Introduction
Epilepsy is a chronic neurological disorder characterized by abnormal electrical activity that generates recurrent and unpredictable seizures. It is defined by the occurrence of at least two unprovoked seizures with an interval of more than 24 hours or a single seizure with a high risk (>60%) of recurrence.1,2 Globally, epilepsy affects more than 70 million people, with 80% of them in low and middle-income countries.3 It is a highly prevalent disease that impacts the quality of life of patients not only due to seizures, but also because of associated psychosocial difficulties.4,5
The stigma associated with epilepsy is a serious problem that can affect the patients’ physical and psychological well-being. It can be classified into two main categories: enacted stigma-effective external stigma and felt stigma - or perceived stigma.1,2,6 Enacted stigma refers to the real and visible discrimination that individuals with epilepsy face due to their condition, while felt stigma involves fear of stigmatization, resulting from the anticipation of discrimination.4 Clinically, stigma can be further subdivided as self-stigma, attribution stigma, and other stigma nomenclatures, however the patient manifestations are similar and involve social withdrawal and behavior compatible with some psychiatric disorders, such as depression and anxiety.4,6 All types of stigma have significant impact on patients’ quality of life, contributing to a cycle of exclusion and psychological suffering.
There is evidence that people with epilepsy (PWE) have a significantly higher prevalence of psychiatric comorbidities, such as major depression, anxiety, stress and psychogenic seizures, than the general population.4,7 Recent systematic reviews and meta-analyses found that the overall prevalence of active depression in subjects with epilepsy was 23.1%, while anxiety affected about 20.2% of individuals with the condition.8,9 These psychiatric comorbidities have a significant negative impact on health-related quality of life, often augmenting the impact of epileptic seizures.8 It has been suggested that shared neurobiological mechanisms, such as neurotransmitter disorders and endocrinopathy, may contribute to the interrelationship between epilepsy and psychiatric disorders.7
From a public health perspective, the stigma and discrimination associated with epilepsy have significant implications. The international league against epilepsy and other international organizations highlight that despite advances in the control of epileptic seizures, social stigma remains a major challenge, and often more difficult to overcome than the seizure itself.4
Objective
The aim of this article is to define the precise stigma prevalence across all studies involving prevalence outcomes; with a focus on potentially uncovering insights regarding which approaches and settings may contribute to higher or lower stigmatization. Also, the correlation with psychiatric comorbidities in real-life scenarios, investigation of prevalence alignment, implications for clinical practice and public health policies are explored and reviewed.
Methods
Systematic review
A systematic review was conducted by two authors independently, according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses statement, to identify studies assessing prevalence of perceived, internalized or self-reported stigma in PWE.10 The databases PubMed and Scopus were searched from inception to May 31, 2024. Observational and interventional studies were eligible, whereas reviews, meta-analyses, correspondence and guidelines were excluded. The following search strategies were used on PubMed: “stigma [tiab] AND prevalence [tw] AND epilepsy [tw] AND cross-sectional [tiab]”; and Scopus: “(TITLE-ABS-KEY (stigma) AND TITLE-ABS-KEY (prevalence) AND TITLE-ABS-KEY (epilepsy) AND TITLE-ABS-KEY (cross-sectional))”; with no language restrictions. Data extraction was conducted by all authors, employing standardized questionnaires including year, country, sample size, patient characteristics, assessment instrument, stigma prevalence and conclusions.
Meta-analysis
Meta-analyses, for all studies and subgroups, were executed with random effects models weighted by inverse variance applying the Hartung-Knapp correction for Clopper-Pearson confidence intervals. Interstudy variance (τ) was estimated using restricted maximum likelihood. Studies were regionally grouped. As a significant number of data came from Africa, while the rest was mainly from Asia, the defined groups were Africa and Asia (mostly). Cochran’s Q was calculated to test for subgroup differences. Funnel plotting with Egger’s test was used to evaluate publication bias. A study density world map was constructed for global publication distribution analysis. All statistical analyses were done in R ver. 4.3.3 (R Foundation for Statistical Computing, Vienna, Austria).11
Results
Database search yielded 105 records, which after the screening process described in Fig. 1 resulted in 22 primary studies evaluating stigma prevalence in PWE that were pooled in the meta-analysis. A meta prevalence of 35% (29%; 41%), including 6,072 participants, was reached. There was a significant difference between subgroups (p=0.03).
