If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Neurology and Institute of Health Science, Gyeongsang National University College of Medicine, Jinju, Republic of KoreaDepartment of Neurology, Gyeongsang National University Changwon Hospital, Changwon, Republic of Korea
Department of Neurology and Institute of Health Science, Gyeongsang National University College of Medicine, Jinju, Republic of KoreaDepartment of Neurology, Gyeongsang National University Hospital, Jinju, Republic of Korea
Department of Neurology and Institute of Health Science, Gyeongsang National University College of Medicine, Jinju, Republic of KoreaDepartment of Neurology, Gyeongsang National University Hospital, Jinju, Republic of Korea
Corresponding author at: Department of Neurology, Gyeongsang National University Hospital, Gyeongsang National University College of Medicine, 816-15 Jinjudae-ro, Jinju, 52727, Korea.
Affiliations
Department of Neurology and Institute of Health Science, Gyeongsang National University College of Medicine, Jinju, Republic of KoreaDepartment of Neurology, Gyeongsang National University Hospital, Jinju, Republic of Korea
This meta-analysis is for anxiety disorders (ADs) in outpatient epilepsy clinics.
•
The current prevalence of any ADs in people with epilepsy (PWE) was 26.1 %.
•
The most prevalent subtype of ADs in PWE was generalized anxiety disorder (GAD).
•
The current prevalence of GAD was 18.2 %.
•
In terms of the prevalence, agoraphobia and social phobia followed GAD.
Abstract
Background and purpose
Although anxiety disorders (ADs) occur frequently in people with epilepsy (PWE) and impair quality of life and treatment outcomes, current efforts to categorize and investigate AD subtypes in PWE remain insufficient. Thus, the present meta-analysis aimed to determine the current prevalence rates of any AD type and various AD subtypes in PWE managed by outpatient clinics.
Methods
MEDLINE, EMBASE, Cochrane Library, Web of Science, and SCOPUS were searched to identify and select studies that assessed the prevalence of ADs or individual AD subtypes in adult PWE under the routine care of outpatient epilepsy clinics in tertiary hospitals. Only studies that used gold-standard diagnostic tools for assessing ADs were included in this meta-analysis.
Results
The database search ultimately identified 15 studies, of which 9 provided current prevalence rates of any type of AD. The pooled estimated prevalence of any AD was 26.1 %. Of the 15 total studies, 13 provided current prevalence rates of generalized anxiety disorder (GAD), revealing an overall estimated prevalence of 18.2 %. In terms of current prevalence rates, GAD was highest, followed by agoraphobia, social phobia, panic disorder, and obsessive-compulsive disorder.
Conclusions
Among PWE managed in the outpatient epilepsy clinics of tertiary care hospitals, the current prevalence of any AD was 26.1 %, and GAD was the most prevalent subtype of AD.
Prevalence of self-reported epilepsy or seizure disorder and its associations with self-reported depression and anxiety: results from the 2004 Health Styles Survey.
]. Because anxiety has significant impacts on quality of life (QOL) and treatment outcomes in PWE, the identification of anxiety is crucial for the management of this disorder. Anxiety disorders (ADs) in PWE include a variety of subtypes such as generalized anxiety disorder (GAD), agoraphobia (AP), social phobia (SP), panic disorder (PD), obsessive-compulsive disorder (OCD), and post-traumatic stress disorder [PTSD; 7, 8]. However, the general categorization of ADs has yet to be fully established. For example, the International Classification of Diseases, 10th revision (ICD-10), classifies OCD and PTSD as distinct categories separate from ADs [
], PWE may be more likely to experience ADs. In fact, the current prevalence of any type of AD in PWE under routine care in outpatient epilepsy clinics ranges from 8.0 %–52.3 % [
Combined detection of depression and anxiety in epilepsy patients using the Neurological Disorders Depression Inventory for Epilepsy and the World Health Organization well-being index.
]. Thus, the integration of data provided by epilepsy clinics in a meta-analysis will provide useful information for comparing the current prevalence rates of ADs between PWE and the general population. Furthermore, the use of reliable diagnostic tools is necessary to determine the true prevalence of anxiety in epidemiological studies. For example, a meta-analysis of ADs and depressive disorders in PWE found that the diagnostic tools used in the primary studies potentially affect the results of epidemiological studies [
]. Those authors also reported that the prevalence of anxiety is 8.1 % in studies that used unstructured clinician assessments to diagnose anxiety, whereas the prevalence was 27.3 % in studies that administered structured clinical interviews.
