Epilepsy & Behavior 35 (2014) 1–5
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Risk factors for psychological distress in community-treated epilepsy Cameron J. Lacey a,b,⁎, Michael R. Salzberg a, Wendyl J. D'Souza b a b
Department of Psychiatry, St Vincent's Hospital Melbourne, University of Melbourne, 59 Victoria Parade, Fitzroy, 3065 Melbourne, Victoria, Australia Department of Medicine, St Vincent's Hospital Melbourne, University of Melbourne, 59 Victoria Parade, Fitzroy, 3065 Victoria, Australia
a r t i c l e
i n f o
Article history: Received 20 February 2014 Revised 28 March 2014 Accepted 31 March 2014 Available online xxxx Keywords: Epilepsy Psychiatric comorbidity Community Risk factors
a b s t r a c t The study aimed to determine risk factors for psychological distress in a community-treated sample of patients with epilepsy. This study investigated the Tasmanian Epilepsy Register participants. Participants included were as follows: aged 13 years and over, able to complete the individual computer-assisted participant interview, and diagnosed with epilepsy following an epilepsy specialist review of the diagnostic epilepsy interview, which was interpreted using standardized diagnostic guidelines. Psychological distress was assessed with the Kessler-10 questionnaire. Risk factors were grouped into four domains: sociodemographic factors, diseaserelated factors, psychological factors, and treatment-related factors. High or very high levels of psychological distress were reported by 22% of the participants, with 7.8% having very high distress. The regression model showed that psychological distress was significantly associated with female gender (F = 18.1, p b 0.001), diabetes mellitus (F = 8.7, p = 0.003), intellectual disability (F = 7.1, p = 0.06), and not receiving phenytoin (F = 5.1, p = 0.02). While the model was significant (F = 5.78, p b 0.001), only 11% of the variance of the K-10 score was explained by these factors (adjusted R-squared = 0.11). This study identifies female gender and comorbid medical conditions as risk factors for psychological distress and the use of phenytoin as a protective factor. The few factors identified and the limited variance explained suggest that a focus on epilepsy-related variables is unlikely to explain key influences underlying psychiatric comorbidity in patients with epilepsy. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The association between epilepsy and psychiatric comorbidity is well recognized with increased rates of a range of psychiatric disorders including depression, generalized anxiety disorder, and panic disorder [1]. Psychiatric comorbidity is associated with decreased quality-oflife, diminished medication adherence, poorer treatment outcomes, increased health service use, increased cognitive complaints, increased risk of other chronic diseases such as cardiovascular disease, and suicide [2–6]. The risk factors that contribute to psychological distress in patients with epilepsy remain unclear. Most research has been derived from hospital- or tertiary-based populations and is vulnerable to important sampling biases. For example, the rate of depression was found to be 58% in surgical patients [7] compared with 11% in community samples [8]. Furthermore, there are discrepancies in reported risk factors for psychiatric comorbidity between hospital-based samples and community studies. This may be due to the practical challenges of obtaining valid epilepsy-related variables in representative community-based studies with large sample sizes. ⁎ Corresponding author at: Department of Psychological Medicine, University of Otago, Christchurch, PO Box 4345, Christchurch, New Zealand. Tel.: +64 3 3720400; fax: +64 3 3720407. E-mail address: [email protected] (C.J. Lacey).
http://dx.doi.org/10.1016/j.yebeh.2014.03.025 1525-5050/© 2014 Elsevier Inc. All rights reserved.
The Tasmanian Epilepsy Register (TER) is one of a handful of studies utilizing a sample of people with community-treated epilepsy, and initial results confirmed that the rate of psychological distress is greater than that in the general population [4]. The TER is a sufficiently large, well-classified sample of people with epilepsy to investigate risk factors for psychological distress which may inform efforts at reducing this important health disparity [9]. The study aimed to determine risk factors for psychological distress in a community-treated sample of patients with epilepsy and to determine if rates of psychological distress differ across treatment settings.
2. Methods A description of the Tasmanian Epilepsy Register methodology has been previously published [10]. This study examined “psychological distress” in the TER population as revealed by the Kessler-10 (K-10) psychiatric screening tool. For convenience, the study used the term “psychological distress” to denote the symptoms assessed by the K-10 instrument [11]. The K-10 is an ideal measure to begin investigating psychiatric comorbidity as it captures most depressive and anxiety illness, has the advantages of brevity, and allows comparison with large community surveys that also employed this instrument [12]. While the concept of “psychological distress” includes both depression and anxiety disorders, research findings in the general population have
2
C.J. Lacey et al. / Epilepsy & Behavior 35 (2014) 1–5
established that there is some overlap in risk factors for both conditions [13]. 2.1. Inclusion criteria The inclusion criteria for this study were the following: TER participants of at least secondary school age (13 years and over), able to complete the individual computer-assisted participant interview, and diagnosed with epilepsy following an epilepsy specialist (WD) review of the diagnostic epilepsy interview, which was interpreted using standardized diagnostic guidelines [14]. Participants who were unable to be interviewed due to intellectual disability or communication difficulties were excluded. 2.2. Psychological distress Psychological distress was assessed with the K-10, a 10-question screening scale developed for the US National Health Interview Survey. The K-10 performs well in detecting DSM-IV anxiety and mood disorders (as validated by the Composite International Diagnostic Interview (CIDI)), with a high area under receiver operating characteristic curve (AUC) = 0.90. As in the Australian National Health Surveys [12], the K10 total was grouped into the following levels of psychological distress: low (K-10 = 10–15), moderate (K-10 = 16–21), high (K-10 = 22–30), and very high (K-10 = 31–50) or as a dichotomous variable in which participants were classified as having either low–moderate or high– very high levels of psychological distress. 2.3. Risk factors The classification proposed by Hermann et al. was modified to group potential risk factors divide into four domains [15]. 2.3.1. Sociodemographic factors Socioeconomic status was obtained from the participants' postcodes using the Index of Relative Socioeconomic Advantage/Disadvantage (SEIFA) 2001 developed by the Australian Bureau of Statistics [16]. This provides an estimate of each individual's socioeconomic status based on a measure of the relative social and economic well-being of the population of their postcode area taken as a whole. This, in turn, is derived from attributes such as the proportions in each area with low income, low educational attainment, high unemployment, and jobs in relatively unskilled occupations. High scores on the Index of Relative Socioeconomic Advantage/Disadvantage indicate higher socioeconomic status and less disadvantage. This study transformed SEIFA values into quintiles for the Tasmanian population as previously described in the Tasmanian Epilepsy Register methodology [10]. This study also examined the association with the three main geographical/administrative regions of Tasmania (Southern, Northern, and North-western) determined by participants' postcodes. Postcodes of participants were also used to determine their geographical remoteness using the Australian Standard Geographical Classification (ASGC) ‘Remoteness Structure’. This measures the remoteness based on the physical distance to the nearest urban center [17] and is classified into the following: major cities of Australia, inner regional Australia, outer regional Australia, remote Australia, very remote Australia, and migratory. 2.3.2. Disease-related factors Participants completed a detailed epilepsy diagnostic telephone questionnaire administered by trained interviewers. This provided detailed descriptive seizure data to enable an epilepsy specialist to determine the presence of epilepsy, seizure-onset type (generalized, focal, or uncertain), presence of an idiopathic generalized epilepsy (IGE) syndrome, age at onset of epilepsy, duration of epilepsy, seizure frequency, and antecedent epilepsy risk factors [14]. A diagnosis of epilepsy was made by blind interpretation on two occasions by an experienced epileptologist