African subgroup
Of the 13 eligible African studies, 84.6% were conducted in hospital settings, and 53.8% employed the Kilifi stigma scale of epilepsy, 12–17 known for its high internal consistency as measured by Cronbach’s α, to assess epilepsy-related stigma, although one used an adapted version.1 Other assessment tools were the three item stigma scale,7,18,19 study-specific stigma scales2,20 and the family interview schedule.6
With an overall meta-prevalence of 40% (34%; 46%); and 3,738 patients in total, the African subgroup corresponded to 59.1% of the weight of the overall meta-analysis, as shown in Fig. 2. Studies focused on the associations between epilepsy-dependent variables such as seizure frequency, number of antiepileptic drugs in use, epilepsy related stigma, length of treatment, treatment response and type of epilepsy syndrome; and correlated those variables with independent variables such as medication nonadherence, depression, stress or anxiety disorders, school attendance, functional seizures and marriage rate.
Associative evidence between epilepsy related stigma and the independent variables was found in most investigations. Four studies identified a significant association with depression7,14,18-one of which found a correlation with depression but not with anxiety;19 three studies found an association with treatment nonadherence;12,13,17 two studies identified an association with higher seizure frequencies,6,16 and the other four studies each found associations with disrupted school attendance, 1 psychogenic seizures,2 lower marriage rates,20 and stress.15
Asia (mostly) subgroup
In the Asia subgroup, which also had one study from America and one study from Europe, there was a substantial South Korean representation, mainly from research of the Korean quality of life in epilepsy group. Out of the nine studies, 44.4% used the stigma scale-revised, 5,8,9,21 whereas the other five studies applied unique questionnaires, including the epilepsy stigma scale,22 stigma scale for epilepsy (which preceded the stigma scale revised),23 adapted internalized stigma of epilepsy,24 a study-specific scale,4 and the liverpool stigma scale.25 All studies were conducted in hospital settings.
The subgroup meta-analysis identified a pooled prevalence of 28% (21%; 37%), accounting for 2,334 total patients. Associations between epilepsy related stigma and depression,24 polypharmacy,4,22 higher seizure frequency,22 lower self-reported quality of life,23 social anxiety,8,9 and poor disclosure management21 were found. An investigation comparing types of epilepsy found that for both genetic generalized epilepsies and mesial temporal lobe epilepsy with hippocampal sclerosis - an epilepsy syndrome known to be typically refractory to treatment - there was significant prediction-type correlation between stigma and polypharmacy, seizure frequency and psychiatric comorbidities.4
One of the included studies compared psychogenic non-epileptic seizures (PNES) and epileptic seizures and found higher degrees of social impairment among people with PNES.25 This may reinforce the case report findings, as in it the patient had both conditions. In one of the Korean studies,5 although no direct link was made, lack of epilepsy knowledge was associated with both epilepsy related stigma and depression, suggesting reverse causality. Further information was summarized in Table 1.
Overall
It is well known that PWE experience a higher prevalence of psychiatric comorbidities. Here, besides describing a more precise stigma prevalence among PWE, we reinforce the evidence that significant associations exist between epilepsy related stigma and depression, stress, anxiety, social anxiety and functional-or psychofunctional or psychogeneic seizures, although reverse associations are also possible. This may further specify the etiology of these psychopathological events and lead to the realization that these patients are at risk and would possibly benefit from tailored management policies.
Stigma scales comparison
Two major trends were identified in the stigma scales used by eligible studies: a preference for simpler questions, oftentimes in sets of three; and the employment of grading systems for responses, ranging from simple yes/no answers to eight possible response grades in a Likert approach. The most frequent types of stigma assessed were felt-or internalized-(90%) and enacted- or actual-(40%) stigmas. Further details from the stigma scales are displayed in Table 2.
Case report
A 16-year-old female with a medical history of epilepsy and intellectual disability presented to the primary care clinic with frequent syncopal events that had begun 2 months prior. The episodes were preceded by malaise, lightheadedness, and nausea; without loss of sphincter control, motor jerks, tongue biting or vomiting. The first episode had occurred in school, and their frequency had been increasing, which led to irregular school attendance.