ADs in PWE have rarely been investigated as a primary topic in systematic reviews and meta-analyses. Thus, the present meta-analysis aimed to assess the current prevalence rates of any AD and AD subtypes in PWE under routine care in epilepsy clinics by identifying and selecting proper epidemiological studies.
2. Methods
2.1 Search strategy
The present meta-analysis was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses as a guideline [
]. The protocol for this meta-analysis was prospectively published on the International Prospective Register of Systematic Reviews (registration number: CRD42019139242). A search strategy was established after selecting the following keywords and phrases to identify relevant articles: “epilepsy”, “epilepsies”, “anxiety disorders”, “anxiety”, “anxieties”, “anxious”, “GAD”, “panic”, “PD”, “agoraphobia”, “phobia”, “obsessive-compulsive”, “OCD”, “post-traumatic stress”, “post-traumatic stress”, “PTSD”, “diagnosis, differential”, “diagnosis”, “M.I.N.I.”, “MINI”, “Mini-International Neuropsychiatric Interview”, “S.C.I.D.”, “SCID”, “structured clinical interview”, “C.I.D.I.”, “CIDI”, and “Composite International Diagnostic Interview”. The search strategy used for this meta-analysis is presented as an appendix.
Several electronic databases, including MEDLINE, EMBASE, Cochrane Library, Web of Science, and SCOPUS, were searched on 28 July 2019. Additionally, other bibliographies, such as reference lists and gray literature, were also examined to avoid missing relevant studies. Publication date and language were not limited. All included studies were articles or proceedings, and the data were exported to EndNote X9.2 [
Only studies that fulfilled all of the following criteria were selected for this meta-analysis: conducted as original research, included adult subjects, provided the prevalence of any AD and/or AD subtype, performed in outpatient epilepsy clinics in tertiary hospitals, and used a structured interview for diagnosing ADs. Studies that investigated specific subgroups of PWE (e.g., children, the elderly, women, drug-resistant epilepsy, temporal lobe epilepsy, idiopathic generalized epilepsy, juvenile myoclonic epilepsy, patients under preoperative evaluation, first-ever seizure, and newly diagnosed epilepsy) were excluded from the meta-analysis because the compositions of these PWE populations differ from those of PWE populations treated in the outpatient clinics of tertiary hospitals. Epilepsy outpatient clinics for routine care typically manage PWE with various causes but not specific groups of PWE.
An initial screening process and subsequent selection process were performed to identify the studies eligible for the present meta-analysis. During the initial screening process, two reviewers (OY Kwon and TW Yang) independently reviewed the titles and abstracts of the selected studies, of which the appropriate studies were then included in the subsequent selection process. During this selection process, the two reviewers independently reviewed the full-length texts of the studies identified in the initial screening. If there were any disagreements between the two reviewers regarding a specific article, then that study was reappraised through discussion or consultation with the third author (DH Kim).
2.3 Data synthesis and analysis
2.3.1 Data extraction
The data extracted from the selected studies included study information (author, year, publication type, area, diagnostic tool, purpose, and exclusion criteria), demographic characteristics of the study patients (number, age, and sex), duration of epilepsy, and information associated with effect size (prevalence, type of prevalence, event number, and sample number). In this meta-analysis, GAD, AP, SP, PD, OCD, and PTSD were categorized as subtypes of AD based on the DSM-IV classification. If there were discrepancies among the reviewers regarding the extracted data, they were discussed until an agreement was reached.
2.3.2 Main analysis
All data were assessed using R software (version 3.6.1, The R Foundation for Statistical Computing; Vienna, Austria; [
]). All information extracted from the selected studies was combined, and then the pooled estimates of current prevalence rates and confidence intervals (CI) for any AD, GAD, AP, SP, PD, and OCD were independently calculated. It was not possible to include PTSD in the meta-analysis, because only one study providing the current prevalence of this disorder was identified. It is also important to note that the primary studies included in this meta-analysis may have been homogeneous, because all were conducted in relatively similar environments, and gold standard structured interviews were administered to diagnose ADs. Thus, the fixed-effects model was applied in a predetermined manner.
2.3.3 Heterogeneity
Forest plots were conducted to evaluate heterogeneity via visual analysis. Additionally, Cochrane Q and I2 values were computed for statistical evaluation. If the I2 was > 50 %, then the included studies were considered to be substantially heterogeneous.
2.3.4 Publication bias
If more than 10 studies are included in a meta-analysis, then they should be evaluated for publication bias. A funnel plot was constructed for graphical evaluation, and Egger’s regression intercept was calculated for statistical evaluation.
