Epilepsy Research (2014) 108, 305—315
journal homepage: www.elsevier.com/locate/epilepsyres
Epilepsy beyond seizure: A population-based study of comorbidities Anbesaw W. Selassie, Dulaney A. Wilson, Gabriel U. Martz, Georgette G. Smith, Janelle L. Wagner, Braxton B. Wannamaker ∗ Medical University of South Carolina, United States Received 17 June 2013; received in revised form 25 October 2013; accepted 1 December 2013 Available online 18 December 2013
KEYWORDS Epilepsy; Comorbidity; Epidemiology; Migraine; Mortality; Risk set
∗
Summary Comorbid conditions may affect the quality of life in persons with epilepsy (PWE) more than seizures. Using legally mandated healthcare encounter data, somatic, psychiatric, and neurodevelopmental comorbidities in a large population-based cohort of PWE, were compared to persons with migraine (PWM), a similar neurologic condition, and lower extremity fracture (PWLF), otherwise healthy controls. 64,188 PWE, 121,990 PWM, and 89,808 PWLF were identified from inpatient, outpatient, and emergency department from 2000 to 2011. Epilepsy was ascertained with ICD-9-CM code 345; migraine with 346; fracture of the tibia, fibula, and ankle with 823 and 824. Common comorbidities of epilepsy were identified from the literature. Differences in prevalence among PWE, PWM, and PWLF were assessed by comparison of 95% confidence intervals (CI) constructed under the assumption of independence and normal approximation. The association of the comorbid conditions with epilepsy and migraine, compared to lower extremity fracture, were evaluated with polytomous logistic regression controlling for demographic and mortality covariables. PWE had significantly elevated prevalence of comorbidities compared with PWM and PWLF. Compared with PWLF, the adjusted odds ratios (OR) of having both somatic and psychiatric/neurodevelopmental comorbidities were 5.44 (95% CI = 5.25—5.63) and 2.49 (95% CI = 2.42—2.55) in PWE and PWM, respectively. The association with epilepsy was the strongest for cognitive dysfunction (OR = 28.1; 95% CI = 23.3—33.8); autism spectrum disorders (OR = 22.2; 95% CI = 16.8—29.3); intellectual disability (OR = 12.9; 95% CI = 11.6—14.3); and stroke (OR = 4.2; 95% CI = 4.1—4.4). The absolute risk increase in PWE compared with PWM for any somatic or psychiatric/neurodevelopmental comorbidity was 58.8% and 94.3%, respectively. Identifying comorbidities that are strongly and consistently associated with seizures, particularly disorders with shared underlying pathophysiology, is critical in identifying specific research and practice goals that may ultimately improve the quality of life for PWE. This study contributes to that effort by providing population-based comorbidity data for PWE compared with PWM and PWLF. © 2013 Elsevier B.V. All rights reserved.
Corresponding author at: 135 Cannon Street, Suite 303, MSC 835, Charleston, SC 29425, United States. Tel.: +1 843 876 1140. E-mail address: [email protected] (B.B. Wannamaker).
0920-1211/$ — see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.eplepsyres.2013.12.002
306
A.W. Selassie et al.
Introduction
Study setting and population
Various somatic, psychiatric, and neurodevelopmental conditions associated with epilepsy, other than seizures, have been recognized for many years and can impact quality of life more than seizures (Baca et al., 2011; Kobau et al., 2008; Linehan et al., 2011) and reduce rehabilitation efficacy (Chiappedi et al., 2011). These conditions are said to be comorbidities of epilepsy when they occur more frequently in persons with epilepsy (PWE) than in the general population (Thurman et al., 2011). In most instances, it is not known whether the comorbid condition was caused by epilepsy or its treatments, led to epilepsy, or has a common underlying cause with epilepsy (Seidenberg et al., 2009). Epidemiological studies can facilitate a better understanding of the burden that these comorbidities impose on PWE. In the clinic, PWE with multiple comorbidities are common and the treatment of comorbid conditions is often more difficult than treating the seizures. This study undertakes an analysis of statewide healthcare encounter data to address comorbidities associated with epilepsy in order to inform clinical practice and public health policy. The database is very large and includes all age groups, diverse socioeconomic and demographic groups thus permitting robust cohort sizes for subset analyses and comparison with other medical conditions. Following the recommendations of the ILAE (Thurman et al., 2011), a control group of persons with lower extremity fracture (PWLF) without any clinical evidence of pathological fractures were selected and should represent an otherwise healthier patient population with only intermittent healthcare needs and fewer counts of comorbid conditions relatively comparable to the general population. A neurological comparative group, people with migraine (PWM), was also selected to address the research gaps identified in comorbidities of epilepsy (National Institutes of Health, 2012). PWM has many similar characteristics to PWE regarding chronicity, earlylife onset, intermittent and variable symptoms, medical therapy including and rescue and prophylactic medications, genetic or unknown etiologies, and an increased risk of psychological factors with bidirectional association with migraine, compared to healthy controls (Balottin et al., 2011; Kasteleijn-Nolst Trenite and Parisi, 2012; Swartz et al., 2000). This study hypothesizes PWE would have higher odds of chronic comorbid conditions than PWM as compared with PWLF independent of demographic and clinical characteristics.
SC residents hospitalized or treated and released from the ED and OPD from January 1, 2000 to December 31, 2011 were eligible for the study. Cases and control groups were ascertained using International Classification of Diseases, 9th revision, Clinical Modification (ICD-9-CM) diagnosis codes for epilepsy (345.x), migraine (346.x), and fracture of the tibia, fibula, or ankle (823.x and 824.x) assigned by clinical providers at discharge (American Medical Association, 2009). PWLF without any history of seizure represent healthcare controls that approximate the pattern of comorbidity in the general population. Such fractures occur as isolated trauma without damage to adjacent vital organs and are less likely to be secondary to underlying pathologies of the bone. Persons with epilepsy and migraine were classified as PWE with comorbid migraine; therefore, no PWM had epilepsy.
Methods Data sources This study utilized the South Carolina (SC) statewide hospital discharge and emergency department (ED) visit datasets, which includes hospital-based outpatient department (OPD) visits reported via the uniform billing system (UB-04). In SC, health care providers are legally required to submit data to the Office of Research and Statistics (ORS), the SC Budget and Control Board (Weis et al., 2006). The Medical University of South Carolina Institutional Review Board exempted this study.
Definitions PWE were defined based on at least one ICD-9-CM code for epilepsy (345.x), excluding petit mal status (345.2) and grand mal status (345.3), during the study period. Prevalent comorbidities demonstrated in the epilepsy literature (Rai et al., 2012; Thurman et al., 2011) were chosen for the current novel comparative analysis between epilepsy, migraine, and lower extremity fracture (LEF). The ICD-9-CM codes used to define these comorbidities are listed in Appendix A. Race-ethnicity was categorized into non-Hispanic white, non-Hispanic black, Hispanic, and all others. Mortality status from any cause was determined for the total cohort through December 31, 2011 based on the SC Multiple Causes of Death Data file. All diagnosis fields were searched for the specified comorbid conditions. The conditions were further grouped as somatic or psychiatric/neurodevelopmental disorders. Using polytomous logistic regression the association of epilepsy with the comorbid conditions was examined in comparison with migraine and LEF adjusting for demographic characteristics and mortality status. The cohort size, frequency of clinical encounters and the number of comorbid conditions per capita are shown in Table 1. The remaining variables were classified as shown in Table 2.
Statistical analysis Data were analyzed with SAS software package, V9.3.1 (SAS Institute, 2011). Demographic characteristics and the comorbid conditions were treated as independent variables. In accordance with product-multinomial sampling design, marginal frequencies corresponding to cases (PWE) and comparison groups (PWM and PWLF) were fixed a priori (Fienberg, 1989). Proportions corresponding to demographic characteristics and the specific comorbid conditions were compared among PWE, PWM, and PWLF by constructing 95% confidence intervals (CI) under the assumption of independence and normal approximation (Table 1). Overlapping confidence intervals suggest no significant difference in proportions. Taking epilepsy, migraine, and LEF as nominal response categories, association with comorbid conditions and demographics were examined in polytomous logistic regression.
Epilepsy beyond seizure Table 1
307
Proportionate distribution of comorbid conditions.