After repeated attempts to establish rapport, she reported having troubled relationships and difficulty learning. Since age 6, she had been taking sodium valproate and phenobarbital, with the current daily doses of 1,000 and 200 mg, respectively. Family history was positive for epilepsy with motor seizures. Physical examination revealed a tearful patient with communication difficulties marked by dysarthria and dyspraxia. The neurological exam had no abnormalities.
A diagnosis of PNES was proposed; whereas the diagnostic criteria for major depressive disorder were met. To further evaluate and confirm the first, the patient was referred to a tertiary neurology center. The antiepileptic drugs were maintained and the patient was started on fluoxetine 20 mg, twice a day.
After a month, the patient returned for follow-up consultation. She still appeared depressed, cried during the interview, but indicated a significant improvement in how she felt during the last month. The focal seizures, which she had been treating since an early age, had not happened during this interval. However, she reported that the PNES still occurred, albeit with a slightly lower frequency. During the interview, the patient’s mother described more precisely how the prejudice was being used against her daughter, mainly through bullying.
Between the two consults, she had an appointment with a neurologist, who started decreasing the phenobarbital dose to 100 mg, once daily. We maintained phenobarbital weaning, keeping the 100 mg daily dose; and increased the dose of Fluoxetine to 40 mg, twice daily.
Since one of the sources of stigma was identified, and it had gone on for long, solving the prejudice system that had formed without harming the patient would be difficult, so we discussed the possibilities the patient’s family had, including taking legal action or changing schools. The patient’s mother decided to move to another city, due to multiple social factors.
Conclusion
Summary of main findings
This study highlights the burden of stigma in PWE, which is may be directly linked to the load of psychiatric diseases in these patients. Given the evidence manifested by the meta-prevalence and associative studies, epilepsy should be considered a risk factor for the development of major depression, anxiety disorders, social phobia and stress. A bidirectional relationship may be intertwined in the pathogenesis of these associations, especially depression. Notwithstanding, these patients’ quality of life may be compromised in a setting which may already be marked by significant social and economic difficulties, since about 80% of the PWE live in low and middle-income countries.3
Besides psychiatric comorbidities, cognitive impairment is frequent in PWE. The need for an exclusion criterion of not understanding the questionnaire, present on must studies, as described in Table 1, reinforces the high comorbidity between the two conditions. Although there is a lack of robust evidence confirming this association, it may be hypothesized that frequent intense stigma manifestations may lead to social withdrawal, which could lead to poor social and cognitive skill development.
Two of the included papers found an association between stigma and psychogenic seizures - also known as PNES.2,25 Incidentally, the patient in the reported case developed both organic and suspected psychogenic seizures during a period in which prolonged stigmas occurred. Given the circumstances, although confirmation bias should not be ignored, the systematic approach employed in data extraction reduces the probability that evidence may have been selectively picked. Hence, observational studies, coupled with a descriptive basis, make for a compelling argument in favor of associating these psychopathological entities - stigma and PNES.
It is also worth highlighting sex differences in psychological disorders, and how epilepsy may affect this correlation. In PWE, there were conflicting results regarding sex differences in prevalence and risk factors for anxiety. As reported by Lee et al.,8 an investigation found that anxiety was more common in women than in men, and a greater frequency of seizures in the past year was a risk factor for anxiety in men, but not in women. Another study, however, revealed that anxiety prevalences were similar between male and female patients, and that seizure frequency was linked to anxiety symptoms in women, but not in men.8 As such, the relation between sex and anxiety remains elusive.
Conversely, the two clinical investigations that enrolled only female participants20,22 had results that did not differ significantly from other eligible trials. In fact, Pattnaik et al.22 displayed a stigma prevalence below that of most studies, although direct comparison is not possible since it employed a unique stigma scale; and Teuwen et al.20 reported a discreetly higher prevalence than its subgroup. Therefore, a subgroup meta-analysis moderated by sex was not conducted, also considering that meta-analyses based on only two studies are not robust.
The multidimensional impact of epilepsy on mental health, especially when there is associated depression, must be considered. Despite growing literature on direct association of epilepsy, depression and quality of life, an implicit consequence is cognitive impairment. The case here reported has both staged and felt stigma, and bullying seemed to affect academic performance. This stigma may be correlated both with the fact that she had epilepsy with global developmental delay and, above all, the cycle reinforcement by depression which seemed to be partially generated by stigma. The fear of going to school, as the patient is aware of the discrimination and exclusion suffered, feeds a vicious cycle which ends up interfering with daily activities.