2.4 Quality assessment of the literature
The risk of bias tool for prevalence studies developed by Hoy et al. [
] was used for quality assessment of the primary studies. This tool is a 10-item inventory in which Items 1–4 evaluate external validity (target population, sampling frame, random selection, and nonresponse bias) and Items 5–10 internal validity (data collection from the subjects, case definition, study instrument, mode of data collection, prevalence period, and numerator and denominator). When evaluating each study, the evaluator should assess individual items in terms of low or high risk of bias, such that one point is given for a low risk of bias, and no points are given otherwise. The summary assessment that combines the scores of all 10 items represents the degree of bias risk. The summary assessment score ranges from 0 to 10, such that a higher value is indicative of a lower risk of bias.
3. Results
3.1 Identification of relevant studies
A flow chart depicting the article search and selection procedures is shown in Fig. 1. A total of 3110 possible studies were identified and selected for the present meta-analysis: 739 from MEDLINE, 1092 from EMBASE, 124 from Cochrane Library, 45 from Web of Science, and 1110 from SCOPUS. Immediately following the search procedure, 537 duplicate studies were excluded by an automatic check performed using EndNote X9.2 software [
], and then another 2532 studies were eliminated during the initial screening process. The full texts of the remaining 41 studies were reviewed, resulting in exclusion of another 26 studies for the following reasons: duplicated data (n = 9), lack of a structured interview (n = 6), inclusion of a specific epilepsy group (n = 3), no mention of prevalence duration (n = 3), no information regarding prevalence (n = 3), pre-screening using an inventory (n = 1), and different durations of prevalence (n = 1). Ultimately, 15 studies were selected for the present meta-analysis.
Fig. 1Flow diagram of the selection of the relevant studies. A total of 3110 research articles were identified by searching five databases. Of these articles, 537 duplicate studies were removed, and an additional 2532 studies that did not satisfy the selection criteria were excluded from the analyses. The full texts of the remaining 41 studies were reviewed, and 26 of these were excluded for various reasons. Thus, a total of 15 studies were ultimately selected for the meta-analysis.
The characteristics of the selected studies are summarized in Table 1. All 15 studies were articles conducted in Asia (n = 6), Europe (n = 5), North America (n = 2), Africa (n = 1), and Australia (n = 1) and published from 2005 to 2019. More specifically, two studies each were performed in Korea and the United States and one study each in Australia, China, Denmark, France, Nigeria, Poland, Thailand, and Turkey; one study was a cooperative work between Germany and Italy. Of the 15 selected studies, AD was diagnosed using the Mini-International Neuropsychiatric Interview (MINI) in 12 and the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID) in 3. The diagnoses were based on DSM-IV criteria in 10 studies and both the DSM-IV and ICD-10 criteria in 5 studies.
Table 1Characteristics of the 15 studies selected for the meta-analysis.
Turkey is a country that spans two continents, Asia and Europe. Because of the geopolitical proximity of Turkey to other European countries, it was classified as a European country.
SCID
DSM-IV
Prevalence of PsyC
Yes
29.0
9.3
42.0
14.4
10.3
Mula 2008
Europe
Germany, Italy
MINI
DSM-IV
Features of IDD
Yes
43.5
14.4
43.6
19.5
14.8
Jones 2005
North America
USA
MINI
DSM-IV
Prevalence of PsyC
Yes
39.0
11.9
66.1
17.9
12.7
SCID: the Structured Clinical Interview for DSM-IV Axis I Disorders, MINI: the Mini-International Neuropsychiatric Interview; DSM-IV: the Diagnostic and Statistical Manual of Mental Disorders, 4th edition; ICD-10: the International Classification of Diseases, 10th revision; STAI: the State-Trait Anxiety Inventory; PsyC: psychiatric comorbidities; LAEP: the Liverpool Adverse Event Profile; GAD-7: the Generalized Anxiety Disorder 7; NDDI-E: the Neurological Disorder Depression Inventory for Epilepsy; QOL: quality of life, WHO-5: the 5-item World Health Organization Well-Being Index; HADS: the Hospital Anxiety and Depression Scale; AED: antiepileptic drug; AEs: adverse events; IDD: interictal dysphoric disorder; NA: not available.
* Turkey is a country that spans two continents, Asia and Europe. Because of the geopolitical proximity of Turkey to other European countries, it was classified as a European country.
