Cross sectional study of reporting of epileptic seizures to general practitioners

Introduction

Epilepsy is the most common neurological condition encountered in primary health care in the United Kingdom.1 2 General practitioners have assumed much of the responsibility for the continuing management of this condition.3 As the lifetime prevalence of this condition is 2-5% of the population, a general practitioner will care for about 10 patients receiving treatment but will have another 15–25 patients who have had seizures in the past but are no longer receiving treatment.2

A key aspect of monitoring patients with epilepsy is the reporting of seizures to their doctor. This information is used to determine the most appropriate antiepileptic drug and its dose and is also the basis of advice on lifestyle—for example, driving, employment, and leisure activities. It is therefore unsurprising that several researchers have suggested that seizures—and even the diagnosis of epilepsy—might be concealed. Van der Lugt, for example, reported that only a small proportion of people disqualified from military service because of epilepsy reported their epilepsy when applying for a driving licence.4 Taylor et al said that a “considerable proportion of patients may continue to drive illegally.”5 Moreover, Hopkins and Scambler reported that 12 (19%) out of 62 subjects ineligible to hold a UK driving licence because of uncontrolled epilepsy were driving, in two cases without a licence.6 Similar problems have been found in other countries. A study in the United States found that 28% of drivers with epilepsy would not inform their doctor if they had a seizure,7 and an assessment of notification to the driver licensing authorities in Australia asserted that: “patients with epilepsy withhold the diagnosis from the authorities, and worse, may withhold information about the occurrence of seizures from their advising doctors.”8

Concealment from employers has also been reported. In a study in the United Kingdom 53% of people with epilepsy and in employment would choose concealment as a strategy.9 Scheinder and Conrad identified concealment as one of several processes by which individuals manage their epilepsy.10 In a study of patients' reactions to fictional situations, Tröster found that disclosure of epilepsy depended on the perceived risk of detection and the anticipated consequences of disclosure; about 30% of patients would never reveal the diagnosis whatever the situation.11

These studies suggest that the concealment of seizures or epilepsy is common and that this is likely to compromise the care of epileptic patients. However, no studies have attempted to quantify the extent to which seizures are concealed. We compared the reported seizures from two questionnaires completed by the same patient, one administered by their general practitioner and one self completed anonymously.

Participants and methods

This was a general practice based cross sectional study into the reporting of seizures to a patient's doctor and therefore removes the bias of other hospital clinic based studies. Two linked questionnaires were designed to examine expected differences in the reporting of seizures. The first questionnaire was completed by the general practitioner with the patient during a routine consultation. The second questionnaire was sent to the patient two weeks after the initial consultation. At this point the first questionnaire was anonymised and linked to the second questionnaire by a unique number.

The first questionnaire investigated patient characteristics: age, sex, socioeconomic group, and current occupational status and, in addition, information on type of epilepsy, date of diagnosis, treatment, and the date and time (day or night) of last seizure. The second, anonymous, questionnaire was identical to the first apart from the addition of the hospital anxiety and depression scale12 together with a measure of perceived stigma.13 The hospital anxiety and depression scale has been shown to be a sensitive and specific screening tool for mood disorders in patients with somatic illnesses. Scores of 8 or above were taken as predictive of depression or anxiety.12 Perceived stigma was measured as the positive response to one or more questions asking whether other people were uncomfortable with them, treated them as inferior, or preferred to avoid them.13

On the basis of an unpublished audit of practice data from 1994–5 we calculated that the sample size necessary to detect a difference in the reported seizure rates at 5% significance levels and 90% power was at least 85. General practitioners were recruited from Norwich and the county of Norfolk to give a study population representative of urban and rural practice which reflected the general population. Each general practitioner was asked to recruit only four patients. The Norwich district ethics committee approved the protocol for the study.

Patients were recruited by 31 general practitioners during a routine review of their epilepsy between 1 September 1996 and 31 March 1997. Consecutive patients over 16 years of age with a diagnosis of epilepsy were invited to take part in the study. Any patient who had had epilepsy diagnosed by a neurologist could be included, whether they were receiving antiepileptic drugs or not. Only patients who were personally responsible for the care of their epilepsy were included. Patients were informed, by their general practitioner and the patient information sheet, that the study was investigating control of seizures. Informed consent was obtained by the recruiting general practitioner. The seizure rate in the participants was analysed by directly comparing the dates of seizures given in both questionnaires.

Results

A total of 122 patients was recruited by the general practitioners. The reported response rate was over 90%. Of these 122 patients, 111 (94%) returned the second, confidential questionnaire. These comprised 52 (48%) men and 59 (52%) women with a mean (SD) age of 48.5 (16) years (range 16–81 years) and a mean duration of illness of 21.5 (15.5) years. Sixty two (51%) subjects were in socioeconomic groups 4 and 5. The age and sex of the 111 respondents mirrored that found in previous studies of epilepsy in general practice (analysis of variance W=68, P=0.87).1 14 15 Analysis of the ages, sex, and type of epilepsy or frequency of seizures for the 11 non-respondents showed no significant differences from respondents (Mann-Whitney test w=330.5, P=0.94). In particular, five patients reported a seizure during the past 12 months and six did not.

Of the 111 respondents who completed the anonymous questionnaire, 42 admitted to their general practitioner that they had experienced a seizure during the past 12 months and 60 reported a seizure on the anonymous questionnaire (table 1). One patient reported a seizure to the general practitioner but not anonymously. No patients experienced seizures during the interval between completing the questionnaires. This difference was significant (P<0.001) and comparison of proportions indicated a 16% (95% confidence interval 9% to 23%) difference in the number of reported seizures.

Table 2 gives data on possession of driving licence and employment status according to reporting of seizures to general practitioners and anonymously. Six patients (out of 42) who reported a seizure to their general practitioner also reported having a driving licence, but 24 out of 60 patients with anonymously reported uncontrolled epilepsy possessed a licence (table 2). This difference was significant (P<0.001). Although patients were asked if they possessed a driving licence, they were not asked if they were currently driving. Therefore, we cannot assume that all patients with a licence were driving. However, several patients volunteered information in their questionnaire that indicated whether they were driving and the figures were adjusted accordingly.

After patients who had retired (including those retired on medical grounds) had been allowed for, the self reported unemployment rate for the study population was 32% (27 patients) of the economically active population (85 patients). Patients who had uncontrolled and unconcealed seizures (42) had an unemployment rate of 47% whereas patients with uncontrolled but concealed seizures (18) had an unemployment rate of 17% (difference in proportions 30% (95% confidence interval 27% to 33%), P<0.001). At the time of the study the national unemployment rate was 9%.

Discussion

We found that reporting of seizures to general practitioners was significantly lower than reporting in an anonymous questionnaire. About a sixth of patients concealed seizures from their general practitioner. The anonymous reporting equates to an annual seizure rate of 53%. This is higher than the generally reported annual seizure rate of about 30%,16 although similar rates have been reported in other studies.14 17

Scambler found that secrecy and concealment was epileptic patients' first choice strategy for managing their seizures.18 He developed the hidden distress model as an explanation for this strategy and introduced the concepts of felt and enacted stigma. Enacted stigma refers to episodes of discrimination against people with epilepsy on the grounds of social and cultural unacceptability. Felt stigma has two referents: the shame of having epilepsy and the fear of encountering enacted stigma.