applying standardized guidelines [14]. The diagnostic interview was a modified version of two diagnostic questionnaires, each previously shown to have substantial to very high agreement with physicianbased diagnoses in classifying seizure types and broad epilepsy-onset types [18,19]. This modified version showed almost perfect agreement in diagnosing epilepsy (κ = 0.94), seizure-onset types (κ = 0.84), simple or complex partial seizures (κ = 0.87), any generalized nonconvulsive seizure (κ = 0.82), IGE (κ = 0.82) and substantial agreement for secondarily generalized seizures (κ = 0.74), and generalized tonic–clonic seizures (κ = 0.79) [14]. The frequency of both convulsive and nonconvulsive seizures in the last 12 months was grouped into the following: none, less than monthly, or more than monthly. The epilepsy diagnostic interview also elicited information about antecedent epilepsy risk factors including the presence of other medical conditions that have been associated with seizures. These included each participant's history of febrile convulsion, serious head injury before first seizure, cerebrovascular accident (CVA), brain tumor, brain surgery, meningitis or encephalitis, cerebral palsy, intellectual disability, multiple sclerosis, diabetes mellitus, coma, polio, and arteriovenous malformation (AVM). The presence of these conditions was summed into the variable “number of antecedent epilepsy risk factors”. 2.3.3. Psychological factors Although alcohol abuse results in psychiatric disorder by both neurobiological and psychological mechanisms, it will be considered here for convenience [20]. Alcohol use was assessed with the Alcohol Use Disorders Identification Test (AUDIT) questionnaire, which has been widely used to screen for hazardous and harmful drinking [21]. The AUDIT scores of eight or greater were considered positive for hazardous and harmful drinking [21]. 2.3.4. Treatment-related factors The Health Insurance Commission provided data on the individual anticonvulsant medications dispensed in the 12-month study period for each participant. These were recorded as dichotomous variables as well as combined into the total number of anticonvulsant medications. The Health Insurance Commission records “prescribing doctor provider type” for all prescriptions, and this utilizes the vocation speciality recorded with each doctor's medical registration information. For the 12-month study period, these data were used to estimate the setting in which the Tasmanian Epilepsy Register participants received their medical care as studies suggest that the medical practitioner writing anticonvulsant drug prescriptions is most likely to also be responsible for disease supervision and follow-up [22]. Patients receiving care from a general practitioner only were compared with those receiving care from a general practitioner and/or a specialist. 2.4. Statistical analysis Univariate statistics were initially used to test for significant associations with psychological distress. The associations between K-10 total (continuous outcome) and predictor variables are presented using Spearman's rank correlation, independent t tests, and ANOVA tests for ordinal, dichotomous, and categorical variables, respectively. The associations between “high or very high” psychological distress (dichotomous outcome) and individual predictor variables utilized independent t tests, Mann–Whitney U test, and chi-squared test for continuous, ranked, and categorical variables, respectively. Boxplots show median, quartiles, range, and outliers (defined by 1.5 times the interquartile range outside the quartiles). General linear regression (SPSS© Version 19) was utilized to assess predictors of the level of psychological distress. Any predictor variable with a p value b 0.1 was selected for inclusion in the linear regression. This study was granted ethical approval by the University of Tasmania Human Research Ethics Committee.
C.J. Lacey et al. / Epilepsy & Behavior 35 (2014) 1–5
3. Results 3.1. Study population Of the 1180 patients enrolled on the Tasmanian Epilepsy Register, 43 withdrew prior to interview, 36 died, and 254 were excluded (90 were b13 years old, 164 had intellectual disability or communication difficulties), leaving 847 potential subjects. Six hundred sixty patients completed the K-10 (response rate = 78%), and 554 of these were diagnosed with epilepsy following specialist review of the epilepsy diagnostic interview. The mean age of participants was 51 years (range = 13–102 years), with 48% of male gender, similar to the Tasmanian Epilepsy Register cohort. There were no significant differences between participants and those excluded for mean age, gender ratio, and socioeconomic status.
3
of psychological distress between participants with IGE compared with those without IGE. There was no significant correlation between age at onset or duration of epilepsy and level of psychological distress. Convulsive seizures occurring more than monthly in the last 12 months were associated with a higher level of psychological distress (p = 0.001, see Fig. 1). However, there was no association between nonconvulsive seizure frequency in the last 12 months and level of psychological distress. The majority of participants (56%) had at least one antecedent epilepsy risk factor (see Supplementary Table). Only the presence of CVA (p = 0.006), intellectual disability (p = 0.002), and diabetes mellitus (p = 0.008) were associated with increased psychological distress. Antecedent epilepsy risk factors were summed to give the “number of antecedent epilepsy risk factors”. There was a significant association between higher number of risk factors and increased level of psychological distress (p = 0.006).
3.2. Psychological distress 3.5. Alcohol abuse High or very high levels of psychological distress were reported by 22% of the participants, with 7.8% having very high distress.
Hazardous or harmful alcohol use was reported in only 4% of the participants and was not associated with level of psychological distress.
3.3. Sociodemographic risk factors 3.6. Treatment-related risk factors Females had a higher rate of high–very high psychological distress compared with males (28.3% versus 18.5%, p = 0.003). There was a small but significant correlation between lower levels of psychological distress and greater age (r = 0.29, p = 0.001). There was no significant correlation between psychological distress and socioeconomic status or geographical region. Almost all participants had a “remoteness” classification of either “inner” or “outer regional Australia”. Only two participants lived in “remote Australia”, and these were not included in the analysis because of the small number. There was no association between living in “inner regional Australia” and increased psychological distress compared with living in “outer regional Australia”. 3.4. Disease-related risk factors There were no differences in the mean level of psychological distress between participants who were diagnosed with epilepsy following specialist review of the diagnostic epilepsy interview data and those with uncertain or other diagnoses. The associations between disease-related risk factors and psychological distress are presented in Table 1. There was no significant association between seizure-onset type and level of psychological distress. There was no significant difference in the level
Only the use of phenytoin was associated with lower levels of psychological distress (p = 0.001). Participants who received phenytoin were more likely to be male (p = 0.07), report less frequent convulsive seizures (p b 0.001), and receive only one anticonvulsant medication (p = 0.004). No single anticonvulsant was associated with increased psychological distress. The mean number of anticonvulsant medications was 1.4 (range = 0–5). Higher number of anticonvulsant medications was associated with increasing psychological distress (p = 0.03). The majority of participants were treated by a general practitioner only (77%). However, participants receiving treatment from specialists alone or in combination with a general practitioner (23%) had an increased rate of high–very high psychological distress compared with those seeing only a general practitioner (p = 0.03). 3.7. Multivariate associations The variables with univariate association with psychological distress (p b 0.1) were entered into linear regression modeling with age and gender entered as covariates. These variables were geographical
Table 1 Epilepsy-related risk factors and psychological distress. Characteristic
Low–moderate psychological distress
High–very high psychological distress
p
Seizure-onset type Generalized Focal Uncertain
101 (79%) 282 (76%) 46 (82%)
27 (21%) 88 (24%) 10 (18%)
NSa
Epilepsy syndrome IGE Other syndromes
98 (78%) 331 (77%)
27 (22%) 98 (23%)
NSa
Convulsive seizure frequency (last 12 months) None 136 (72%) Less than monthly 284 (82%) More than monthly 9 (50%)
52 (28%) 64 (18%) 9 (50%)
0.01a
Nonconvulsive seizure frequency (last 12 months) None 244 (76%) Less than monthly 147 (81%) More than monthly 38 (73%)
76 (24%) 35 (19%) 14 (27%)
NSa
Numbers are n (%). IGE = idiopathic generalized epilepsy. a χ2.
Fig. 1. Boxplot of K-10 total by convulsive seizure frequency.
4
C.J. Lacey et al. / Epilepsy & Behavior 35 (2014) 1–5
remoteness, convulsive seizure frequency, CVA, intellectual disability, diabetes mellitus, number of antecedent epilepsy risk factors, use of phenytoin, and number of anticonvulsant medications. The regression model showed that psychological distress was significantly associated with female gender (F = 18.1, p b 0.001), diabetes mellitus (F = 8.7, p = 0.003), intellectual disability (F = 7.1, p = 0.06), and not receiving phenytoin (F = 5.1, p = 0.02). The strongest predictor of psychological distress was female gender; however, this explained only 3% of the variance of the K-10 score. While the model was significant (F = 5.78, p b 0.001), only 11% of the variance of the K-10 score was explained by these factors (adjusted R-squared = 0.11).