Of the 15 studies, the most common objective (n = 5) was to investigate the prevalence of psychiatric comorbidities, whereas 2 studies were performed to validate the GAD-7 scale. The primary aims of the remaining 8 studies varied among the following goals: the GAD-7 and the Neurological Disorder Depression Inventory for Epilepsy (NDDI-E) for QOL, Liverpool Adverse Event Profile for depression and anxiety, effects of antiepileptic drug (AED)-related adverse events (AEs), factors for psychiatric comorbidities, features of interictal dysphoric disorder, validation of the NDDI-E and the 5-item World Health Organization (WHO) Well-Being Index, validation of the Hospital Anxiety and Depression Scale, and validation of the State-Trait Anxiety Inventory. Exclusion criteria for the recruitment of PWE were enforced in 13 studies. In the 15 studies, the mean age of the PWE ranged from 26.6–50.5 years old (median = 39.0 years), the proportion of females ranged from 32.3 %–67.9 % (median = 43.6 %), and the mean duration of epilepsy ranged from 8.7–19.5 years (median = 16.0 years).
The sample sizes of, and current prevalence rates reported by, the AD studies are summarized in Table 2. The sample sizes of the individual studies ranged from 50 to 243 (median = 132). Any AD type was assessed in 9 studies, GAD in 13, AP in 5, SP in 4, PD in 7, OCD in 7, and PTSD in 1, with current prevalence rates of 8.0–52.3 % for any AD (median = 14.9 %), 0.0–34.1 % for GAD (median = 10.6 %), 1.0–15.5 % for AP (median = 2.0 %), 2.0–10.9 % for SP (median = 7.2 %), 1.0–13.5 % for PD (median = 2.1 %), 0.7–3.4 % for OCD (median = 2.0 %), and 5.7 % for PTSD, reported by the only study [
The pooled estimate of the current prevalence of any AD was 26.1 % (p < 0.01, 95 % confidence interval [CI]: 23.4, 29.1), which was based on nine studies [
Combined detection of depression and anxiety in epilepsy patients using the Neurological Disorders Depression Inventory for Epilepsy and the World Health Organization well-being index.
]. The total number of PWE included in those nine studies was 1166, with 50–188 PWE per study (median = 124) (Fig. 2).
Fig. 2Pooled estimate of the current prevalence of any AD in PWE. In nine studies pertaining to any AD, the total number of recruited PWE was 1166, and the current prevalence of any AD was 26.1 %. The heterogeneity among the included studies is evident in the forest plot.
Of the various AD subtypes, GAD had the highest current prevalence (pooled estimate = 18.2 %, p < 0.01, 95 % CI: 16.2, 20.3), which was based on 13 studies [
Combined detection of depression and anxiety in epilepsy patients using the Neurological Disorders Depression Inventory for Epilepsy and the World Health Organization well-being index.
]. The total number of PWE included in those 13 studies was 1771, with 50–243 PWE per study (median = 132) (Fig. 3).
Fig. 3Pooled estimate of the current prevalence of GAD in PWE. In 13 studies pertaining to GAD, the total number of recruited PWE was 1771, and the current prevalence of GAD was 18.2 %. The heterogeneity among the included studies is evident in the forest plot.
The pooled estimates of the current prevalence rates of ADs other than GAD (i.e., AP, SP, PD, and OCD) were also assessed. For AP, the prevalence was 12.5 % (p < 0.01, 95 % CI: 9.8, 15.8), which was based on five studies [
]. The total number of PWEs included in those five studies was 685, with 80–188 PWE per study (median = 147) (Fig. 4-A). For SP, the prevalence was 8.6 % (p < 0.01, 95 % CI: 6.5, 11.4), which was based on four studies [
]. The total number of PWE included in those four studies was 559, with 50–188 PWE per study (median = 161) (Fig. 4-B). For PD, the prevalence was 6.7 % (p < 0.01, 95 % CI: 4.9, 9.0), which was based on seven studies [
]. The total number of PWE included in those seven studies was 835, with 50–188 PWE per study (median = 100) (Fig. 4-C). For OCD, the prevalence was 2.4 % (p < 0.01, 95 % CI: 1.6, 3.7), which was based on seven studies [
]. The total number of PWE included in those seven studies was 953, with 50–188 PWE per study (median = 147) (Fig. 4-D). A meta-analysis of PTSD was not possible because only a single study provided the prevalence of PTSD [
Fig. 4Pooled estimates of the current prevalence rates of AP, SP, PD, and OCD in PWE. In five studies with AP information, the total number of recruited PWE was 685, and the current prevalence of AP was 12.5 % (A). In four studies with SP information, the total number of recruited PWE was 559, and the current prevalence of SP was 8.6 % (B). In seven studies with PD information, the total number of PWE was 835, and the current prevalence of PD was 6.7 % (C). In seven studies with OCD information, the total number of PWE was 953, and the current prevalence of OCD was 2.4 % (D). The heterogeneities among the included studies are evident in the individual forest plots for each AD type, except OCD.