Comorbidity type
Epilepsy (n = 64,188) % (95% CI)
Migraine (n = 121,990) % (95% CI)
Lower ext fracture (n = 89,808) % (95% CI)
5-year median number of visits (IQR)* Median count of comorbidity (IQR)* % with polymorbidity (≥6 conditions) Somatic disorders Cardiovascular disease Intestinal problems Asthma/pulmonary disease Gastric reflux Anemia Stroke Diabetes Peptic ulcer Traumatic brain injury Nutritional deficiency GI bleed Osteoporosis Vision loss Hearing loss Parkinson’s disease HIV/AIDS Multiple sclerosis Migraine Psychiatric/neurodevelopmental disorders Depression Anxiety Psychoses Alcoholism Drug abuse Suicidal ideation/attempt Intellectual disability Schizophrenia Alzheimer’s dementia Personality disorder Cognitive dysfunction ADHD Autism spectrum disorder
15 (8—27) 4 (2—7) 38.5 (38.1—38.9)
13 (6—24) 3 (1—5) 20.7 (20.4—20.9)a
7 (4—14) 2 (0—4) 14.3 (14.0—14.5)a,b
61.7 (61.3—62.1) 38.2 (37.9—38.6) 36.4 (36.0—36.7) 34.5 (34.2—34.9) 28.9 (28.5—29.2) 27.1 (26.8—27.5) 24.8 (24.4—25.1) 23.8 (23.5—24.2) 14.5 (14.2—14.7) 13.4 (13.1—13.6) 10.4 (10.2—10.7) 6.0 (5.9—6.2) 3.4 (3.2—3.5) 2.9 (2.8—3.1) 1.8 (1.7—1.9) 1.3 (1.2—1.4) 1.1 (1.1—1.2) 12.1 (11.9—12.4)
46.2 (45.9—46.5)a 34.5 (34.2—34.7)a 30.5 (30.2—30.7)a 33.6 (33.3—33.8)a 16.7 (16.4—16.9)a 8.0 (7.9—8.2)a 13.7 (13.5—13.9)a 23.7 (23.4—23.9) 7.1 (6.9—7.2)a 2.9 (2.8—3.0)a 6.3 (6.2—6.4)a 3.1 (3.0—3.2)a 1.1 (1.1—1.2)a 1.6 (1.5—1.6)a 0.3 (0.3—0.3)a 0.5 (0.4—0.5)a 1.1 (1.1—1.2) 100.0 (100.0—100.0)a
41.4 (41.1—41.8)a,b 24.7 (24.4—25.0)a,b 21.5 (21.2—21.7)a,b 21.0 (20.8—21.3)a,b 13.6 (13.4—13.8)a,b 7.3 (7.1—7.5)a,b 15.8 (15.5—16.0)a,b 13.5 (13.3—13.7)a,b 9.2 (9.0—9.4)a,b 4.5 (4.4—4.6)a,b 4.7 (4.5—4.8)a,b 5.2 (5.1—5.4)a,b 1.0 (1.0—1.1)a 1.4 (1.3—1.5)a 0.6 (0.5—0.6)a,b 0.4 (0.4—0.5)a 0.4 (0.4—0.4)a,b 4.5 (4.4—4.6)a,b
31.3 (30.9—31.6) 29.1 (28.7—29.4) 21.5 (21.2—21.8) 18.1 (17.8—18.4) 16.9 (16.6—17.2) 8.8 (8.6—9.0) 7.3 (7.1—7.5) 5.6 (5.4—5.7) 5.0 (4.8—5.2) 4.3 (4.2—4.5) 3.6 (3.5—3.7) 3.4 (3.3—3.6) 1.3 (1.2—1.4)
28.5 (28.3—28.8)a 29.0 (28.8—29.3) 11.3 (11.2—11.5)a 5.7 (5.6—5.9)a 9.8 (9.6—9.9)a 5.1 (5.0—5.2)a 0.2 (0.2—0.2)a 1.2 (1.1—1.3)a 0.5 (0.4—0.5)a 2.4 (2.3—2.5)a 0.1 (0.1—0.1)a 2.1 (2.1—2.2)a 0.0 (0.0—0.0)a
15.3 (15.1—15.5)a,b 13.3 (13.1—13.5)a,b 6.4 (6.2—6.5)a,b 10.1 (9.9—10.3)a,b 7.1 (6.9—7.2)a,b 2.7 (2.6—2.8)a,b 0.4 (0.4—0.5)a,b 1.3 (1.2—1.4)a 1.8 (1.8—1.9)a,b 0.9 (0.8—0.9)a,b 0.1 (0.1—0.2)a 1.8 (1.7—1.9)a,b 0.1 (0.0—0.1)a,b
* a b
Inter-quartile range, P < 0.05 Kruskal—Wallis H-test. Significantly different from epilepsy group. Significantly different from migraine group.
This modeling technique is an extension of the standard logistic regression that allows direct comparison of the parameter estimates across the response categories because of the model’s ability to generate pooled covariance matrix and smaller standard errors (Biesheuvel et al., 2008; Stokes et al., 1997). Parameter estimates derived from polytomous logistic regression are similar to those derived from two separate dichotomous logistic models (epilepsy vs. LEF and migraine vs. LEF); however, parameter estimates from separate logistic models cannot be directly compared. Multicollinearity among covariables was absent based on the deviations of the regression coefficients and their standard errors in the fitted univariate and multivariable models (Darlington et al., 1998). Variables were entered
simultaneously in the model. The adjusted odds ratios and 95% CIs are reported.
Results From 2000 through 2011, 64,188 PWE, 121,990 PWM, and 89,808 PWLF met inclusion criteria for the study (Table 1). Median per capita encounters/year was 1.2 and 2.1 times more frequent in PWE than PWM and PWLF, respectively. PWE had significantly higher proportions of comorbid conditions than PWM and PWLF. Polymorbidity (having ≥6 comorbid conditions) was 1.86 and 2.69 times more frequent in PWE than PWM and PWLF, respectively.
308
Table 2
Associations of somatic and psychiatric/neurodevelopmental comorbidities in persons with epilepsy, migraine and lower extremity fracture.
Characteristics
Epilepsy (n = 64,188) Frequency (%)
Somatic or psychiatric/neurodevelopmental comorbidity Both 36,271 (56.5) 5.44 Psych/neurodev. only 3,563 (5.6) 3.70 Somatic only 18,052 (28.1) 2.11 None 6,302 (9.8) 1.00 Mortality status Deceased 11,989 (18.7) 1.80 Alive 1.00 52,199 (81.3) Age group Mean age (s.d.)* 41.6 ± 22.5 0—5 5,641 (8.5) 2.38 6—12 3,252 (5.1) 1.17 13—18 4,386 (6.8) 1.45 19—34 12,867(20.1) 1.70 35—64 28,718 (44.7) 1.32 65 and older 9,504 (14.8) 1.00 Race/ethnicity Black 23,134 (36.1) 1.47 Hispanic 574 (0.9) 1.27 Other 856 (1.3) 1.18 White 39,624 (61.7) 1.00 Gender Male 31,256 (48.7) 1.13 Female 32,932 (51.3) 1.00 Payer Uninsured 6,516 (10.2) 0.76 Medicare 21,024 (32.8) 1.79 Medicaid 18,551 (28.9) 1.76 Commercial 18,097 (28.2) 1.00 *
Adjusted for all covariables in the model. P < 0.05.
Lower extremity fracture (n = 89,808) Frequency (%) Reference
(5.25—5.63) (3.51—3.90) (2.04—2.18)
49,160 (40.3) 5,758 (4.7) 45,326 (37.2) 21,746 (17.8)
2.49 (2.42—2.55) 1.51 (1.44—1.58) 1.89 (1.84—1.94) 1.00
24,988 (27.8) 4,279 (4.8) 33,482 (37.3) 27,059 (30.1)
1.00
(1.74—1.86)
4,379 (3.6) 117,611 96.4)
0.54 (0.51—0.56) 1.00
8,001 (8.9) 81,807 (91.1)
1.00
35.9 ± 15.9 1,901 (1.6) 5,931 (4.9) 13,120 (10.8) 45,345 (37.1) 51,086 (41.9) 4,607 (3.7)
0.87 2.27 4.20 5.84 2.92 1.00
(0.81—0.93) (2.15—2.40) (3.98—4.42) (5.58—6.12) (2.80—3.04)
38.8 ± 22.7 6,476 (7.2) 7,834 (8.7) 8,477 (9.4) 18,848 (21.0) 36,979 (41.2) 11,194 (12.5)
(1.43—1.51) (1.14—1.42) (1.08—1.29)
32,673 (26.7) 1,047 (0.9) 2,050 (1.7) 86,220 (70.7)
0.90 (0.88—0.92) 0.92 (0.86—0.99) 0.80 (0.74—0.88) 1.00
25,206 (28.1) 1,188 (1.3) 2,023 (2.3) 61,391 (68.3)
1.00
(1.11—1.15)
25,300 (20.7) 96,690 (79.3)
0.28 (0.28—0.29) 1.00
44,440 (49.5) 43,368 (50.5)
1.00
(0.73—0.78) (1.74—1.85) (1.71—1.81)
19,212 15,720 26,128 60,930
0.77 (0.75—0.79) 0.67 (0.65—0.69) 0.94 (0.92—0.96) 1.00
16,280 18,647 16,641 38,240
1.00
(2.25—2.52) (1.10—1.24) (1.37—1.53) (1.62—1.77) (1.27—1.38)
(15.7) (12.9) (21.4) (50.0)
(18.1) (20.8) (18.5) (42.6)
1.00
A.W. Selassie et al.