4. Discussion This study identifies female gender and a history of other medical conditions (CVA, diabetes, and intellectual disability) as risk factors for psychological distress and the use of phenytoin as a protective factor. However, these factors predicted only 11% of the variance in psychological distress and suggest that further research utilizing a broader range of potential risk factors is required to identify potential contributors to psychological distress. The few factors identified and the limited variance explained suggest that focusing solely on epilepsy-related variables is unlikely to explain the key influences underlying psychiatric comorbidity in patients with epilepsy. Psychological factors have been more consistently associated with psychiatric comorbidity in previous reviews [15], and a more detailed study, incorporating recent findings of risk factors for depression in the general population, is underway. Although not directly comparable, the rate of psychological distress found in this study is consistent with that found in other community studies. One of the more rigorous community studies reported a higher rate of depression in a USA community sample of persons with epilepsy (36%) than in healthy controls (12%), although similar to a control group of patients with asthma (28%) [23]. This is also consistent with a population-based study which estimated a 32% rate of psychiatric comorbidity in a nationwide sample of patients with epilepsy using the Norwegian Prescription Database [24]. Previous studies reporting a higher rate of psychiatric comorbidity in tertiary care compared with primary care relied on indirect comparison of different sample groups and differing study designs. In fact, this comparison has been estimated across at least three different treatment settings: patients being assessed with video-EEG or for epilepsy surgery, patients attending epilepsy outpatient clinics, other groups of patients with epilepsy including support groups, and patients attending general practitioners. The novel recruitment strategy (national prescription database), validated epilepsy diagnostic classification methods, and use of a standardized measure of psychiatric comorbidity across these differing treatment settings allow the first direct measure of this phenomenon and support the gradient of psychological distress from primary to specialized care. This gradient of distress across treatment settings highlights the importance of sample selection and potential for ascertainment bias. Seizure type and epilepsy syndrome are often poorly addressed in community studies, with few studies attempting this and most relying on patient self-report [25,26] or clinical record review [27]. The benefit of utilizing the Tasmanian Epilepsy Register is the availability of validated epilepsy syndrome and seizure-onset data that reflect the interpretation by an epilepsy specialist of seizure symptoms and signs directly from the patients and a witness. Ideally, the use of a structured diagnostic interview combined with results of additional investigations (including EEG and MRI) would be combined to provide greater diagnostic precision. However, this study's finding of no association between psychological distress and seizure-onset type or presence of IGE syndrome suggests that any contribution of these factors to the risk of depression is likely to be small if it exists at all. There is a much larger literature derived from more highly selected samples with more accurate seizure type and epilepsy syndrome characterization, with many advocating
that psychiatric comorbidity is more common in focal epilepsy than in other types and syndromes [28], although this is also debated [29]. There are several limitations to this study. Primarily, it is crosssectional; thus, it is not possible to draw inferences about causation or the nature of the association between psychological distress and potential risk factors. As previously noted, the K-10 identifies symptoms of anxiety or depression but does not generate a psychiatric diagnosis. While there is some overlap in risk factors for both conditions in the general population and in people with chronic disease [13,30], this has not been well established in people with epilepsy. This study also relies on self-report for the assessment of seizure frequency. Accurately determining the frequency of seizures is not simple, particularly for recall of seizures in which people lose awareness or have cognitive dysfunction features or seizures that occur exclusively during sleep [31]. Additionally, for those experiencing multiple seizure types, it is difficult to determine the individual frequency of each. Finally, the presence of depression has been shown to bias perception of seizure frequency after adjusting for seizure type and recency [26]. There are also potential limitations with the method of identifying medications used by participants. It was not possible to determine dosage, adherence, and blood levels, which may confer additional risk for development of adverse effects including psychological distress. Additionally, as medications for the entire year were recorded in the Health Insurance Commission database, this may capture sequential trials of individual medications rather than reflecting true polydrug therapy. Furthermore, only small numbers of participants received some anticonvulsants that have been associated with depression, such as primidone, phenobarbitone, and topiramate, or predate their introduction in Australia (levetiracetam), and, subsequently, there is limited power to detect associations between individual medications and psychological distress. The few risk factors for psychological distress identified in this study are largely consistent with previous studies of psychiatric comorbidity in people with epilepsy although there are notable differences. The association between female gender and psychological distress is consistent with risk factors for depression identified in the general population [32], people with chronic disease [33], and other community samples with epilepsy [23,34]. Living in an urban region has also been established as a risk factor for depression in the general population [32]; however, this study did not find evidence of an association between psychological distress and geographical remoteness. This may reflect the limitation of the measure of remoteness for Tasmania [17]. However, there may also be other factors contributing to the decision where people with epilepsy live, such as the proximity to health care. At least part of the failure to demonstrate further risk factors may be attributable to the lack of key sociodemographic measures in the TER, such as employment and relationship status, which have been consistently linked with depression in the general population [32]. The association of additional comorbidities (diabetes, intellectual disability, and stroke) with psychological distress reinforces an emerging awareness of the need to consider the range of comorbidities seen in people with epilepsy [5]. This study examined only conditions that are risk factors for development of epilepsy, but 25% of the participants reported having two or more such conditions, and over half (56%) reported at least one. Even higher rates of comorbidity were reported in a household survey in the US, the National Comorbidity Survey — Replication (NCS-R), with 41% of people with epilepsy reported having four or more physical comorbidities [35]. This may be explained by the greater number of physical comorbidities screened for in the NCS-R study. The association of phenytoin with lower psychological distress is an intriguing finding, and whether this is a true drug effect or whether psychological distress leads to help-seeking, and increased likelihood of medication switch to ‘newer’ anticonvulsant medication warrants further investigation. It is also possible that the phenytoin's apparent protective factor is attributable to the associations between phenytoin use and male gender, lower convulsive seizure frequency, and anticonvulsant monotherapy. There is some literature suggesting an
C.J. Lacey et al. / Epilepsy & Behavior 35 (2014) 1–5
antidepressant effect for phenytoin, which may explain its apparent protective effect against psychological distress found in this analysis [36]. There is also some evidence from other studies that other anticonvulsant medications, such as lamotrigine, may protect against depression and anxiety [37,38]. However, the limited numbers of study participants receiving lamotrigine (n = 97) and the study's limited ability to capture the use of newer anticonvulsant medication with proposed therapeutic psychotropic effects limit the study's power to detect such effects. Understanding the factors associated with psychological distress may assist developing strategies to optimize pharmacotherapy given the association between depression comorbidity and treatment resistance [3,39,40]. 5. Conclusions Despite finding associations between gender, intellectual disability, diabetes mellitus, and psychological distress and a protective effect of phenytoin, little of the variance in psychological distress is linked with these factors. In future studies of psychiatric comorbidity in people with epilepsy, there remains a need for adoption of a broader range of risk factors and, in particular, detailed psychological variables [15]. Ethical publication statement We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.yebeh.2014.03.025. Acknowledgments Cameron Lacey has received support from a Lundbeck Neuropsychiatry Research Fellowship. Wendyl D'Souza has received support from a Menzies Research Institute NHMRC Capacity Building Grant and an NHMRC Health Professional Fellowship. The Tasmanian Epilepsy Register was supported by The Booth Estate Launceston, The Royal Hobart Hospital Research Foundation, GSK Neurology, and the Clifford Craig Medical Research Trust, North West Tasmania. We also thank our research assistant Pam McDonald for her considerable efforts in participant liaison, data processing, and management. Conflict of interest statement Associate Professor D'Souza has received travel, investigatorinitiated, and speaker honoraria from UCB Pharma; educational grants from Novartis Pharmaceuticals, Pfizer Pharmaceuticals, and SanofiSynthelabo; and educational, travel, and fellowship grants from GSK Neurology Australia and honoraria from SciGen Pharmaceuticals. The remaining authors have no conflicts of interest. References [1] Ettinger AB, Kanner AM. Psychiatric issues in epilepsy: a practical guide to diagnosis and treatment. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2006. [2] Taylor RS, Sander JW, Taylor RJ, Baker GA. Predictors of health-related quality-of-life and costs in adults with epilepsy: a systematic review. Epilepsia 2011;52:2168–80. [3] Hitiris N, Mohanraj R, Norrie J, Sills GJ, Brodie MJ. Predictors of pharmacoresistant epilepsy. Epilepsy Res 2007;75:192–6. [4] Lacey CJ, Salzberg MR, Roberts H, Trauer T, D'Souza WJ. Psychiatric comorbidity and impact on health service utilization in a community sample of patients with epilepsy. Epilepsia 2009;50:1991–4. [5] Ottman R, Lipton RB, Ettinger AB, Cramer JA, Reed ML, Morrison A, et al. Comorbidities of epilepsy: results from the Epilepsy Comorbidities and Health (EPIC) survey. Epilepsia 2011;52:308–15. [6] Christensen J, Vestergaard M, Mortensen PB, Sidenius P, Agerbo E. Epilepsy and risk of suicide: a population-based case–control study. Lancet Neurol 2007;6:693–8. [7] Victoroff JMD. DSM-III-R psychiatric diagnoses in candidates for epilepsy surgery: lifetime prevalence. Neuropsychiatry Neuropsychol Behav Neurol 1994;7:87–97.
