differently?

M301_9

Which, if any, of the following jobs do you think that people with epilepsy should not be employed in?

  • Teacher

  • Nurse

  • Police Officer

  • Solicitor

  • Shop assistant

  • Long-distance lorry driver

  • Fire fighter

  • Factory worker

  • Armed Forces

  • None of these

M301_10b

Suppose a new colleague comes to work with you with one of conditions on this card. Could you put these in order that would cause you greatest concern? Please start with the condition that would cause you the most concern, and finish with the condition that would cause you the least concern.

  • A person older than 50 years

  • A wheelchair user

  • A person with epilepsy who had two seizures in the last year

  • A person who had a heart attack a year ago

  • A person who has been off sick with stress/depression for 3 months in the last year

  • A person whose face has been disfigured by burns

M301_11c

How strongly do you agree or disagree with the following statements:

  • a)  People with epilepsy are as intelligent as everyone else.
  • b)  People with epilepsy have more personality problems than other people.
  • c)  It is frightening for others to see someone having an epileptic seizure.
  • d)  Children with epilepsy should be allowed to play with other children.
  • e)  Doctors can successfully treat epilepsy with drugs in most cases.
  • f)  A person with epilepsy can be as successful as other people in their chosen career.
  • g)  A person with epilepsy can lead a normal life.

M301_12M

Which of the items listed on this card do you think are the main causes of epilepsy?

  • Accident (e.g., head injury)

  • Brain disorder

  • Problem with nervous system

  • Mental illness

  • Stress/pressure

  • Born with it/birth injury

  • Result of another illness or disease

  • Old age

  • Alcohol or drug abuse

  • An infection caught from other people

  • None of these

aM301_7 and M301_8 were open questions; interviewers recorded responses verbatim.

b To minimise response bias at M301_10, respondents were randomly assigned to receive prompt cards on which the order of listing of the six conditions varied.

c Possible response options were “strongly agree,”“agree,”“neither agree nor disagree,”“disagree,”“strongly disagree.”

Ancillary