a
Migraine (n = 121,990) Frequency (%) ORa (95% CI)
ORa (95% CI)
Epilepsy beyond seizure Table 1 shows the distribution of the various comorbid conditions among PWE, PWM and PWLF. Of 35 conditions initially identified, four were dropped for very low counts (cysticercosis, onchocerciasis/toxocariasis, celiac disease, and somatoform disorders). The remaining conditions were grouped into two major categories: 18 somatic and 13 psychiatric/neurodevelopmental conditions. The most common somatic comorbidity was cardiovascular disease (CVD), a group that includes hypertension, ischemic heart disease and diseases of pulmonary circulation. Depression and anxiety were the most prevalent psychiatric comorbidities. PWE had significantly higher proportions of somatic and psychiatric/neurodevelopmental conditions than PWM; PWM had significantly higher proportions than PWLF. Table 2 shows the frequency and multivariable adjusted odds ratio (OR) of clinical and demographic risk characteristics in PWE and PWM compared with PWLF. The odds of both somatic and psychiatric/neurodevelopmental comorbidities was significantly higher in PWE than PWLF (OR = 5.44; 95% Confidence Interval [CI] = 5.25—5.63). The corresponding OR in PWM compared to PWLF was 2.49 (95% CI = 2.42—2.55). The absolute risk increase (ARI) of having both somatic and psychiatric/neurodevelopmental comorbidities in PWE compared to PWM is 18.5% [((5.44 − 2.49)/2.49) × 100]. The odds of psychiatric/neurodevelopmental comorbidities without a somatic comorbidity were 3.70 (95% CI = 3.51—3.90) and 1.51 (95% CI = 1.44—1.58) in PWE and PWM respectively, with an ARI of 145.0% in PWE. Although somatic comorbidities without psychiatric/neurodevelopmental comorbidity among PWE were significantly higher than in PWM, the gap was narrower (ARI = 11.6%). Compared to PWLF, the odds of mortality from any cause in PWE was 1.80 (95% CI = 1.74—1.86) vs 0.54 (95% CI = 0.51—0.56) among PWM, resulting in an ARI of 233.3% in PWE compared to PWM. There were important differences regarding the association of epilepsy and migraine with demographic characteristics, specifically in age, race/ethnicity, and payer status. PWE were 1.47 (95% CI = 1.43—1.51) times more likely to be black, and 1.79 and 1.76 times more likely to have coverage by Medicare and Medicaid, respectively, compared to PWLF. The corresponding ratios were inversely associated in PWM. Compared with PWLF, PWE were 2.4 times more likely to be between ages 0 and 5 compared to PWLF and PWM were 5.8 times more likely to be between ages 19 and 34. Among the 18 conditions listed as somatic comorbidities, the adjusted odds ratios were significantly higher in PWE than PWM except for intestinal problems, gastric reflux and peptic ulcer (Table 3). Stroke showed the strongest association with epilepsy, with PWE 4.20 (95% CI = 4.06—4.34) times more likely than PWLF to have been diagnosed with stroke. The odds of stroke in PWM were 1.65 (95% CI = 1.59—1.71) resulting in an ARI in PWE of 154.6% relative to PWM. Other neurological diseases with higher odds of occurrence included multiple sclerosis (MS) (OR 2.25; 95% CI = 1.98—2.54) and Parkinson’s disease (PD) (OR 2.45; 95% CI = 1.98—2.54). The corresponding odds in PWM were 1.68 (95% CI = 1.50—1.89) and 1.01 (95% CI = 0.88—1.16) for MS and PD respectively. The ARI of MS and PD among PWE compared to PWM is 33.9 and 142.6% for, respectively. Stark differences between PWE and PWM occurred in intellectual disability (ID), cognitive dysfunction, and autism
309 spectrum disorders (ASD). Compared to PWM, the ARI in PWE was 2791.7% higher for ID, 1487% higher for cognitive dysfunction, and 1730.6% higher for ASD. In all of the remaining psychiatric/neurodevelopmental comorbidities, the odds ratios were significantly higher in PWE compared to PWM, most notably in Alzheimer’s and schizophrenia with ARI of 388.3% and 259.0% in PWE, respectively. The strengths of the observed associations of epilepsy and migraine with the specific comorbid conditions are summarized in Table 4. The adjusted association of epilepsy with ID, cognitive dysfunction, and ASD was profoundly strong (OR ≥ 10). The association of epilepsy with psychiatric comorbidities and seven somatic comorbidities (stroke, MS, Parkinson’s, vision loss, HIV/AIDS, migraine, and nutritional deficiencies) was very strong (OR 2.0—9.99). The association of epilepsy with diabetes remained in the weak range while osteoporosis was inversely associated. The remaining conditions were strongly associated with epilepsy (OR 1.5—1.99). In regard to migraine, five somatic (diabetes, TBI, nutritional deficiencies, osteoporosis, and Parkinson’s disease) and four psychiatric/neurodevelopmental comorbidities (schizophrenia, alcoholism, Alzheimer’s dementia, and ID) showed neutral or inverse association. Six somatic (intestinal problems, asthma, gastric reflux, stroke, peptic ulcer and MS) and six psychiatric/neurodevelopmental comorbidities (depression, anxiety, personality disorder, psychosis, cognitive dysfunction and ADHD) were strongly associated with migraine while the rest showed weak associations.
Discussion This large, population-based, retrospective cohort study examined the association of somatic and psychiatric/neurodevelopmental comorbidities in PWE and PWM compared with PWLF. Somatic comorbidities occurred in 84.6% of PWE and 77.5% of PWM; psychiatric/neurodevelopmental comorbidities occurred in 62.1% of PWE and 45% of PWM. PWE had strongly elevated odds of having both somatic and psychiatric/neurodevelopmental comorbidities compared to PWM and PWLF. Furthermore, PWE had 80% higher mortality from all causes than PWLF and 126% higher than PWM. There was 18-fold increase in ASD, 16-fold increase in cognitive dysfunction, and 27-fold increase in ID among PWE compared to PWM. Epilepsy was significantly positively associated with all 31 specific comorbid conditions while migraine was significantly positively associated with 20 of 30 conditions. Unlike studies that reported the associations of epilepsy with comorbid conditions from self-report or subsets of population groups, our study evaluated multivariable adjusted associations of specific comorbidities in PWE compared with PWM and PWLF using clinically ascertained diagnoses. The study findings are consonant with the research priorities of the IOM report (England et al., 2012). To our knowledge, this is the first study of a statewide population with large minority and socioeconomically disadvantaged groups to examine the relationships of specific comorbidities with epilepsy compared with another neurological disorder and healthy controls.
310
A.W. Selassie et al.
Table 3 Association of common comorbidities in persons with epilepsy and migraine compared to lower extremity fracture controls. Comorbid condition
Epilepsy Adjusteda OR
Somatic comorbidities Cardiovascular disease Intestinal problems Asthma/pulmonary Gastric reflux Anemia Stroke Diabetes Peptic ulcer Traumatic brain injury Nutritional deficiency GI bleed Osteoporosis Vision loss Hearing loss Parkinson’s disease HIV/AIDS Multiple sclerosis Migraine
1.97 1.54 1.61 1.59 1.91 4.20 1.29 1.57 1.59 2.27 1.80 0.87 2.47 1.79 2.45 2.17 2.25 3.37
Psychiatric and neurodevelopmental Depression 2.12 Anxiety 2.29 Psychoses 3.17 Schizophrenia 3.15 Personality disorder 3.61 Alcoholism 1.77 Drug abuse 2.37 Suicidal ideation/attempt 2.95 Alzheimer’s dementia 2.19 Intellectual disability 12.88 Cognitive dysfunction 28.09 ADHD 2.31 Autism spectrum disorder 22.15 a
Migraine Adjusteda (95% CI)
OR
(95% CI)
(1.92—2.03) (1.50—1.58) (1.57—1.65) (1.55—1.63) (1.86—1.96) (4.06—4.34) (1.26—1.33) (1.52—1.61) (1.54—1.67) (2.18—2.37) (1.72—1.87) (0.83—0.91) (2.28—2.68) (1.67—1.93) (2.20—2.73) (1.91—2.45) (1.98—2.54) (3.23—3.52)
1.45 1.59 1.53 1.78 1.18 1.65 0.89 1.78 0.81 0.83 1.42 0.65 1.33 1.42 1.01 1.13 1.68
(1.42—1.49) (1.56—1.63) (1.50—1.56) (1.74—1.81) (1.15—1.21) (1.59—1.71) (0.86—0.91) (1.73—1.82) (0.78—0.83) (0.79—0.88) (1.36—1.48) (0.62—0.69) (1.22—1.45) (1.31—1.53) (0.88—1.16) (0.98—1.29) (1.50—1.89) —
(2.06—2.17) (2.23—2.35) (3.06—3.28) (2.94—3.37) (3.34—3.89) (1.71—1.82) (2.29—2.45) (2.81—3.10) (2.05—2.33) (11.59—14.30) (23.33—33.82) (2.16—2.47) (16.77—29.26)
1.62 1.87 1.51 0.85 1.67 0.60 1.19 1.44 0.61 0.48 1.77 1.56 1.21
(1.59—1.66) (1.82—1.91) (1.46—1.56) (0.78—0.92) (1.55—1.80) (0.58—0.62) (1.15—1.23) (1.38—1.52) (0.56—0.68) (0.41—0.57) (1.38—2.26) (1.46—1.67) (0.81—1.80)
Adjusted for age, race, gender, insurance status, and mortality status and number of comorbid conditions.