The meta-analyses of any AD and the various AD subtypes each involved 5–13 studies. For OCD, the primary studies were not statistically heterogeneous, according to a Q-value of 4.72 (p = 0.58) and I2 of 0.0 % (Fig. 4-D). However, the primary studies assessing any AD and other AD subtypes were substantially heterogeneous. The Q-value and I2 were 133.38 (p < 0.01) and 94.0 % for any AD (Fig. 2), 86.44 (p < 0.01) and 86.1 % for GAD (Fig. 3), 5.35 (p = 0.15) and 43.9 % for SP (Fig. 4-A), 25.85 (p < 0.01) and 84.5 % for AP (Fig. 4-B), and 25.44 (p < 0.01) and 76.4 % for PD (Fig. 4-C), respectively.
3.5 Publication bias
A funnel plot of the 13 studies that reported the current prevalence rates of GAD revealed that the distribution of the effect sizes was asymmetric and, thus, the risk of publication bias high (Fig. 5). According to the funnel plot, it is possible that studies with small sample sizes and high event rates may not have been published. Egger's regression was calculated to evaluate the publications and revealed a bias of −4.5365871 (p < 0.001), which confirmed the high risk of publication bias. A publication bias analysis was not possible for any AD type or the other AD subtypes because the numbers of primary studies were fewer than 10.
Fig. 5Funnel plot of the included studies that reported the current prevalence of GAD in PWE. In the funnel plot of the 13 studies that provided current prevalence rates for GAD, the distribution of the effect sizes was asymmetric and, thus, the risk of publication bias high.
In the present study, the risk of bias for each of the 15 selected studies was evaluated using Hoy's inventory; all 15 studies had a score of 8 out of 10 points as a “summary assessment”. Because the PWE were recruited from epileptic clinics, the study subjects do not represent the national population. As a result, the risk of bias in those studies was high for the “target population” item, and a score was not assigned to them. In addition, all 15 studies had a high risk for the “random selection” item.
4. Discussion
Epilepsy and anxiety have a bidirectional association [
Prevalence of self-reported epilepsy or seizure disorder and its associations with self-reported depression and anxiety: results from the 2004 Health Styles Survey.
]. Consistent with the diagnostic tool and findings of that study, the present meta-analysis included only studies that used gold standard diagnostic tests and found that the current prevalence of any AD in PWE under the care of epilepsy clinics is 26.1 %.
In PWE, anxiety increases suicidal ideation as well as the degree of perceived stigma [
]. Because anxiety has a negative impact on various characteristics of PWE, it ultimately degrades QOL. In fact, the negative effects of anxiety on QOL are more potent than those of depression, seizure frequency, and demographic variables [
]. Moreover, in the absence of anxiety and depression, the QOL of people with drug-resistant epilepsy is better than that of anxious or depressed people with well-controlled epilepsy [
] have reported prevalence rates of ADs among the general population. The 12-month prevalence rates of any AD, GAD, AP, PD, and separation anxiety disorder are 8.4–21.3 %, 0.2–4.3 %, 0.1–10.5 %, 0.3–3.1 %, and 0.4–8.0 %, respectively, and the life-time prevalence rates are 7.6–33.7 %, 0.3–6.2 %, 0.4–2.6 %, 0.5–5.2 %, and 0.6–13.0 %, respectively. Data regarding the current prevalence of AD in the general population can help with interpretation of the results of the present meta-analysis. According to the Epidemiologic Catchment Area Survey, a population-based study conducted in the United States, the current prevalence of any AD is 7.3 % [
]. Similarly, a meta-analysis of 87 studies reported that the mean prevalence of any AD is 7.3 % (95 % CI: 4.8, 29.1) and the individual prevalence rates 0.9–28.3 % [
]. According to an integration of cross-sectional population surveys conducted in 29 countries by the WHO, the current prevalence of GAD in the general population is 0.8 % [
]. Those authors also reported a prevalence range across these 29 nations of 0.1–3.0 %. The WHO study included an extensive general population survey from the United States that reported a current GAD prevalence of 1.6 % [
]. The Epidemiologic Catchment Area Survey also reported the current prevalence rates of phobias, PD, and OCD to be 6.3 %, 0.5 %, and 1.3 %, respectively [
], and although several studies have evaluated the relationship between epilepsy and ADs, the proportions of different AD subtypes in PWE remain unclear. Thus, the present study aimed to elucidate this issue by integrating information from primary studies and found that the current prevalence of any AD in epilepsy clinics ranges from 8.0 %–52.3 % [
Combined detection of depression and anxiety in epilepsy patients using the Neurological Disorders Depression Inventory for Epilepsy and the World Health Organization well-being index.