Implications and limitations
Although reverse causality cannot be ruled out, the high prevalence of stigma directed toward PWE and its association with psychiatric comorbidities represent a definitive reason why stigma in people living with epilepsy should be mitigated. Moreover, the alignment between a stigma prevalence of 35% and the previously reported depression prevalence of 23.1% in populations with epilepsy reinforces the hypothesis of a possible causal relationship between the two disorders. The direct clinical implications of such findings may be the increase in difficulty of managing these patients and the need for recognition of the conditions as comorbidities, whereas one should be investigated if the other is identified.
The global distribution of studies related to stigma in PWE may serve as a hypothesis generating basis. The reason that almost all published investigations were from Africa and Asia might represent a difference in scientific knowledge trends between these regions. Other continents may have a more treatment-oriented type of research, instead of associative or individual feelings-focused investigations, as were observed in Africa and Asia. Another possibility is the lack of visibility of stigma in these regions or, optimistically, its actually reduced presence, which could serve as an example of how to improve outcomes.
This study reviews multiple stigma scales with different approaches that were ultimately aligned in investigating how patients felt about themselves and about others. The presented data may contribute to delineating the ideal evaluation tool for each setting. However, different sensibilities, both related to the scale itself or the cut-off points used by the studies, for each situation should not be overlooked as they may increase or decrease stigma prevalences. Another point worth emphasizing are the evaluation tools used for the psychiatric comorbidities and the statistical approaches employed, which should also be considered in a case-by-case basis.
Limitations of this review include the design of the eligible studies as most were observational. The funnel plot displayed only discrete asymmetry, with one more study with higher prevalence as compared to the equivalent lower prevalence side of the plot, shown in Fig. 2 (p=0.22). This nonsignificant lower number of low stigma prevalence studies may be due to the reduced prevalence of epilepsy in high income countries, which could thus reduce disease burden and research urgency.3
As depicted in Fig. 3, the two countries with highest representation in this review were Ethiopia and South Korea, with nine and five studies, respectively. In both cases, there were recurring clinical investigations. In the first case, for example, three studies had their population sample enrolled from the Emanuel Mental Hospital, at varying timepoints. Similarly, the South Korean investigations were conducted by the same group, regularly during 4 years, employing a network of hospitals to increase robustness of data. Although comparable, each study had a particular purpose, such as testing whether a different writing manner for epilepsy would influence stigma. Assuredly, recognition of stigma against PWE as a major public health problem by patients and health professionals in these countries was a major influence.
The opposite might also hold, and the lower burden of epilepsy in highly developed countries may have the converse implication, reducing the number of related publications. This is reinforced by a characteristic observed in most included studies: lower-than-average socioeconomic conditions. The case presented is an example, and also proof that these patients can be managed in the primary care setting.
The identified subgroup differences between Africa and Asia are potentially due to three characteristics: 1) a myriad of sociocultural differences which may either hasten or slow the propagation of stigmatized ideas; 2) differences in stigma scale sensibility, with the manner in which some questions are asked potentially interfering in how they are interpreted, thus resulting in diverging sensibilities; and 3) economic disparity and the ensuing information barriers. Further specificities of the two settings should be investigated with a focus on if different public health policies, for example, may have influenced these results.
There seems to be an association between stigma in PWE and psychiatric disorders, mainly depression, social interaction anxiety, social phobia and stress, while also contributing to the development of neurological complications such as PNES. Particularly for depression, this association is sufficiently strong for some studies to claim that “early screening and management of depression among PWE should be mandatory”.15 This aggregated evidence reinforces that epilepsy is indeed a neuropsychiatric disorder, having such complexity that neither of the areas isolated might be sufficient to provide sizeable, accurate and thorough evaluations; calling for a multidisciplinary approach.
Stigmas among PWE are highly prevalent and comorbid to several psychiatric diseases, some of which are yet to be causatively linked to stigma but already have established evidence of association and plausibility. This systematic review summarizes clinical investigations whose focuses were not necessarily stigma but that measured its prevalence with objective instruments nonetheless. Most of these studies were from low-income countries, which is consistent with the disease distribution worldwide. There is an increasing need for awareness of the challenges faced by these patients from social and medical perspectives. Tailored management strategies would be of high benefit.