Somatic comorbidities in epilepsy Cardiovascular diseases (CVD) were the most prevalent somatic comorbidity overall (61.7%, 95% CI = 61.3—62.1) and strongly associated with epilepsy (OR = 1.97; 95% CI = 1.92—2.03), similar to the association seen in a US study of adults with seizures (Strine et al., 2005). Hypertension was the dominant comorbidity in the CVD category in our dataset contributing for the high prevalence (data not shown). Severe hypertension has been implicated in new onset seizure (Hesdorffer et al., 1996). Of the somatic comorbidities, stroke (27.1% of PWE) showed the strongest association with epilepsy (OR = 4.20; 95% CI = 4.06—4.34). This finding concurs with prior studies (Gaitatzis et al., 2012; Nuyen et al., 2006; Tellez-Zenteno et al., 2005). Antecedent stroke has been well-established as a cause of epilepsy (Benbir et al., 2006; Hussain et al., 2006; Mousali et al., 2009). However, epilepsy as a
antecedent risk factor of stroke remains tenuous because the common risk factors of stroke such as hypertension, diabetes, and dyslipidemia are prevalent in PWE making the association with pre-existing epilepsy uncertain (Cleary et al., 2004; Seidenberg et al., 2009; Shinton et al., 1987). Migraine (12.1% of PWE) was the second most strongly associated comorbidity with epilepsy (OR = 3.37; 95% CI = 3.23—3.52). Our study showed stronger association between migraine and epilepsy than other population-based studies (Ottman et al., 2011; Strine et al., 2005). Migraine shares many of the features of epilepsy, which obscures discrimination between the two conditions and indicates possible shared genetic and physiological mechanisms and bidirectional associations between the two (Kasteleijn-Nolst Trenite and Parisi, 2012; Seidenberg et al., 2009). PD and MS were less prevalent but strongly associated with epilepsy. The association of PD (OR = 2.45; 95% CI = 1.98—2.54) was similar to that seen in a
Epilepsy beyond seizure
311
Table 4 Summary of associations of epilepsy and migraine with specific comorbid conditions by their the strengths of association. Adjustment
Profoundly strong OR ≥ 10
Very strong 2.0 ≤ OR < 10.0
Strong 1.5 ≤ OR < 2.0
Weak 1.0 < OR < 1.5
Neutral or inverse OR ≤ 1.0
Epilepsy
Age, race, gender, insurance status, mortality status
Intellectual disability, cognitive dysfunction, ASD
Stroke, nutritional deficiency, vision loss, Parkinson’s disease, HIV/AIDS, MS, migraine, depression, anxiety, psychoses, schizophrenia, personality disorder, drug abuse, suicidal ideation/attempt, Alzheimer’s, ADHD
Cardiovascular, intestinal, asthma, gastric reflux, anemia, peptic ulcer, TBI, GI bleed, hearing loss, alcoholism
Diabetes
Osteoporosis
Migraine
Age, race, gender, insurance status, mortality status
Intestinal, asthma, gastric reflux, stroke, peptic ulcer, ms, depression, anxiety, personality disorder psychoses, cognitive dysfunction, ADHD
Cardiovascular, anemia, vision loss, hearing loss, GI bleed, HIV/AIDS, drug abuse, suicidal ideation, ASD
Diabetes, TBI, nutritional deficiency, osteoporosis, Parkinson’s, schizophrenia, alcoholism, Alzheimer’s, intellectual disability
population-based UK study (Gaitatzis et al., 2004a, 2012). However, the association of PD may be confounded by stroke and dementia that both share common underlying risk factors. We found a significant positive association of MS with epilepsy (OR = 2.25; 95% CI = 1.91—2.45) while others reported associations ranging from non-significant to a 3-fold increase in reference to the general population (Nicoletti et al., 2003; Nyquist et al., 2002). The underlying pathophysiological link between epilepsy and MS is speculated to involve cortical and subcortical demyelination and inflammation (Kelley and Rodriguez, 2009). The prevalence of diabetes in PWE (24.8%; 95% CI = 24.4—25.1) was significantly higher than the prevalence in the comparison groups. The adjusted association between epilepsy and diabetes in this study (OR = 1.29; 95% CI = 1.26—1.33) approximates the findings reported by Gaitatzis et al. (2004a).
Psychiatric and mental health comorbidities in epilepsy Depression was the most prevalent psychiatric comorbidity (31.3%) and was strongly associated with epilepsy (OR = 2.12;
95% CI = 2.06—2.17). Other studies reported similar prevalence and association (Hesdorffer et al., 2006; Rai et al., 2012). Although social stigma and unemployment may contribute to the high prevalence of depression among PWE, recent evidence suggest shared underlying neurochemical abnormalities may contribute to the bidirectional association between epilepsy and depression (Jensen, 2011; Seidenberg et al., 2009). Anxiety was the second most prevalent psychiatric comorbidity (29.1% in PWE) and was strongly associated with epilepsy (OR = 2.29; 95% CI = 2.23—2.35). The lifetime prevalence of anxiety disorders among PWE varies between 10 and 25% (Gaitatzis et al., 2004b) although higher rates have been reported with refractory epilepsy. The OR reported in our study is comparable to other population-based studies (Hesdorffer et al., 2012). The prevalence in PWM was similar (29.0%) despite significantly weaker association compared with epilepsy. Alcoholism and drug abuse have been identified as common mental health problems in PWE (Hillbom et al., 2003) as was evident in our cohort. Alcoholism was identified in 18.1% of PWE and was strongly associated with epilepsy (OR = 1.77; 95% CI = 1.71—1.82) consistent with prior reports
312 (Chan, 1985; Hauser et al., 1988). However, alcoholism often precedes epilepsy; the risk of seizure with daily alcohol consumption is reported to increase from 3-fold to 20-fold (Ng et al., 1988). In our study, drug abuse was identified in 16.9% of PWE and was strongly associated with epilepsy (OR = 2.37; 95% CI = 2.29—2.45). There is limited information in the literature regarding the prevalence of drug abuse in PWE. The very strong association of schizophrenia with epilepsy (OR = 3.61; 95% CI = 3.34—3.89) despite its low prevalence (5.6% in PWE) indicate the importance of schizophrenia in PWE, suggesting the need for coordinated patient care with a neuropsychiatry team. Previously reported prevalence rates of schizophrenia range from 4.3 to 18% (Gaitatzis et al., 2004b). A comparable longitudinal population-based study from Denmark reported a relative risk of 2.48 in PWE (95% CI = 2.20—2.80) (Qin et al., 2005). The mechanism by which epilepsy and schizophrenia are associated may relate to cortical dysgenesis or diffuse brain lesions that underlie both conditions or confounding sociobehavioral factors (Sachdev, 1998). Similarly, despite the low prevalence (4.3% in PWE), personality disorder was very strongly associated with epilepsy (OR = 3.61; 95% CI = 3.34—3.89) in our study. The prevalence of personality disorder among PWE is reported to range from 5 to 18% depending on cortical lesion and type of epilepsy with higher rates in temporal lobe epilepsy (Devinsky, 2003). Suicidal ideation and attempt among PWE has been evaluated extensively. In our study, the prevalence of suicidal ideation in PWE was 8.8% and the observed association with epilepsy was strong (OR = 2.95; 95% CI = 2.81—3.10). This result is comparable with a meta-analysis of 23 studies (Jones et al., 2003) and a study that examined suicidal ideation as function of onset of epilepsy (Hesdorffer et al., 2012). Risk of suicidal ideation among PWE is significantly increased in the presence of other psychiatric comorbidities, especially major depression and anxiety disorders (Gandy et al., 2013). Suicidal ideation may also be associated with unprovoked seizures in children independent of depression (Hesdorffer et al., 2006). The prevalence of Alzheimer’s dementia (AD) was 5.0% in PWE with strong association (OR = 2.19; 95% CI = 2.05—2.33). A health interview survey found an unadjusted prevalence ratios ranging from of 4.3 to 8.1 times the prevalence in the general population, although a prospective study showed no significant increased risk in PWE (Gaitatzis et al., 2004a; Scarmeas et al., 2009; Tellez-Zenteno et al., 2005). The attributed underlying pathophysiological phenomenon between epilepsy and AD is a progressive neurodegenerative process that begins with cognitive decline (Amatniek et al., 2006; Palop and Mucke, 2009).
Neurodevelopmental comorbidities in epilepsy While the prevalence of cognitive dysfunction was only 3.6% in PWE, the association was profoundly strong (OR = 28.09; 95% CI = 23.33—33.82). Cognitive dysfunction is predominately diagnosed in young children with epilepsy, although memory deficits are also common in temporal lobe epilepsy in adults (Jensen, 2011). In our cohort of PWE, 55% of cognitive dysfunction was diagnosed in age ≤5, 26% in ages 6—18, and 9% in ages 19—34. In a community-based study,
A.W. Selassie et al. 26.4% of children had evidence of cognitive decline, mostly attributable to epileptic encephalopathies (Berg et al., 2008). Similarly, our study noted a prevalence rate of 7.3% in ID among PWE and a profoundly strong association with epilepsy (OR = 12.88; 95% CI = 11.59—1.40). The prevalence of ID in SC is reported to be 1.13% in the general population (Massey and McDermott, 1996). The prevalence of autism spectrum disorders (ASD) in PWE was low (1.3%) but the association was profoundly strong (OR = 22.15; 95% CI = 16.77-29.26). ASD are a common comorbidity of epilepsy in children and approximately 75% of those with ASD were children and adolescents giving a prevalence of 4.3% for those 18 and under. The prevalence of ASD in the general US population of children with epilepsy is undetermined but a community-based study of childhood-onset epilepsy reported 5% prevalence of ASD (Berg et al., 2011). A UK study found a prevalence of 8.1% (95% CI = 2.2—25.9) in non-institutionalized adults with epilepsy (Rai et al., 2012). The prevalence of attention deficit/hyperactivity disorder (ADHD) was 3.4% and the association of ADHD with epilepsy was strong (OR = 2.31; 95% CI 2.16—2.47). ADHD has been described as one of the commonly co-occurring comorbidities of epilepsy in children (Parisi et al., 2010). The prevalence in our study was 12.8% (95% CI = 11.2—14.4) in children 18 and younger, similar to that seen in descriptive study of 175 children and adolescents with epilepsy (Dunn et al., 2003). ADHD has been associated with increased risk for developing unprovoked seizure (Hesdorffer et al., 2004). Factors that may contribute to the association between epilepsy and ADHD are chronic seizures, repetitive epileptiform discharges, and therapeutic medications (Parisi et al., 2010).