5
[8] Mensah SA, Beavis JM, Thapar AK, Kerr M. The presence and clinical implications of depression in a community population of adults with epilepsy. Epilepsy Behav 2006;8:213–9. [9] Barry JJ, Ettinger AB, Friel P, Gilliam FG, Harden CL, Hermann B, et al. Consensus statement: the evaluation and treatment of people with epilepsy and affective disorders. Epilepsy Behav 2008;13(Suppl. 1):S1–S29. [10] D'Souza WJ. Epidemiology of epilepsy in Tasmania: a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Epidemiology at Massey University at Massey University, Wellington, New Zealand; 2008. [11] Kessler RC, Andrews G, Colpe LJ, Hiripi E, Mroczek DK, Normand SL, et al. Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychol Med 2002;32:959–76. [12] Australian Bureau of Statistics. Information Paper: Use of the Kessler Psychological Distress Scale in ABS Health Surveys. Canberra: 2012 4817.0.55.001. [13] Kendler KS, Gardner CO, Prescott CA. Toward a comprehensive developmental model for major depression in women. Am J Psychiatry 2002;159:1133–45. [14] D'Souza WJ, Stankovich J, O'Brien TJ, Bower S, Pearce N, Cook MJ. The use of computer-assisted-telephone-interviewing to diagnose seizures, epilepsy and idiopathic generalized epilepsy. Epilepsy Res 2010;91:20–7. [15] Hermann BP, Seidenberg M, Bell B. Psychiatric comorbidity in chronic epilepsy: identification, consequences, and treatment of major depression. Epilepsia 2000;41(Suppl. 2):31–41. [16] Statistics ABo. Information paper census of population and housing socio-economic indexes for areas Australia 2001. Belconnen, ACT: Australian Bureau of Statistics; 2001. [17] Trewin D. Australian Standard Geographical Classification (ASGC) — ABS Catalogue No. 1216.0. Canberra: Australian Bureau of Statistics; 2005. [18] Ottman R, Hauser WA, Stallone L. Semistructured interview for seizure classification: agreement with physicians' diagnoses. Epilepsia 1990;31:110–5. [19] Reutens DC, Howell RA, Gebert KE, Berkovic SF. Validation of a questionnaire for clinical seizure diagnosis. Epilepsia 1992;33:1065–71. [20] Foulds J, Wells JE, Lacey C, Adamson S, Sellman JD, Mulder R. A comparison of alcohol measures as predictors of psychological distress in the New Zealand population. IJADR 2013;2(1):59–67. [21] Babor TF, Higgins-Biddle JC, Saunders JB, Maristela G, Monteiro MG. The alcohol use disorders identification test guidelines for use in primary care. Dependence DoMHaS. WHO/MSD/MSB/01.6a; 2001. [22] Rutgers MJ. Epilepsy in general practice: the Dutch situation. Epilepsia 1986;27:734–8. [23] Ettinger A, Reed M, Cramer J, Epilepsy Impact Project G. Depression and comorbidity in community-based patients with epilepsy or asthma. Neurology 2004;63:1008–14. [24] Karouni M, Arulthas S, Larsson PG, Rytter E, Johannessen SI, Landmark CJ. Psychiatric comorbidity in patients with epilepsy: a population-based study. Eur J Clin Pharmacol 2010;66:1151–60. [25] Thapar A, Kerr M, Harold G. Stress, anxiety, depression, and epilepsy: investigating the relationship between psychological factors and seizures. Epilepsy Behav 2009;14:134–40. [26] Cramer JA, Blum D, Reed M, Fanning K. The influence of comorbid depression on seizure severity. Epilepsia 2003;44:1578–84. [27] Stefanello S, Marin-Leon L, Fernandes PT, Li LM, Botega NJ. Psychiatric comorbidity and suicidal behavior in epilepsy: a community-based case–control study. Epilepsia 2010;51:1120–5. [28] Adams SJ, O'Brien TJ, Lloyd J, Kilpatrick CJ, Salzberg MR, Velakoulis D. Neuropsychiatric morbidity in focal epilepsy. Br J Psychiatry 2008;192:464–9. [29] Hoppe C, Elger CE. Depression in epilepsy: a critical review from a clinical perspective. Nat Rev Neurol 2011;7:462–72. [30] Vink D, Aartsen MJ, Schoevers RA. Risk factors for anxiety and depression in the elderly: a review. J Affect Disord 2008;106:29–44. [31] Cook MJ, O'Brien TJ, Berkovic SF, Murphy M, Morokoff A, Fabinyi G, et al. Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study. Lancet Neurol 2013;12:563–71. [32] Goodwin R, Jacobi F, Bittner A, Wittchen H. Epidemiology of mood disorders. In: D.J.S., editor. The American Psychiatric Publishing Textbook of Mood Disorders. Arlington, Va: American Psychiatric Publishing; 2006. p. 33–54. [33] Clarke DM, Currie KC. Depression, anxiety and their relationship with chronic diseases: a review of the epidemiology, risk and treatment evidence. Med J Aust 2009;190: S54–60. [34] Gaitatzis A, Trimble MR, Sander JW. The psychiatric comorbidity of epilepsy. Acta Neurol Scand 2004;110:207–20. [35] Kessler RC, Lane MC, Shahly V, Stang PE. Accounting for comorbidity in assessing the burden of epilepsy among US adults: results from the National Comorbidity Survey Replication (NCS-R). Mol Psychiatry 2012;17:748–58. [36] Brummett BH, Krystal AD, Ashley-Koch A, Kuhn CM, Zuchner S, Siegler IC, et al. Sleep quality varies as a function of 5-HTTLPR genotype and stress. Psychosom Med 2007;69:621–4. [37] Ettinger AB. Psychotropic effects of antiepileptic drugs. Neurology 2006;67:1916–25. [38] Weintraub D, Buchsbaum R, Resor Jr SR, Hirsch LJ. Psychiatric and behavioral side effects of the newer antiepileptic drugs in adults with epilepsy. Epilepsy Behav 2007;10:105–10. [39] Petrovski S, Szoeke CE, Jones NC, Salzberg MR, Sheffield LJ, Huggins RM, et al. Neuropsychiatric symptomatology predicts seizure recurrence in newly treated patients. Neurology 2010;75:1015–21. [40] Kanner AM, Byrne R, Chicharro A, Wuu J, Frey M. A lifetime psychiatric history predicts a worse seizure outcome following temporal lobectomy. Neurology 2009;72:793–9.