]. Furthermore, the pooled estimated current prevalence of any AD in PWE in the present meta-analysis was 26.1 %, which was relatively higher than the prevalence rates reported by the abovementioned population-based studies. More than five of the studies in the present meta-analysis estimated the current prevalence rates of GAD, AP, SP, PD, and OCD. Of these subtypes, the overall estimated prevalence of GAD was the highest at 18.2 %, which was much higher than the prevalence reported by the abovementioned population-based studies [
]. Following GAD, the prevalences of AP, SP, PD, and OCD were highest, with pooled estimates of 12.5 %, 8.6 %, 6.7 %, and 2.4 %, respectively (Fig. 6). These overall estimates were higher than those reported by a study from the United States [
]. Taken together, the present results will contribute to a more accurate picture of the prevalence of comorbid ADs in outpatient epilepsy clinics and may help with the management of epilepsy in actual practice. Additionally, these findings will be a useful reference for future research investigating ADs in PWE.
Fig. 6Comparisons of the pooled estimated prevalences among any AD and the AD subtypes in PWE. In nine studies related to any AD, the current prevalence of any AD was 26.1 %. Comparisons of the AD subtypes revealed that the overall estimated prevalence of GAD was 18.2 %, which was the highest of the AD subtypes, followed by AP, SP, PD, and OCD in that order.
It is important to note that the criteria used to classify AD subtypes have changed over time. As a result, the criteria used for the classification of AD subtypes at a specific time point may affect the prevalence of any AD in different epidemiological studies. In the present meta-analysis, nine of the studies provided an overall estimate for any AD, of which seven used the DSM-IV and two used both the DSM-IV and ICD-10 classification systems to determine the diagnoses. The DSM-IV includes OCD and PTSD in the AD category [
], and thus, the seven studies that used only the DSM-IV defined “any AD” as GAD, AP, SP, PD, OCD, and PTSD, in a manner consistent with the purposes of the present meta-analysis. The ICD-10 categorizes OCD and PTSD as separate disorders, different than AD [
]. The two studies that diagnosed ADs based on both the DSM-IV and ICD-10 criteria reported the prevalence of OCD separately from those for ADs. In those two studies, none of the PWE had PTSD [
Combined detection of depression and anxiety in epilepsy patients using the Neurological Disorders Depression Inventory for Epilepsy and the World Health Organization well-being index.
People may experience anxiety as a result of their diseases, and there is a specific relationship between epilepsy and anxiety. High incidences of seizures, recent seizures, and symptomatic focal epilepsy are factors that contribute to anxiety in PWE, which can develop into an AD [
]. Therefore, accurate information will likely improve anxiety related to specific epileptic conditions, and controlling seizures can improve ictal anxiety.
Diagnostic lag may also occur in PWE and can induce anxiety that may evolve into GAD [
]. Even after diagnosis, PWE feel epilepsy-related anxiety that includes fears of future seizures, disease progression, and/or AED-related AEs. These anxiety symptoms result in the persistence of GAD in PWE [
]. Furthermore, complex partial seizures, high seizure frequency, recent seizures, multiple AEDs, and AED-related AEs are associated with the development of GAD. However, if the epilepsy is idiopathic, it is less likely that GAD will occur [
Ictal fear is a seizure that is similar to the panic attacks associated with PD. Although it is difficult to distinguish ictal fear from panic attacks, there are some differences. For example, the former is briefer and often progresses to other seizures, whereas the latter is often accompanied by depression and is associated with the severity of depression [
]. Similarly, AP and SP in PWE are different than in people without epilepsy. In PWE, a common phobia is fear of seizures or outdoor accidents, which becomes a variation of AP or SP [
]. If accompanied by severe psychological trauma during the first seizure, PWE may also experience PTSD potentially resulting in psychological strain and interfering with treatment compliance. In this case, proper psychiatric treatment is essential [
]. Accordingly, a better understanding of the specific associations between AD subtypes and epilepsy will be useful for the differential diagnosis of ADs in PWE. The present findings regarding the prevalence of AD subtypes in PWE will provide additional information about these differential diagnoses.