Strengths and limitations Our study has several strengths. First, data come from a legally mandated, multifaceted medical database of all healthcare encounters spanning 12 years. Use of a medical database overcomes the drawback of relying on self-report used in many epidemiological studies of epilepsy. Second, the findings are generalizable because cases and controls are representative of the referent population. Third, case ascertainment relied on ICD-9-CM codes following the ILAE guidelines and listing of routinely associated comorbidities for epidemiological studies (International League Against Epilepsy Commission on Epidemiology and Prognosis, 1993). Fourth, the methodological and analytical approaches of comparing epilepsy, migraine, and LEF as polytomous discrete responses and the very large sample size provided robust estimates and within group comparisons. Fifth, comparison with a similar chronic neurological disease (migraine) allowed estimation of ARI attributable to epilepsy. Despite the aforementioned strengths, there are important limitations worth noting. First, the data do not include PWE who receive healthcare in military and veteran affairs hospitals that may have higher rates of epilepsy. This limitation, however, applies to all public health data systems in the US. Second, analysis relied on administrative data designed for billing third-party providers and a
Epilepsy beyond seizure
313
coding practice responsive to the policies of providers, possibly influencing estimation of the comorbid conditions. Third, with wide variability in skillset and diagnostic resources among the hospitals, the accuracy of the 4th and 5th digit of the diagnosis codes that provide much of the desired diagnostic specificity might be unreliable. Nonetheless, previous validation studies involving medical chart review of randomly selected records conducted for various prior projects using the SC data system found that coding error is very low and unlikely to sway the observed results (data not shown). Fourth, we have not accounted for venue of care. Inpatient hospital encounters may emphasize specific comorbidities more than others, and identify more comorbidities overall due to more extensive workup. However, this bias is likely to affect all population groups equally; those with epilepsy, migraine or LEF who require hospitalization are likely to have more comorbidities compared with those treated and released in the ED. Fifth, we are unable to obtain data on encounters in physician offices and other outpatient settings resulting in a bias towards more complex cases. In summary, this study provides population-based distribution and associations of a set of somatic and psychiatric/neurodevelopmental comorbidities among PWE in comparison with PWM and PWLF. Epilepsy was positively associated with all of the comorbidities examined except for osteoporosis. Further, comorbidities, particularly neurodevelopmental comorbidities, were more strongly associated with epilepsy than with migraine or LEF.
Acknowledgements This study was funded by cooperative agreement DP003251 from the National Center for Chronic Disease Prevention and Health Promotion, the Centers for Disease Control and Prevention. The findings of the study and opinions presented in the manuscript are those of the authors and do not reflect the opinion of the funding agencies. The authors recognize the support of Mr. Chris Finney from the Office of Research and Statistics, South Carolina Budget and control Board, for selecting and organizing the data from multiple administrative data sources.
Appendix A. ICD-9-CM codes for case and control group and comorbid condition definitions ICD-9-CM code(s) Cohort status Case: Epilepsy Control 1: Migraine Control 2: Tibia, Fibula, Ankle fractures Comorbidity ADHD Alcoholism
Alzheimer’s dementia
345.1, 345.4—345.9 346 823, 824
314.0 291.1, 291.2, 291.5, 291.8, 291.9, 303.90, 303.93, 305.00, 305.03, V11.3 331.0
Appendix (Continued ) ICD-9-CM code(s) Anemia Anxiety Asthma/pulmonary disease Autism spectrum disorder Cardiovascular disease Celiac disease Cognitive dysfunction Cysticercoids Depression Diabetes
Drug abuse
Gastric reflux GI bleed Hearing loss HIV/AIDS Intestinal problems Intellectual disability Migraine Multiple sclerosis Nutritional deficiencies Onchocerciasis/toxocariasis Osteoporosis Parkinson’s disease Peptic ulcer Personality disorder Psychoses Schizophrenia Somatoform disorder Stroke Suicidal ideation/attempt Traumatic brain injury Vision loss
280.1—281.9, 285.9 300.0, 300.7 490—496 299.0 401—405, 410—417, 420—429 579.0 315, V40.0 123.1 300.4, 309.0, 309.1, 311 250.00—250.33, 250.40—250.73, 250.90—250.93 292.0, 292.82—292.89, 292.9, 304.00—304.93, 305.20—305.93 530.81 578 389 042, 044 560—569 317—319 346 340 260—269 125.3, 128.0 733.0, V82.81 332 531—535 301 293.8, 297—298.9, 299.10, 299.11 295 300.8 430—438 300.9, V62.84 800, 801, 803, 804, 850—854, 959.01 369
References Amatniek, J.C., Hauser, W.A., DelCastillo-Castaneda, C., Jacobs, D.M., Marder, K., Bell, K., et al., 2006. Incidence and predictors of seizures in patients with Alzheimer’s disease. Epilepsia 47 (5), 867—872, http://dx.doi.org/10.1111/ j. 1528-1167.2006.00554.x. American Medical Association, 2009. International Classification of Diseases, 9th revision, Clinical Modification. Saunders, Chicago, IL. Baca, C.B., Vickrey, B.G., Caplan, R., Vassar, S.D., Berg, A.T., 2011. Psychiatric and medical comorbidity and quality of life outcomes in childhood-onset epilepsy. Pediatrics 128 (6), e1532—e1543, http://dx.doi.org/10.1542/peds.2011-0245. Balottin, U., Chiappedi, M., Rossi, M., Termine, C., Nappi, G., 2011. Childhood and adolescent migraine: a neuropsychiatric disorder? Med. Hypotheses 76 (6), 778—781, http://dx.doi.org/ 10.1016/j.mehy.2011.02.016.
314 Benbir, G., Ince, B., Bozluolcay, M., 2006. The epidemiology of post-stroke epilepsy according to stroke subtypes. Acta Neurol. Scand. 114 (1), 8—12, http://dx.doi.org/10.1111/j. 1600-0404.2006.00642.x. Berg, A.T., Caplan, R., Hesdorffer, D.C., 2011. Psychiatric and neurodevelopmental disorders in childhood-onset epilepsy. Epilepsy Behav. 20 (3), 550—555, http://dx.doi.org/10.1016/ j.yebeh.2010.12.038. Berg, A.T., Langfitt, J.T., Testa, F.M., Levy, S.R., DiMario, F., Westerveld, M., et al., 2008. Global cognitive function in children with epilepsy: a community-based study. Epilepsia 49 (4), 608—614, http://dx.doi.org/10.1111/j. 1528-1167.2007.01461.x. Biesheuvel, C.J., Vergouwe, Y., Steyerberg, E.W., Grobbee, D.E., Moons, K.G., 2008. Polytomous logistic regression analysis could be applied more often in diagnostic research. J. Clin. Epidemiol. 61 (2), 125—134, http://dx.doi.org/10.1016/ j.jclinepi.2007.03.002. Chan, A.W., 1985. Alcoholism and epilepsy. Epilepsia 26 (4), 323—333. Chiappedi, M., Beghi, E., Ferrari-Ginevra, O., Ghezzo, A., Maggioni, E., Mattana, F., et al., 2011. Response to rehabilitation of children and adolescents with epilepsy. Epilepsy Behav. 20 (1), 79—82, http://dx.doi.org/10.1016/j.yebeh.2010.10.029. Cleary, P., Shorvon, S., Tallis, R., 2004. Late-onset seizures as a predictor of subsequent stroke. Lancet 363 (9416), 1184—1186, http://dx.doi.org/10.1016/S0140-6736(04)15946-1. Darlington, G.A., Armtage, P., Colton, T., 1998. Collinearity Encyclopedia of Biostatistics. John Wiley & Sons, Chichester, West Sussex, UK, pp. 788—789. Devinsky, O., 2003. Psychiatric comorbidity in patients with epilepsy: implications for diagnosis and treatment. Epilepsy Behav. 4 (Suppl. 4), S2—S10. Dunn, D.W., Austin, J.K., Harezlak, J., Ambrosius, W.T., 2003. ADHD and epilepsy in childhood. Dev. Med. Child Neurol. 45 (1), 50—54. England, M.J., Liverman, C.T., Schultz, A.M., Strawbridge, L.M., 2012. Epilepsy across the spectrum: promoting health and understanding. A summary of the Institute of Medicine report. Epilepsy Behav. 25 (2), 266—276, http://dx.doi.org/10.1016/ j.yebeh.2012.06.016. Fienberg, S.E., 1989. Sampling Models. In: Press, T.M. (Ed.), The Analysis of Cross-Classified Categorical Data. The Massachusetts Institute of Technology, Cambridge, MA, pp. 29—32. Gaitatzis, A., Carroll, K., Majeed, A., J., W.S., 2004a. The epidemiology of the comorbidity of epilepsy in the general population. Epilepsia 45 (12), 1613—1622, http://dx.doi.org/10.1111/j. 0013-9580.2004.17504.x. Gaitatzis, A., Sisodiya, S.M., Sander, J.W., 2012. The somatic comorbidity of epilepsy: a weighty but often unrecognized burden. Epilepsia 53 (8), 1282—1293, http://dx.doi.org/10.1111/j. 1528-1167.2012.03528.x. Gaitatzis, A., Trimble, M.R., Sander, J.W., 2004b. The psychiatric comorbidity of epilepsy. Acta Neurol. Scand. 110 (4), 207—220, http://dx.doi.org/10.1111/j. 1600-0404.2004.00324.x. Gandy, M., Sharpe, L., Perry, K.N., Miller, L., Thayer, Z., Boserio, J., et al., 2013. Rates of DSM-IV mood, anxiety disorders, and suicidality in Australian adult epilepsy outpatients: a comparison of well-controlled versus refractory epilepsy. Epilepsy Behav. 26 (1), 29—35, http://dx.doi.org/10.1016/j.yebeh.2012.10.023. Hauser, W.A., Ng, S.K., Brust, J.C., 1988. Alcohol, seizures, and epilepsy. Epilepsia 29 (Suppl. 2), S66—S78. Hesdorffer, D.C., Caplan, R., Berg, A.T., 2012. Familial clustering of epilepsy and behavioral disorders: evidence for a shared genetic basis. Epilepsia 53 (2), 301—307, http://dx.doi.org/10.1111/j. 1528-1167.2011.03351.x. Hesdorffer, D.C., Hauser, W.A., Annegers, J.F., Rocca, W.A., 1996. Severe, uncontrolled hypertension and adult-onset seizures: a case-control study in Rochester, Minnesota. Epilepsia 37 (8), 736—741.