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Risk factors for psychological distress in community-treated epilepsy Cameron J. Lacey a,b,⁎, Michael R. Salzberg a, Wendyl J. D'Souza b a b
Department of Psychiatry, St Vincent's Hospital Melbourne, University of Melbourne, 59 Victoria Parade, Fitzroy, 3065 Melbourne, Victoria, Australia Department of Medicine, St Vincent's Hospital Melbourne, University of Melbourne, 59 Victoria Parade, Fitzroy, 3065 Victoria, Australia
a r t i c l e
i n f o
Article history: Received 20 February 2014 Revised 28 March 2014 Accepted 31 March 2014 Available online xxxx Keywords: Epilepsy Psychiatric comorbidity Community Risk factors
a b s t r a c t The study aimed to determine risk factors for psychological distress in a community-treated sample of patients with epilepsy. This study investigated the Tasmanian Epilepsy Register participants. Participants included were as follows: aged 13 years and over, able to complete the individual computer-assisted participant interview, and diagnosed with epilepsy following an epilepsy specialist review of the diagnostic epilepsy interview, which was interpreted using standardized diagnostic guidelines. Psychological distress was assessed with the Kessler-10 questionnaire. Risk factors were grouped into four domains: sociodemographic factors, diseaserelated factors, psychological factors, and treatment-related factors. High or very high levels of psychological distress were reported by 22% of the participants, with 7.8% having very high distress. The regression model showed that psychological distress was significantly associated with female gender (F = 18.1, p b 0.001), diabetes mellitus (F = 8.7, p = 0.003), intellectual disability (F = 7.1, p = 0.06), and not receiving phenytoin (F = 5.1, p = 0.02). While the model was significant (F = 5.78, p b 0.001), only 11% of the variance of the K-10 score was explained by these factors (adjusted R-squared = 0.11). This study identifies female gender and comorbid medical conditions as risk factors for psychological distress and the use of phenytoin as a protective factor. The few factors identified and the limited variance explained suggest that a focus on epilepsy-related variables is unlikely to explain key influences underlying psychiatric comorbidity in patients with epilepsy. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The association between epilepsy and psychiatric comorbidity is well recognized with increased rates of a range of psychiatric disorders including depression, generalized anxiety disorder, and panic disorder [1]. Psychiatric comorbidity is associated with decreased quality-oflife, diminished medication adherence, poorer treatment outcomes, increased health service use, increased cognitive complaints, increased risk of other chronic diseases such as cardiovascular disease, and suicide [2–6]. The risk factors that contribute to psychological distress in patients with epilepsy remain unclear. Most research has been derived from hospital- or tertiary-based populations and is vulnerable to important sampling biases. For example, the rate of depression was found to be 58% in surgical patients [7] compared with 11% in community samples [8]. Furthermore, there are discrepancies in reported risk factors for psychiatric comorbidity between hospital-based samples and community studies. This may be due to the practical challenges of obtaining valid epilepsy-related variables in representative community-based studies with large sample sizes. ⁎ Corresponding author at: Department of Psychological Medicine, University of Otago, Christchurch, PO Box 4345, Christchurch, New Zealand. Tel.: +64 3 3720400; fax: +64 3 3720407. E-mail address: [email protected] (C.J. Lacey).
http://dx.doi.org/10.1016/j.yebeh.2014.03.025 1525-5050/© 2014 Elsevier Inc. All rights reserved.
The Tasmanian Epilepsy Register (TER) is one of a handful of studies utilizing a sample of people with community-treated epilepsy, and initial results confirmed that the rate of psychological distress is greater than that in the general population [4]. The TER is a sufficiently large, well-classified sample of people with epilepsy to investigate risk factors for psychological distress which may inform efforts at reducing this important health disparity [9]. The study aimed to determine risk factors for psychological distress in a community-treated sample of patients with epilepsy and to determine if rates of psychological distress differ across treatment settings.
2. Methods A description of the Tasmanian Epilepsy Register methodology has been previously published [10]. This study examined “psychological distress” in the TER population as revealed by the Kessler-10 (K-10) psychiatric screening tool. For convenience, the study used the term “psychological distress” to denote the symptoms assessed by the K-10 instrument [11]. The K-10 is an ideal measure to begin investigating psychiatric comorbidity as it captures most depressive and anxiety illness, has the advantages of brevity, and allows comparison with large community surveys that also employed this instrument [12]. While the concept of “psychological distress” includes both depression and anxiety disorders, research findings in the general population have
2
C.J. Lacey et al. / Epilepsy & Behavior 35 (2014) 1–5
established that there is some overlap in risk factors for both conditions [13]. 2.1. Inclusion criteria The inclusion criteria for this study were the following: TER participants of at least secondary school age (13 years and over), able to complete the individual computer-assisted participant interview, and diagnosed with epilepsy following an epilepsy specialist (WD) review of the diagnostic epilepsy interview, which was interpreted using standardized diagnostic guidelines [14]. Participants who were unable to be interviewed due to intellectual disability or communication difficulties were excluded. 2.2. Psychological distress Psychological distress was assessed with the K-10, a 10-question screening scale developed for the US National Health Interview Survey. The K-10 performs well in detecting DSM-IV anxiety and mood disorders (as validated by the Composite International Diagnostic Interview (CIDI)), with a high area under receiver operating characteristic curve (AUC) = 0.90. As in the Australian National Health Surveys [12], the K10 total was grouped into the following levels of psychological distress: low (K-10 = 10–15), moderate (K-10 = 16–21), high (K-10 = 22–30), and very high (K-10 = 31–50) or as a dichotomous variable in which participants were classified as having either low–moderate or high– very high levels of psychological distress. 2.3. Risk factors The classification proposed by Hermann et al. was modified to group potential risk factors divide into four domains [15]. 2.3.1. Sociodemographic factors Socioeconomic status was obtained from the participants' postcodes using the Index of Relative Socioeconomic Advantage/Disadvantage (SEIFA) 2001 developed by the Australian Bureau of Statistics [16]. This provides an estimate of each individual's socioeconomic status based on a measure of the relative social and economic well-being of the population of their postcode area taken as a whole. This, in turn, is derived from attributes such as the proportions in each area with low income, low educational attainment, high unemployment, and jobs in relatively unskilled occupations. High scores on the Index of Relative Socioeconomic Advantage/Disadvantage indicate higher socioeconomic status and less disadvantage. This study transformed SEIFA values into quintiles for the Tasmanian population as previously described in the Tasmanian Epilepsy Register methodology [10]. This study also examined the association with the three main geographical/administrative regions of Tasmania (Southern, Northern, and North-western) determined by participants' postcodes. Postcodes of participants were also used to determine their geographical remoteness using the Australian Standard Geographical Classification (ASGC) ‘Remoteness Structure’. This measures the remoteness based on the physical distance to the nearest urban center [17] and is classified into the following: major cities of Australia, inner regional Australia, outer regional Australia, remote Australia, very remote Australia, and migratory. 2.3.2. Disease-related factors Participants completed a detailed epilepsy diagnostic telephone questionnaire administered by trained interviewers. This provided detailed descriptive seizure data to enable an epilepsy specialist to determine the presence of epilepsy, seizure-onset type (generalized, focal, or uncertain), presence of an idiopathic generalized epilepsy (IGE) syndrome, age at onset of epilepsy, duration of epilepsy, seizure frequency, and antecedent epilepsy risk factors [14]. A diagnosis of epilepsy was made by blind interpretation on two occasions by an experienced epileptologist