When diagnosing psychiatric disorders, including ADs, structured interviews such as the MINI and SCID are considered the gold standard. The MINI evaluates current psychopathology and is often used by researchers in clinical trials due to its simplicity [
The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10.
]. The SCID is primarily used as a diagnostic tool for research rather than in clinical settings because it takes more time to complete than does the MINI [
A fixed-effects model was used as the predetermined method of analysis, because the included studies were conducted only in routine outpatient epilepsy clinics. However, in the statistical analyses, there was substantial heterogeneity among the included studies of ADs, excluding OCD. This is likely because the underlying PWE populations in the selected studies differed in many aspects, even though all PWE were treated in outpatient epilepsy clinics, contributing to the heterogeneity. In the general population, ADs are more prevalent in females than males and in elderly than younger adults [
]. Thus, it is also possible that the proportion of female subjects, patients’ ages, and cultural differences contributed to the heterogeneity. In the 15 studies assessed, the proportions of females (32.3–67.9 %) and mean ages (26.6–50.5 years) of the PWE ranged widely, and the studies were conducted in various continents.
The present meta-analysis has several limitations that should be noted. First, because the focus was on epilepsy clinics, population-based studies were excluded, and thus, the AD-related information may be limited. Second, only the current prevalence rates of ADs were provided, precluding identification of lifetime and past-year prevalence rates. Additionally, most of the studies included in this meta-analysis did not use control groups, rendering it difficult to identify differences between PWE and the general population. Despite these limitations, the present results may provide useful information that will be referred to when treating PWE in clinical practice. In this meta-analysis, studies that investigated the current prevalence of ADs in PWE under routine care in outpatient epilepsy clinics were included and assessed, and only studies that used gold standard methods for diagnosing ADs were included. The results provided information about the prevalence rates of various AD subtypes in PWE that will further the current understanding of the relationship between epilepsy and anxiety.
Declaration of Competing Interest
The authors report no financial interests or potential conflicts of interest related to the present study.
Acknowledgments
The authors have no acknowledgments to present.
Appendix A. Search Strategies
MEDLINE
#1. epilepsy[MeSH Terms] OR (epilepsy[tiab] OR epilepsies[tiab])
#2. ("anxiety disorders"[MeSH Terms] OR anxiety[MeSH Terms] OR “stress disorder, post-traumatic”[MeSH Terms]) OR (anxiety[tiab] OR anxieties[tiab] OR anxious[tiab] OR GAD[tiab] OR panic[tiab] OR PD[tiab] OR agoraphobia[tiab] OR phobia[tiab] OR “obsessive compulsive”[tiab] OR “obsessive-compulsive”[tiab] OR OCD[tiab] OR “posttraumatic stress”[tiab] OR “post-traumatic stress”[tiab] OR PTSD[tiab])
#3. ("diagnosis, differential"[MeSH Terms] OR diagnosis[tiab]) OR (M.I.N.I.[tiab] OR MINI[tiab] OR "Mini International Neuropsychiatric Interview"[tiab]) OR (S.C.I.D.[tiab] OR SCID[tiab] OR "structured clinical interview"[tiab]) OR (C.I.D.I.[tiab] OR CIDI[tiab] OR "Composite International Diagnostic Interview"[tiab])
#4. #1 AND #2 AND #3
EMBASE
#1. (epilepsy/exp) OR (epilepsy:ab,ti OR epilepsies:ab,ti)