A.W. Selassie et al. Hesdorffer, D.C., Hauser, W.A., Olafsson, E., Ludvigsson, P., Kjartansson, O., 2006. Depression and suicide attempt as risk factors for incident unprovoked seizures. Ann. Neurol. 59 (1), 35—41, http://dx.doi.org/10.1002/ana.20685. Hesdorffer, D.C., Ludvigsson, P., Olafsson, E., Gudmundsson, G., Kjartansson, O., Hauser, W.A., 2004. ADHD as a risk factor for incident unprovoked seizures and epilepsy in children. Arch. Gen. Psychiatry 61 (7), 731—736, http://dx.doi.org/10.1001/ archpsyc.61.7.731. Hillbom, M., Pieninkeroinen, I., Leone, M., 2003. Seizures in alcohol-dependent patients: epidemiology, pathophysiology and management. CNS Drugs 17 (14), 1013—1030. Hussain, S.A., Haut, S.R., Lipton, R.B., Derby, C., Markowitz, S.Y., Shinnar, S., 2006. Incidence of epilepsy in a racially diverse, community-dwelling, elderly cohort: results from the Einstein aging study. Epilepsy Res. 71 (2-3), 195—205, http://dx.doi.org/10.1016/j.eplepsyres.2006.06.018. International League Against Epilepsy Commission on Epidemiology and Prognosis, 1993. Guidelines for epidemiologic studies on epilepsy. Commission on epidemiology and prognosis, international league against epilepsy. Epilepsia 34 (4), 592—596. Jensen, F.E., 2011. Epilepsy as a spectrum disorder: implications from novel clinical and basic neuroscience. Epilepsia 52 (Suppl. 1), 1—6, http://dx.doi.org/10.1111/j.1528-1167.2010.02904.x. Jones, J.E., Hermann, B.P., Barry, J.J., Gilliam, F.G., Kanner, A.M., Meador, K.J., 2003. Rates and risk factors for suicide, suicidal ideation, and suicide attempts in chronic epilepsy. Epilepsy Behav. 4 (Suppl. 3), S31—S38. Kasteleijn-Nolst Trenite, D., Parisi, P., 2012. Migraine in the borderland of epilepsy: migralepsy an overlapping syndrome of children and adults? Epilepsia 53 (Suppl. 7), 20—25, http://dx.doi.org/10.1111/j. 1528-1167.2012.03711.x. Kelley, B.J., Rodriguez, M., 2009. Seizures in patients with multiple sclerosis: epidemiology, pathophysiology and management. CNS Drugs 23 (10), 805—815, http://dx.doi.org/ 10.2165/11310900-000000000-00000. Kobau, R., Zahran, H., Thurman, D.J., Zack, M.M., Henry, T.R., Schachter, S.C., et al., 2008. Epilepsy surveillance among adults — 19 States, Behavioral Risk Factor Surveillance System, 2005. MMWR Surveill. Summ. 57 (6), 1—20. Linehan, C., Tellez-Zenteno, J.F., Burneo, J.G., Berg, A.T., 2011. Future directions for epidemiology in epilepsy. Epilepsy Behav. 22 (1), 112—117, http://dx.doi.org/ 10.1016/j.yebeh.2011.06.006. Massey, P.S., McDermott, S., 1996. State-specific rates of mental retardation — United States, 1993. MMWR 45 (3), 61—65. Mousali, Y.M., Obeid, T.H., Al-Gahtani, H.A., Al-Janadi, M.A., Attar, A.A., Dammas, H.O., 2009. Epilepsy in patients 60 years and above. Neurosciences (Riyadh) 14 (1), 53—55. National Institutes of Health, 2012. Epilepsy Research Benchmarks: Progress Report 2007—2012. National Institute of Neurological Disorders and Stroke, Bethesda, MD. Ng, S.K., Hauser, W.A., Brust, J.C., Susser, M., 1988. Alcohol consumption and withdrawal in new-onset seizures. N. Engl. J. Med. 319 (11), 666—673, http://dx.doi.org/10.1056/ NEJM198809153191102. Nicoletti, A., Sofia, V., Biondi, R., Lo Fermo, S., Reggio, E., Patti, F., et al., 2003. Epilepsy and multiple sclerosis in Sicily: a population-based study. Epilepsia 44 (11), 1445—1448. Nuyen, J., Schellevis, F.G., Satariano, W.A., Spreeuwenberg, P.M., Birkner, M.D., van den Bos, G.A., et al., 2006. Comorbidity was associated with neurologic and psychiatric diseases: a general practice-based controlled study. J. Clin. Epidemiol. 59 (12), 1274—1284, http://dx.doi.org/10.1016/j.jclinepi.2006.01.005. Nyquist, P.A., Cascino, G.D., McClelland, R.L., Annegers, J.F., Rodriguez, M., 2002. Incidence of seizures in patients with multiple sclerosis: a population-based study. Mayo Clin. Proc. 77 (9), 910—912, http://dx.doi.org/10.4065/77.9.910.
Epilepsy beyond seizure Ottman, R., Lipton, R.B., Ettinger, A.B., Cramer, J.A., Reed, M.L., Morrison, A., et al., 2011. Comorbidities of epilepsy: results from the Epilepsy Comorbidities and Health (EPIC) survey. Epilepsia 52 (2), 308—315, http://dx.doi.org/10.1111/j. 1528-1167.2010.02927.x. Palop, J.J., Mucke, L., 2009. Epilepsy and cognitive impairments in Alzheimer disease. Arch. Neurol. 66 (4), 435—440, http://dx.doi.org/10.1001/archneurol.2009.15. Parisi, P., Moavero, R., Verrotti, A., Curatolo, P., 2010. Attention deficit hyperactivity disorder in children with epilepsy. Brain Dev. 32 (1), 10—16, http://dx.doi.org/10.1016/ j.braindev.2009.03.005. Qin, P., Xu, H., Laursen, T.M., Vestergaard, M., Mortensen, P.B., 2005. Risk for schizophrenia and schizophrenia-like psychosis among patients with epilepsy: population based cohort study. BMJ 331 (7507), 23, http://dx.doi.org/10.1136/ bmj.38488.462037.8F. Rai, D., Kerr, M.P., McManus, S., Jordanova, V., Lewis, G., Brugha, T.S., 2012. Epilepsy and psychiatric comorbidity: a nationally representative population-based study. Epilepsia 53 (6), 1095—1103, http://dx.doi.org/10.1111/ j. 1528-1167.2012.03500.x. Sachdev, P., 1998. Schizophrenia-like psychosis and epilepsy: the status of the association. Am. J. Psychiatry 155 (3), 325—336. SAS Institute, 2011. Statistical Analytical Software (Version 9.3.1). SAS Institute Inc., Cary, North Carolina. Scarmeas, N., Honig, L.S., Choi, H., Cantero, J., Brandt, J., Blacker, D., et al., 2009. Seizures in Alzheimer disease: who, when, and how common? Arch. Neurol. 66 (8), 992—997, http://dx.doi.org/ 10.1001/archneurol.2009.130. Seidenberg, M., Pulsipher, D.T., Hermann, B., 2009. Association of epilepsy and comorbid conditions. Future Neurol. 4 (5), 663—668, http://dx.doi.org/10.2217/fnl.09.32.