applying standardized guidelines [14]. The diagnostic interview was a modified version of two diagnostic questionnaires, each previously shown to have substantial to very high agreement with physicianbased diagnoses in classifying seizure types and broad epilepsy-onset types [18,19]. This modified version showed almost perfect agreement in diagnosing epilepsy (κ = 0.94), seizure-onset types (κ = 0.84), simple or complex partial seizures (κ = 0.87), any generalized nonconvulsive seizure (κ = 0.82), IGE (κ = 0.82) and substantial agreement for secondarily generalized seizures (κ = 0.74), and generalized tonic–clonic seizures (κ = 0.79) [14]. The frequency of both convulsive and nonconvulsive seizures in the last 12 months was grouped into the following: none, less than monthly, or more than monthly. The epilepsy diagnostic interview also elicited information about antecedent epilepsy risk factors including the presence of other medical conditions that have been associated with seizures. These included each participant's history of febrile convulsion, serious head injury before first seizure, cerebrovascular accident (CVA), brain tumor, brain surgery, meningitis or encephalitis, cerebral palsy, intellectual disability, multiple sclerosis, diabetes mellitus, coma, polio, and arteriovenous malformation (AVM). The presence of these conditions was summed into the variable “number of antecedent epilepsy risk factors”. 2.3.3. Psychological factors Although alcohol abuse results in psychiatric disorder by both neurobiological and psychological mechanisms, it will be considered here for convenience [20]. Alcohol use was assessed with the Alcohol Use Disorders Identification Test (AUDIT) questionnaire, which has been widely used to screen for hazardous and harmful drinking [21]. The AUDIT scores of eight or greater were considered positive for hazardous and harmful drinking [21]. 2.3.4. Treatment-related factors The Health Insurance Commission provided data on the individual anticonvulsant medications dispensed in the 12-month study period for each participant. These were recorded as dichotomous variables as well as combined into the total number of anticonvulsant medications. The Health Insurance Commission records “prescribing doctor provider type” for all prescriptions, and this utilizes the vocation speciality recorded with each doctor's medical registration information. For the 12-month study period, these data were used to estimate the setting in which the Tasmanian Epilepsy Register participants received their medical care as studies suggest that the medical practitioner writing anticonvulsant drug prescriptions is most likely to also be responsible for disease supervision and follow-up [22]. Patients receiving care from a general practitioner only were compared with those receiving care from a general practitioner and/or a specialist. 2.4. Statistical analysis Univariate statistics were initially used to test for significant associations with psychological distress. The associations between K-10 total (continuous outcome) and predictor variables are presented using Spearman's rank correlation, independent t tests, and ANOVA tests for ordinal, dichotomous, and categorical variables, respectively. The associations between “high or very high” psychological distress (dichotomous outcome) and individual predictor variables utilized independent t tests, Mann–Whitney U test, and chi-squared test for continuous, ranked, and categorical variables, respectively. Boxplots show median, quartiles, range, and outliers (defined by 1.5 times the interquartile range outside the quartiles). General linear regression (SPSS© Version 19) was utilized to assess predictors of the level of psychological distress. Any predictor variable with a p value b 0.1 was selected for inclusion in the linear regression. This study was granted ethical approval by the University of Tasmania Human Research Ethics Committee.
C.J. Lacey et al. / Epilepsy & Behavior 35 (2014) 1–5
3. Results 3.1. Study population Of the 1180 patients enrolled on the Tasmanian Epilepsy Register, 43 withdrew prior to interview, 36 died, and 254 were excluded (90 were b13 years old, 164 had intellectual disability or communication difficulties), leaving 847 potential subjects. Six hundred sixty patients completed the K-10 (response rate = 78%), and 554 of these were diagnosed with epilepsy following specialist review of the epilepsy diagnostic interview. The mean age of participants was 51 years (range = 13–102 years), with 48% of male gender, similar to the Tasmanian Epilepsy Register cohort. There were no significant differences between participants and those excluded for mean age, gender ratio, and socioeconomic status.
3
of psychological distress between participants with IGE compared with those without IGE. There was no significant correlation between age at onset or duration of epilepsy and level of psychological distress. Convulsive seizures occurring more than monthly in the last 12 months were associated with a higher level of psychological distress (p = 0.001, see Fig. 1). However, there was no association between nonconvulsive seizure frequency in the last 12 months and level of psychological distress. The majority of participants (56%) had at least one antecedent epilepsy risk factor (see Supplementary Table). Only the presence of CVA (p = 0.006), intellectual disability (p = 0.002), and diabetes mellitus (p = 0.008) were associated with increased psychological distress. Antecedent epilepsy risk factors were summed to give the “number of antecedent epilepsy risk factors”. There was a significant association between higher number of risk factors and increased level of psychological distress (p = 0.006).
3.2. Psychological distress 3.5. Alcohol abuse High or very high levels of psychological distress were reported by 22% of the participants, with 7.8% having very high distress.
Hazardous or harmful alcohol use was reported in only 4% of the participants and was not associated with level of psychological distress.
3.3. Sociodemographic risk factors 3.6. Treatment-related risk factors Females had a higher rate of high–very high psychological distress compared with males (28.3% versus 18.5%, p = 0.003). There was a small but significant correlation between lower levels of psychological distress and greater age (r = 0.29, p = 0.001). There was no significant correlation between psychological distress and socioeconomic status or geographical region. Almost all participants had a “remoteness” classification of either “inner” or “outer regional Australia”. Only two participants lived in “remote Australia”, and these were not included in the analysis because of the small number. There was no association between living in “inner regional Australia” and increased psychological distress compared with living in “outer regional Australia”. 3.4. Disease-related risk factors There were no differences in the mean level of psychological distress between participants who were diagnosed with epilepsy following specialist review of the diagnostic epilepsy interview data and those with uncertain or other diagnoses. The associations between disease-related risk factors and psychological distress are presented in Table 1. There was no significant association between seizure-onset type and level of psychological distress. There was no significant difference in the level
Only the use of phenytoin was associated with lower levels of psychological distress (p = 0.001). Participants who received phenytoin were more likely to be male (p = 0.07), report less frequent convulsive seizures (p b 0.001), and receive only one anticonvulsant medication (p = 0.004). No single anticonvulsant was associated with increased psychological distress. The mean number of anticonvulsant medications was 1.4 (range = 0–5). Higher number of anticonvulsant medications was associated with increasing psychological distress (p = 0.03). The majority of participants were treated by a general practitioner only (77%). However, participants receiving treatment from specialists alone or in combination with a general practitioner (23%) had an increased rate of high–very high psychological distress compared with those seeing only a general practitioner (p = 0.03). 3.7. Multivariate associations The variables with univariate association with psychological distress (p b 0.1) were entered into linear regression modeling with age and gender entered as covariates. These variables were geographical
Table 1 Epilepsy-related risk factors and psychological distress. Characteristic
Low–moderate psychological distress
High–very high psychological distress
p
Seizure-onset type Generalized Focal Uncertain
101 (79%) 282 (76%) 46 (82%)
27 (21%) 88 (24%) 10 (18%)
NSa
Epilepsy syndrome IGE Other syndromes
98 (78%) 331 (77%)
27 (22%) 98 (23%)
NSa
Convulsive seizure frequency (last 12 months) None 136 (72%) Less than monthly 284 (82%) More than monthly 9 (50%)
52 (28%) 64 (18%) 9 (50%)
0.01a
Nonconvulsive seizure frequency (last 12 months) None 244 (76%) Less than monthly 147 (81%) More than monthly 38 (73%)
76 (24%) 35 (19%) 14 (27%)
NSa
Numbers are n (%). IGE = idiopathic generalized epilepsy. a χ2.
Fig. 1. Boxplot of K-10 total by convulsive seizure frequency.
4
C.J. Lacey et al. / Epilepsy & Behavior 35 (2014) 1–5
remoteness, convulsive seizure frequency, CVA, intellectual disability, diabetes mellitus, number of antecedent epilepsy risk factors, use of phenytoin, and number of anticonvulsant medications. The regression model showed that psychological distress was significantly associated with female gender (F = 18.1, p b 0.001), diabetes mellitus (F = 8.7, p = 0.003), intellectual disability (F = 7.1, p = 0.06), and not receiving phenytoin (F = 5.1, p = 0.02). The strongest predictor of psychological distress was female gender; however, this explained only 3% of the variance of the K-10 score. While the model was significant (F = 5.78, p b 0.001), only 11% of the variance of the K-10 score was explained by these factors (adjusted R-squared = 0.11).