#2. (‘anxiety disorders’/exp OR anxiety/exp OR ‘post-traumatic stress disorder’/exp) OR (anxiety:ab,ti OR anxieties:ab,ti OR anxious:ab,ti OR GAD:ab,ti OR panic:ab,ti OR PD:ab,ti OR agoraphobia:ab,ti OR phobia:ab,ti OR ‘obsessive compulsive’:ab,ti OR ‘obsessive-compulsive’:ab,ti OR OCD:ab,ti OR ‘posttraumatic stress’:ab,ti OR ‘post-traumatic stress’:ab,ti OR PTSD:ab,ti)
#3. (‘differential diagnosis’/exp) OR (diagnosis:ab,ti) OR (M.I.N.I.:ab,ti OR MINI:ab,ti OR ‘Mini International Neuropsychiatric Interview’:ab,ti) OR (S.C.I.D.:ab,ti OR SCID:ab,ti OR ‘structured clinical interview’:ab,ti) OR (C.I.D.I.:ab,ti OR CIDI:ab,ti OR ‘Composite International Diagnostic Interview’:ab,ti)
#4. #1 AND #2 AND #3 AND [embase]/lim NOT [medline]/lim
Cochrane Library
#1. (MeSH descriptor: [epilepsy] explode all trees) OR (epilepsy OR epilepsies:ti,ab,kw)
#2. (MeSH descriptor: [anxiety disorders] explode all trees) OR (MeSH descriptor: [anxiety] explode all trees) OR (MeSH descriptor: [stress disorder, post-traumatic] explode all trees) OR (anxiety OR anxieties OR anxious:ti,ab,kw OR GAD OR panic:ti,ab,kw OR PD:ti,ab,kw OR agoraphobia:ti,ab,kw OR phobia:ti,ab,kw OR “obsessive compulsive”:ti,ab,kw OR OCD:ti,ab,kw OR ‘posttraumatic stress’:ti,ab,kw OR ‘post-traumatic stress’:ti,ab,kw OR PTSD:ti,ab,kw)
#3. (MeSH descriptor: [diagnosis, differential] explode all trees) OR
(diagnosis OR M.I.N.I. OR MINI OR "Mini International Neuropsychiatric Interview" OR S.C.I.D. OR SCID OR "structured clinical interview" OR C.I.D.I. OR CIDI OR "Composite International Diagnostic Interview":ti,ab,kw)
#4. #1 AND #2 AND #3
Web of Science
#1. TOPIC: (epilepsy) OR TITLE: (epilepsy OR epilepsies)
#2. TOPIC: ("anxiety disorders" OR anxiety OR “post-traumatic stress disorder”) OR TITLE: (anxiety OR anxieties OR anxious OR GAD OR panic OR PD OR agoraphobia OR phobia OR “obsessive compulsive” OR “obsessive-compulsive” OR OCD OR “posttraumatic stress” OR “post-traumatic stress” OR PTSD)
#3. TOPIC: ("diagnosis, differential") OR TITLE: (diagnosis OR M.I.N.I. OR MINI OR "Mini International Neuropsychiatric Interview" OR S.C.I.D. OR SCID OR "structured clinical interview" OR C.I.D.I. OR CIDI OR "Composite International Diagnostic Interview")
#4. #1 AND #2 AND #3
SCOPUS
#1. INDEXTERMS(epilepsy) OR (TITLE-ABS-KEY(epilepsy) OR TITLE-ABS-KEY(epilepsies))
#2. (INDEXTERMS("anxiety disorders") OR INDEXTERMS("anxiety") OR INDEXTERMS("post-traumatic stress disorder")) OR (TITLE-ABS-KEY(anxiety) OR TITLE-ABS-KEY(anxieties) OR TITLE-ABS-KEY(anxious) OR TITLE-ABS-KEY(GAD) OR TITLE-ABS-KEY(panic) OR TITLE-ABS-KEY(PD) OR TITLE-ABS-KEY(agoraphobia) OR TITLE-ABS-KEY(phobia) OR TITLE-ABS-KEY(“obsessive compulsive”) TITLE-ABS-KEY(“obsessive-compulsive”) OR TITLE-ABS-KEY(OCD) OR TITLE-ABS-KEY(“posttraumatic stress”) OR TITLE-ABS-KEY(“post-traumatic stress”) OR TITLE-ABS-KEY(PTSD))
#3. (INDEXTERMS("diagnosis, differential") OR TITLE-ABS-KEY(diagnosis)) OR (TITLE-ABS-KEY(M.I.N.I.) OR TITLE-ABS-KEY(MINI) OR TITLE-ABS-KEY("Mini International Neuropsychiatric Interview")) OR (TITLE-ABS-KEY(S.C.I.D.) OR TITLE-ABS-KEY(SCID) OR TITLE-ABS-KEY("structured clinical interview")) OR (TITLE-ABS-KEY(C.I.D.I.) OR TITLE-ABS-KEY(CIDI) OR TITLE-ABS-KEY("Composite International Diagnostic Interview"))
#4. #1 AND #2 AND #3
Appendix B. Supplementary data
The following is Supplementary data to this article:
Prevalence of self-reported epilepsy or seizure disorder and its associations with self-reported depression and anxiety: results from the 2004 Health Styles Survey.
Combined detection of depression and anxiety in epilepsy patients using the Neurological Disorders Depression Inventory for Epilepsy and the World Health Organization well-being index.
The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10.
30710-1&id=gr1.jpg){kind=link}
30710-1&id=gr2.jpg){kind=link}
30710-1&id=gr3.jpg){kind=link}
30710-1&id=gr4.jpg){kind=link}
30710-1&id=gr5.jpg){kind=link}
30710-1&id=gr6.jpg){kind=link}