315 Shinton, R.A., Gill, J.S., Zezulka, A.V., Beevers, D.G., 1987. The frequency of epilepsy preceding stroke. Case-control study in 230 patients. Lancet 1 (8523), 11—13. Stokes, M.E., Davis, C.S., Koch, G.G., 1997. Logistic regression II: polytomous response. In: SAS Institute (Ed.), Categorical Data Analysis using the SAS System. 3rd ed. SAS Institute, Cary, NC. Strine, T.W., Kobau, R., Chapman, D.P., Thurman, D.J., Price, P., Balluz, L.S., 2005. Psychological distress, comorbidities, and health behaviors among U.S. adults with seizures: results from the 2002 National Health Interview Survey. Epilepsia 46 (7), 1133—1139, http://dx.doi.org/ 10.1111/j.1528-1167.2005.01605.x. Swartz, K.L., Pratt, L.A., Armenian, H.K., Lee, L.C., Eaton, W.W., 2000. Mental disorders and the incidence of migraine headaches in a community sample: results from the Baltimore Epidemiologic Catchment area follow-up study. Arch. Gen. Psychiatry 57 (10), 945—950. Tellez-Zenteno, J.F., Matijevic, S., Wiebe, S., 2005. Somatic comorbidity of epilepsy in the general population in Canada. Epilepsia 46 (12), 1955—1962, http://dx.doi.org/10.1111/j. 1528-1167.2005.00344.x. Thurman, D.J., Beghi, E., Begley, C.E., Berg, A.T., Buchhalter, J.R., Ding, D., et al., 2011. Standards for epidemiologic studies and surveillance of epilepsy. Epilepsia 52 (Suppl. 7), 2—26, http://dx.doi.org/10.1111/j.1528-1167.2011.03121.x. Weis, M.A., Bradberry, C., Carter, L.P., Ferguson, J., Kozareva, D., 2006. An exploration of human services system contacts prior to suicide in South Carolina: an expansion of the South Carolina Violent Death Reporting System. Inj. Prev. 12 (Suppl. 2), ii17—ii21, http://dx.doi.org/10.1136/ip.2006.012427.
journal homepage: www.elsevier.com/locate/epilepsyres
Epilepsy beyond seizure: A population-based study of comorbidities Anbesaw W. Selassie, Dulaney A. Wilson, Gabriel U. Martz, Georgette G. Smith, Janelle L. Wagner, Braxton B. Wannamaker ∗ Medical University of South Carolina, United States Received 17 June 2013; received in revised form 25 October 2013; accepted 1 December 2013 Available online 18 December 2013
KEYWORDS Epilepsy; Comorbidity; Epidemiology; Migraine; Mortality; Risk set
∗
Summary Comorbid conditions may affect the quality of life in persons with epilepsy (PWE) more than seizures. Using legally mandated healthcare encounter data, somatic, psychiatric, and neurodevelopmental comorbidities in a large population-based cohort of PWE, were compared to persons with migraine (PWM), a similar neurologic condition, and lower extremity fracture (PWLF), otherwise healthy controls. 64,188 PWE, 121,990 PWM, and 89,808 PWLF were identified from inpatient, outpatient, and emergency department from 2000 to 2011. Epilepsy was ascertained with ICD-9-CM code 345; migraine with 346; fracture of the tibia, fibula, and ankle with 823 and 824. Common comorbidities of epilepsy were identified from the literature. Differences in prevalence among PWE, PWM, and PWLF were assessed by comparison of 95% confidence intervals (CI) constructed under the assumption of independence and normal approximation. The association of the comorbid conditions with epilepsy and migraine, compared to lower extremity fracture, were evaluated with polytomous logistic regression controlling for demographic and mortality covariables. PWE had significantly elevated prevalence of comorbidities compared with PWM and PWLF. Compared with PWLF, the adjusted odds ratios (OR) of having both somatic and psychiatric/neurodevelopmental comorbidities were 5.44 (95% CI = 5.25—5.63) and 2.49 (95% CI = 2.42—2.55) in PWE and PWM, respectively. The association with epilepsy was the strongest for cognitive dysfunction (OR = 28.1; 95% CI = 23.3—33.8); autism spectrum disorders (OR = 22.2; 95% CI = 16.8—29.3); intellectual disability (OR = 12.9; 95% CI = 11.6—14.3); and stroke (OR = 4.2; 95% CI = 4.1—4.4). The absolute risk increase in PWE compared with PWM for any somatic or psychiatric/neurodevelopmental comorbidity was 58.8% and 94.3%, respectively. Identifying comorbidities that are strongly and consistently associated with seizures, particularly disorders with shared underlying pathophysiology, is critical in identifying specific research and practice goals that may ultimately improve the quality of life for PWE. This study contributes to that effort by providing population-based comorbidity data for PWE compared with PWM and PWLF. © 2013 Elsevier B.V. All rights reserved.
Corresponding author at: 135 Cannon Street, Suite 303, MSC 835, Charleston, SC 29425, United States. Tel.: +1 843 876 1140. E-mail address: [email protected] (B.B. Wannamaker).
0920-1211/$ — see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.eplepsyres.2013.12.002
306
A.W. Selassie et al.
Introduction
Study setting and population
Various somatic, psychiatric, and neurodevelopmental conditions associated with epilepsy, other than seizures, have been recognized for many years and can impact quality of life more than seizures (Baca et al., 2011; Kobau et al., 2008; Linehan et al., 2011) and reduce rehabilitation efficacy (Chiappedi et al., 2011). These conditions are said to be comorbidities of epilepsy when they occur more frequently in persons with epilepsy (PWE) than in the general population (Thurman et al., 2011). In most instances, it is not known whether the comorbid condition was caused by epilepsy or its treatments, led to epilepsy, or has a common underlying cause with epilepsy (Seidenberg et al., 2009). Epidemiological studies can facilitate a better understanding of the burden that these comorbidities impose on PWE. In the clinic, PWE with multiple comorbidities are common and the treatment of comorbid conditions is often more difficult than treating the seizures. This study undertakes an analysis of statewide healthcare encounter data to address comorbidities associated with epilepsy in order to inform clinical practice and public health policy. The database is very large and includes all age groups, diverse socioeconomic and demographic groups thus permitting robust cohort sizes for subset analyses and comparison with other medical conditions. Following the recommendations of the ILAE (Thurman et al., 2011), a control group of persons with lower extremity fracture (PWLF) without any clinical evidence of pathological fractures were selected and should represent an otherwise healthier patient population with only intermittent healthcare needs and fewer counts of comorbid conditions relatively comparable to the general population. A neurological comparative group, people with migraine (PWM), was also selected to address the research gaps identified in comorbidities of epilepsy (National Institutes of Health, 2012). PWM has many similar characteristics to PWE regarding chronicity, earlylife onset, intermittent and variable symptoms, medical therapy including and rescue and prophylactic medications, genetic or unknown etiologies, and an increased risk of psychological factors with bidirectional association with migraine, compared to healthy controls (Balottin et al., 2011; Kasteleijn-Nolst Trenite and Parisi, 2012; Swartz et al., 2000). This study hypothesizes PWE would have higher odds of chronic comorbid conditions than PWM as compared with PWLF independent of demographic and clinical characteristics.
SC residents hospitalized or treated and released from the ED and OPD from January 1, 2000 to December 31, 2011 were eligible for the study. Cases and control groups were ascertained using International Classification of Diseases, 9th revision, Clinical Modification (ICD-9-CM) diagnosis codes for epilepsy (345.x), migraine (346.x), and fracture of the tibia, fibula, or ankle (823.x and 824.x) assigned by clinical providers at discharge (American Medical Association, 2009). PWLF without any history of seizure represent healthcare controls that approximate the pattern of comorbidity in the general population. Such fractures occur as isolated trauma without damage to adjacent vital organs and are less likely to be secondary to underlying pathologies of the bone. Persons with epilepsy and migraine were classified as PWE with comorbid migraine; therefore, no PWM had epilepsy.
Methods Data sources This study utilized the South Carolina (SC) statewide hospital discharge and emergency department (ED) visit datasets, which includes hospital-based outpatient department (OPD) visits reported via the uniform billing system (UB-04). In SC, health care providers are legally required to submit data to the Office of Research and Statistics (ORS), the SC Budget and Control Board (Weis et al., 2006). The Medical University of South Carolina Institutional Review Board exempted this study.
Definitions PWE were defined based on at least one ICD-9-CM code for epilepsy (345.x), excluding petit mal status (345.2) and grand mal status (345.3), during the study period. Prevalent comorbidities demonstrated in the epilepsy literature (Rai et al., 2012; Thurman et al., 2011) were chosen for the current novel comparative analysis between epilepsy, migraine, and lower extremity fracture (LEF). The ICD-9-CM codes used to define these comorbidities are listed in Appendix A. Race-ethnicity was categorized into non-Hispanic white, non-Hispanic black, Hispanic, and all others. Mortality status from any cause was determined for the total cohort through December 31, 2011 based on the SC Multiple Causes of Death Data file. All diagnosis fields were searched for the specified comorbid conditions. The conditions were further grouped as somatic or psychiatric/neurodevelopmental disorders. Using polytomous logistic regression the association of epilepsy with the comorbid conditions was examined in comparison with migraine and LEF adjusting for demographic characteristics and mortality status. The cohort size, frequency of clinical encounters and the number of comorbid conditions per capita are shown in Table 1. The remaining variables were classified as shown in Table 2.