4. Discussion This study identifies female gender and a history of other medical conditions (CVA, diabetes, and intellectual disability) as risk factors for psychological distress and the use of phenytoin as a protective factor. However, these factors predicted only 11% of the variance in psychological distress and suggest that further research utilizing a broader range of potential risk factors is required to identify potential contributors to psychological distress. The few factors identified and the limited variance explained suggest that focusing solely on epilepsy-related variables is unlikely to explain the key influences underlying psychiatric comorbidity in patients with epilepsy. Psychological factors have been more consistently associated with psychiatric comorbidity in previous reviews [15], and a more detailed study, incorporating recent findings of risk factors for depression in the general population, is underway. Although not directly comparable, the rate of psychological distress found in this study is consistent with that found in other community studies. One of the more rigorous community studies reported a higher rate of depression in a USA community sample of persons with epilepsy (36%) than in healthy controls (12%), although similar to a control group of patients with asthma (28%) [23]. This is also consistent with a population-based study which estimated a 32% rate of psychiatric comorbidity in a nationwide sample of patients with epilepsy using the Norwegian Prescription Database [24]. Previous studies reporting a higher rate of psychiatric comorbidity in tertiary care compared with primary care relied on indirect comparison of different sample groups and differing study designs. In fact, this comparison has been estimated across at least three different treatment settings: patients being assessed with video-EEG or for epilepsy surgery, patients attending epilepsy outpatient clinics, other groups of patients with epilepsy including support groups, and patients attending general practitioners. The novel recruitment strategy (national prescription database), validated epilepsy diagnostic classification methods, and use of a standardized measure of psychiatric comorbidity across these differing treatment settings allow the first direct measure of this phenomenon and support the gradient of psychological distress from primary to specialized care. This gradient of distress across treatment settings highlights the importance of sample selection and potential for ascertainment bias. Seizure type and epilepsy syndrome are often poorly addressed in community studies, with few studies attempting this and most relying on patient self-report [25,26] or clinical record review [27]. The benefit of utilizing the Tasmanian Epilepsy Register is the availability of validated epilepsy syndrome and seizure-onset data that reflect the interpretation by an epilepsy specialist of seizure symptoms and signs directly from the patients and a witness. Ideally, the use of a structured diagnostic interview combined with results of additional investigations (including EEG and MRI) would be combined to provide greater diagnostic precision. However, this study's finding of no association between psychological distress and seizure-onset type or presence of IGE syndrome suggests that any contribution of these factors to the risk of depression is likely to be small if it exists at all. There is a much larger literature derived from more highly selected samples with more accurate seizure type and epilepsy syndrome characterization, with many advocating
that psychiatric comorbidity is more common in focal epilepsy than in other types and syndromes [28], although this is also debated [29]. There are several limitations to this study. Primarily, it is crosssectional; thus, it is not possible to draw inferences about causation or the nature of the association between psychological distress and potential risk factors. As previously noted, the K-10 identifies symptoms of anxiety or depression but does not generate a psychiatric diagnosis. While there is some overlap in risk factors for both conditions in the general population and in people with chronic disease [13,30], this has not been well established in people with epilepsy. This study also relies on self-report for the assessment of seizure frequency. Accurately determining the frequency of seizures is not simple, particularly for recall of seizures in which people lose awareness or have cognitive dysfunction features or seizures that occur exclusively during sleep [31]. Additionally, for those experiencing multiple seizure types, it is difficult to determine the individual frequency of each. Finally, the presence of depression has been shown to bias perception of seizure frequency after adjusting for seizure type and recency [26]. There are also potential limitations with the method of identifying medications used by participants. It was not possible to determine dosage, adherence, and blood levels, which may confer additional risk for development of adverse effects including psychological distress. Additionally, as medications for the entire year were recorded in the Health Insurance Commission database, this may capture sequential trials of individual medications rather than reflecting true polydrug therapy. Furthermore, only small numbers of participants received some anticonvulsants that have been associated with depression, such as primidone, phenobarbitone, and topiramate, or predate their introduction in Australia (levetiracetam), and, subsequently, there is limited power to detect associations between individual medications and psychological distress. The few risk factors for psychological distress identified in this study are largely consistent with previous studies of psychiatric comorbidity in people with epilepsy although there are notable differences. The association between female gender and psychological distress is consistent with risk factors for depression identified in the general population [32], people with chronic disease [33], and other community samples with epilepsy [23,34]. Living in an urban region has also been established as a risk factor for depression in the general population [32]; however, this study did not find evidence of an association between psychological distress and geographical remoteness. This may reflect the limitation of the measure of remoteness for Tasmania [17]. However, there may also be other factors contributing to the decision where people with epilepsy live, such as the proximity to health care. At least part of the failure to demonstrate further risk factors may be attributable to the lack of key sociodemographic measures in the TER, such as employment and relationship status, which have been consistently linked with depression in the general population [32]. The association of additional comorbidities (diabetes, intellectual disability, and stroke) with psychological distress reinforces an emerging awareness of the need to consider the range of comorbidities seen in people with epilepsy [5]. This study examined only conditions that are risk factors for development of epilepsy, but 25% of the participants reported having two or more such conditions, and over half (56%) reported at least one. Even higher rates of comorbidity were reported in a household survey in the US, the National Comorbidity Survey — Replication (NCS-R), with 41% of people with epilepsy reported having four or more physical comorbidities [35]. This may be explained by the greater number of physical comorbidities screened for in the NCS-R study. The association of phenytoin with lower psychological distress is an intriguing finding, and whether this is a true drug effect or whether psychological distress leads to help-seeking, and increased likelihood of medication switch to ‘newer’ anticonvulsant medication warrants further investigation. It is also possible that the phenytoin's apparent protective factor is attributable to the associations between phenytoin use and male gender, lower convulsive seizure frequency, and anticonvulsant monotherapy. There is some literature suggesting an
C.J. Lacey et al. / Epilepsy & Behavior 35 (2014) 1–5
antidepressant effect for phenytoin, which may explain its apparent protective effect against psychological distress found in this analysis [36]. There is also some evidence from other studies that other anticonvulsant medications, such as lamotrigine, may protect against depression and anxiety [37,38]. However, the limited numbers of study participants receiving lamotrigine (n = 97) and the study's limited ability to capture the use of newer anticonvulsant medication with proposed therapeutic psychotropic effects limit the study's power to detect such effects. Understanding the factors associated with psychological distress may assist developing strategies to optimize pharmacotherapy given the association between depression comorbidity and treatment resistance [3,39,40]. 5. Conclusions Despite finding associations between gender, intellectual disability, diabetes mellitus, and psychological distress and a protective effect of phenytoin, little of the variance in psychological distress is linked with these factors. In future studies of psychiatric comorbidity in people with epilepsy, there remains a need for adoption of a broader range of risk factors and, in particular, detailed psychological variables [15]. Ethical publication statement We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.yebeh.2014.03.025. Acknowledgments Cameron Lacey has received support from a Lundbeck Neuropsychiatry Research Fellowship. Wendyl D'Souza has received support from a Menzies Research Institute NHMRC Capacity Building Grant and an NHMRC Health Professional Fellowship. The Tasmanian Epilepsy Register was supported by The Booth Estate Launceston, The Royal Hobart Hospital Research Foundation, GSK Neurology, and the Clifford Craig Medical Research Trust, North West Tasmania. We also thank our research assistant Pam McDonald for her considerable efforts in participant liaison, data processing, and management. Conflict of interest statement Associate Professor D'Souza has received travel, investigatorinitiated, and speaker honoraria from UCB Pharma; educational grants from Novartis Pharmaceuticals, Pfizer Pharmaceuticals, and SanofiSynthelabo; and educational, travel, and fellowship grants from GSK Neurology Australia and honoraria from SciGen Pharmaceuticals. The remaining authors have no conflicts of interest. References [1] Ettinger AB, Kanner AM. Psychiatric issues in epilepsy: a practical guide to diagnosis and treatment. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2006. [2] Taylor RS, Sander JW, Taylor RJ, Baker GA. Predictors of health-related quality-of-life and costs in adults with epilepsy: a systematic review. Epilepsia 2011;52:2168–80. [3] Hitiris N, Mohanraj R, Norrie J, Sills GJ, Brodie MJ. Predictors of pharmacoresistant epilepsy. Epilepsy Res 2007;75:192–6. [4] Lacey CJ, Salzberg MR, Roberts H, Trauer T, D'Souza WJ. Psychiatric comorbidity and impact on health service utilization in a community sample of patients with epilepsy. Epilepsia 2009;50:1991–4. [5] Ottman R, Lipton RB, Ettinger AB, Cramer JA, Reed ML, Morrison A, et al. Comorbidities of epilepsy: results from the Epilepsy Comorbidities and Health (EPIC) survey. Epilepsia 2011;52:308–15. [6] Christensen J, Vestergaard M, Mortensen PB, Sidenius P, Agerbo E. Epilepsy and risk of suicide: a population-based case–control study. Lancet Neurol 2007;6:693–8. [7] Victoroff JMD. DSM-III-R psychiatric diagnoses in candidates for epilepsy surgery: lifetime prevalence. Neuropsychiatry Neuropsychol Behav Neurol 1994;7:87–97.
5
[8] Mensah SA, Beavis JM, Thapar AK, Kerr M. The presence and clinical implications of depression in a community population of adults with epilepsy. Epilepsy Behav 2006;8:213–9. [9] Barry JJ, Ettinger AB, Friel P, Gilliam FG, Harden CL, Hermann B, et al. Consensus statement: the evaluation and treatment of people with epilepsy and affective disorders. Epilepsy Behav 2008;13(Suppl. 1):S1–S29. [10] D'Souza WJ. Epidemiology of epilepsy in Tasmania: a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Epidemiology at Massey University at Massey University, Wellington, New Zealand; 2008. [11] Kessler RC, Andrews G, Colpe LJ, Hiripi E, Mroczek DK, Normand SL, et al. Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychol Med 2002;32:959–76. [12] Australian Bureau of Statistics. Information Paper: Use of the Kessler Psychological Distress Scale in ABS Health Surveys. Canberra: 2012 4817.0.55.001. [13] Kendler KS, Gardner CO, Prescott CA. Toward a comprehensive developmental model for major depression in women. Am J Psychiatry 2002;159:1133–45. [14] D'Souza WJ, Stankovich J, O'Brien TJ, Bower S, Pearce N, Cook MJ. The use of computer-assisted-telephone-interviewing to diagnose seizures, epilepsy and idiopathic generalized epilepsy. Epilepsy Res 2010;91:20–7. [15] Hermann BP, Seidenberg M, Bell B. Psychiatric comorbidity in chronic epilepsy: identification, consequences, and treatment of major depression. Epilepsia 2000;41(Suppl. 2):31–41. [16] Statistics ABo. Information paper census of population and housing socio-economic indexes for areas Australia 2001. Belconnen, ACT: Australian Bureau of Statistics; 2001. [17] Trewin D. Australian Standard Geographical Classification (ASGC) — ABS Catalogue No. 1216.0. Canberra: Australian Bureau of Statistics; 2005. [18] Ottman R, Hauser WA, Stallone L. Semistructured interview for seizure classification: agreement with physicians' diagnoses. Epilepsia 1990;31:110–5. [19] Reutens DC, Howell RA, Gebert KE, Berkovic SF. Validation of a questionnaire for clinical seizure diagnosis. Epilepsia 1992;33:1065–71. [20] Foulds J, Wells JE, Lacey C, Adamson S, Sellman JD, Mulder R. A comparison of alcohol measures as predictors of psychological distress in the New Zealand population. IJADR 2013;2(1):59–67. [21] Babor TF, Higgins-Biddle JC, Saunders JB, Maristela G, Monteiro MG. The alcohol use disorders identification test guidelines for use in primary care. Dependence DoMHaS. WHO/MSD/MSB/01.6a; 2001. [22] Rutgers MJ. Epilepsy in general practice: the Dutch situation. Epilepsia 1986;27:734–8. [23] Ettinger A, Reed M, Cramer J, Epilepsy Impact Project G. Depression and comorbidity in community-based patients with epilepsy or asthma. Neurology 2004;63:1008–14. [24] Karouni M, Arulthas S, Larsson PG, Rytter E, Johannessen SI, Landmark CJ. Psychiatric comorbidity in patients with epilepsy: a population-based study. Eur J Clin Pharmacol 2010;66:1151–60. [25] Thapar A, Kerr M, Harold G. Stress, anxiety, depression, and epilepsy: investigating the relationship between psychological factors and seizures. Epilepsy Behav 2009;14:134–40. [26] Cramer JA, Blum D, Reed M, Fanning K. The influence of comorbid depression on seizure severity. Epilepsia 2003;44:1578–84. [27] Stefanello S, Marin-Leon L, Fernandes PT, Li LM, Botega NJ. Psychiatric comorbidity and suicidal behavior in epilepsy: a community-based case–control study. Epilepsia 2010;51:1120–5. [28] Adams SJ, O'Brien TJ, Lloyd J, Kilpatrick CJ, Salzberg MR, Velakoulis D. Neuropsychiatric morbidity in focal epilepsy. Br J Psychiatry 2008;192:464–9. [29] Hoppe C, Elger CE. Depression in epilepsy: a critical review from a clinical perspective. Nat Rev Neurol 2011;7:462–72. [30] Vink D, Aartsen MJ, Schoevers RA. Risk factors for anxiety and depression in the elderly: a review. J Affect Disord 2008;106:29–44. [31] Cook MJ, O'Brien TJ, Berkovic SF, Murphy M, Morokoff A, Fabinyi G, et al. Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study. Lancet Neurol 2013;12:563–71. [32] Goodwin R, Jacobi F, Bittner A, Wittchen H. Epidemiology of mood disorders. In: D.J.S., editor. The American Psychiatric Publishing Textbook of Mood Disorders. Arlington, Va: American Psychiatric Publishing; 2006. p. 33–54. [33] Clarke DM, Currie KC. Depression, anxiety and their relationship with chronic diseases: a review of the epidemiology, risk and treatment evidence. Med J Aust 2009;190: S54–60. [34] Gaitatzis A, Trimble MR, Sander JW. The psychiatric comorbidity of epilepsy. Acta Neurol Scand 2004;110:207–20. [35] Kessler RC, Lane MC, Shahly V, Stang PE. Accounting for comorbidity in assessing the burden of epilepsy among US adults: results from the National Comorbidity Survey Replication (NCS-R). Mol Psychiatry 2012;17:748–58. [36] Brummett BH, Krystal AD, Ashley-Koch A, Kuhn CM, Zuchner S, Siegler IC, et al. Sleep quality varies as a function of 5-HTTLPR genotype and stress. Psychosom Med 2007;69:621–4. [37] Ettinger AB. Psychotropic effects of antiepileptic drugs. Neurology 2006;67:1916–25. [38] Weintraub D, Buchsbaum R, Resor Jr SR, Hirsch LJ. Psychiatric and behavioral side effects of the newer antiepileptic drugs in adults with epilepsy. Epilepsy Behav 2007;10:105–10. [39] Petrovski S, Szoeke CE, Jones NC, Salzberg MR, Sheffield LJ, Huggins RM, et al. Neuropsychiatric symptomatology predicts seizure recurrence in newly treated patients. Neurology 2010;75:1015–21. [40] Kanner AM, Byrne R, Chicharro A, Wuu J, Frey M. A lifetime psychiatric history predicts a worse seizure outcome following temporal lobectomy. Neurology 2009;72:793–9.