Statistical analysis Data were analyzed with SAS software package, V9.3.1 (SAS Institute, 2011). Demographic characteristics and the comorbid conditions were treated as independent variables. In accordance with product-multinomial sampling design, marginal frequencies corresponding to cases (PWE) and comparison groups (PWM and PWLF) were fixed a priori (Fienberg, 1989). Proportions corresponding to demographic characteristics and the specific comorbid conditions were compared among PWE, PWM, and PWLF by constructing 95% confidence intervals (CI) under the assumption of independence and normal approximation (Table 1). Overlapping confidence intervals suggest no significant difference in proportions. Taking epilepsy, migraine, and LEF as nominal response categories, association with comorbid conditions and demographics were examined in polytomous logistic regression.
Epilepsy beyond seizure Table 1
307
Proportionate distribution of comorbid conditions.
Comorbidity type
Epilepsy (n = 64,188) % (95% CI)
Migraine (n = 121,990) % (95% CI)
Lower ext fracture (n = 89,808) % (95% CI)
5-year median number of visits (IQR)* Median count of comorbidity (IQR)* % with polymorbidity (≥6 conditions) Somatic disorders Cardiovascular disease Intestinal problems Asthma/pulmonary disease Gastric reflux Anemia Stroke Diabetes Peptic ulcer Traumatic brain injury Nutritional deficiency GI bleed Osteoporosis Vision loss Hearing loss Parkinson’s disease HIV/AIDS Multiple sclerosis Migraine Psychiatric/neurodevelopmental disorders Depression Anxiety Psychoses Alcoholism Drug abuse Suicidal ideation/attempt Intellectual disability Schizophrenia Alzheimer’s dementia Personality disorder Cognitive dysfunction ADHD Autism spectrum disorder
15 (8—27) 4 (2—7) 38.5 (38.1—38.9)
13 (6—24) 3 (1—5) 20.7 (20.4—20.9)a
7 (4—14) 2 (0—4) 14.3 (14.0—14.5)a,b
61.7 (61.3—62.1) 38.2 (37.9—38.6) 36.4 (36.0—36.7) 34.5 (34.2—34.9) 28.9 (28.5—29.2) 27.1 (26.8—27.5) 24.8 (24.4—25.1) 23.8 (23.5—24.2) 14.5 (14.2—14.7) 13.4 (13.1—13.6) 10.4 (10.2—10.7) 6.0 (5.9—6.2) 3.4 (3.2—3.5) 2.9 (2.8—3.1) 1.8 (1.7—1.9) 1.3 (1.2—1.4) 1.1 (1.1—1.2) 12.1 (11.9—12.4)
46.2 (45.9—46.5)a 34.5 (34.2—34.7)a 30.5 (30.2—30.7)a 33.6 (33.3—33.8)a 16.7 (16.4—16.9)a 8.0 (7.9—8.2)a 13.7 (13.5—13.9)a 23.7 (23.4—23.9) 7.1 (6.9—7.2)a 2.9 (2.8—3.0)a 6.3 (6.2—6.4)a 3.1 (3.0—3.2)a 1.1 (1.1—1.2)a 1.6 (1.5—1.6)a 0.3 (0.3—0.3)a 0.5 (0.4—0.5)a 1.1 (1.1—1.2) 100.0 (100.0—100.0)a
41.4 (41.1—41.8)a,b 24.7 (24.4—25.0)a,b 21.5 (21.2—21.7)a,b 21.0 (20.8—21.3)a,b 13.6 (13.4—13.8)a,b 7.3 (7.1—7.5)a,b 15.8 (15.5—16.0)a,b 13.5 (13.3—13.7)a,b 9.2 (9.0—9.4)a,b 4.5 (4.4—4.6)a,b 4.7 (4.5—4.8)a,b 5.2 (5.1—5.4)a,b 1.0 (1.0—1.1)a 1.4 (1.3—1.5)a 0.6 (0.5—0.6)a,b 0.4 (0.4—0.5)a 0.4 (0.4—0.4)a,b 4.5 (4.4—4.6)a,b
31.3 (30.9—31.6) 29.1 (28.7—29.4) 21.5 (21.2—21.8) 18.1 (17.8—18.4) 16.9 (16.6—17.2) 8.8 (8.6—9.0) 7.3 (7.1—7.5) 5.6 (5.4—5.7) 5.0 (4.8—5.2) 4.3 (4.2—4.5) 3.6 (3.5—3.7) 3.4 (3.3—3.6) 1.3 (1.2—1.4)
28.5 (28.3—28.8)a 29.0 (28.8—29.3) 11.3 (11.2—11.5)a 5.7 (5.6—5.9)a 9.8 (9.6—9.9)a 5.1 (5.0—5.2)a 0.2 (0.2—0.2)a 1.2 (1.1—1.3)a 0.5 (0.4—0.5)a 2.4 (2.3—2.5)a 0.1 (0.1—0.1)a 2.1 (2.1—2.2)a 0.0 (0.0—0.0)a
15.3 (15.1—15.5)a,b 13.3 (13.1—13.5)a,b 6.4 (6.2—6.5)a,b 10.1 (9.9—10.3)a,b 7.1 (6.9—7.2)a,b 2.7 (2.6—2.8)a,b 0.4 (0.4—0.5)a,b 1.3 (1.2—1.4)a 1.8 (1.8—1.9)a,b 0.9 (0.8—0.9)a,b 0.1 (0.1—0.2)a 1.8 (1.7—1.9)a,b 0.1 (0.0—0.1)a,b
* a b
Inter-quartile range, P < 0.05 Kruskal—Wallis H-test. Significantly different from epilepsy group. Significantly different from migraine group.
This modeling technique is an extension of the standard logistic regression that allows direct comparison of the parameter estimates across the response categories because of the model’s ability to generate pooled covariance matrix and smaller standard errors (Biesheuvel et al., 2008; Stokes et al., 1997). Parameter estimates derived from polytomous logistic regression are similar to those derived from two separate dichotomous logistic models (epilepsy vs. LEF and migraine vs. LEF); however, parameter estimates from separate logistic models cannot be directly compared. Multicollinearity among covariables was absent based on the deviations of the regression coefficients and their standard errors in the fitted univariate and multivariable models (Darlington et al., 1998). Variables were entered
simultaneously in the model. The adjusted odds ratios and 95% CIs are reported.
Results From 2000 through 2011, 64,188 PWE, 121,990 PWM, and 89,808 PWLF met inclusion criteria for the study (Table 1). Median per capita encounters/year was 1.2 and 2.1 times more frequent in PWE than PWM and PWLF, respectively. PWE had significantly higher proportions of comorbid conditions than PWM and PWLF. Polymorbidity (having ≥6 comorbid conditions) was 1.86 and 2.69 times more frequent in PWE than PWM and PWLF, respectively.
308
Table 2
Associations of somatic and psychiatric/neurodevelopmental comorbidities in persons with epilepsy, migraine and lower extremity fracture.
Characteristics
Epilepsy (n = 64,188) Frequency (%)
Somatic or psychiatric/neurodevelopmental comorbidity Both 36,271 (56.5) 5.44 Psych/neurodev. only 3,563 (5.6) 3.70 Somatic only 18,052 (28.1) 2.11 None 6,302 (9.8) 1.00 Mortality status Deceased 11,989 (18.7) 1.80 Alive 1.00 52,199 (81.3) Age group Mean age (s.d.)* 41.6 ± 22.5 0—5 5,641 (8.5) 2.38 6—12 3,252 (5.1) 1.17 13—18 4,386 (6.8) 1.45 19—34 12,867(20.1) 1.70 35—64 28,718 (44.7) 1.32 65 and older 9,504 (14.8) 1.00 Race/ethnicity Black 23,134 (36.1) 1.47 Hispanic 574 (0.9) 1.27 Other 856 (1.3) 1.18 White 39,624 (61.7) 1.00 Gender Male 31,256 (48.7) 1.13 Female 32,932 (51.3) 1.00 Payer Uninsured 6,516 (10.2) 0.76 Medicare 21,024 (32.8) 1.79 Medicaid 18,551 (28.9) 1.76 Commercial 18,097 (28.2) 1.00 *
Adjusted for all covariables in the model. P < 0.05.
Lower extremity fracture (n = 89,808) Frequency (%) Reference
(5.25—5.63) (3.51—3.90) (2.04—2.18)
49,160 (40.3) 5,758 (4.7) 45,326 (37.2) 21,746 (17.8)
2.49 (2.42—2.55) 1.51 (1.44—1.58) 1.89 (1.84—1.94) 1.00
24,988 (27.8) 4,279 (4.8) 33,482 (37.3) 27,059 (30.1)
1.00
(1.74—1.86)
4,379 (3.6) 117,611 96.4)
0.54 (0.51—0.56) 1.00
8,001 (8.9) 81,807 (91.1)
1.00
35.9 ± 15.9 1,901 (1.6) 5,931 (4.9) 13,120 (10.8) 45,345 (37.1) 51,086 (41.9) 4,607 (3.7)
0.87 2.27 4.20 5.84 2.92 1.00
(0.81—0.93) (2.15—2.40) (3.98—4.42) (5.58—6.12) (2.80—3.04)
38.8 ± 22.7 6,476 (7.2) 7,834 (8.7) 8,477 (9.4) 18,848 (21.0) 36,979 (41.2) 11,194 (12.5)
(1.43—1.51) (1.14—1.42) (1.08—1.29)
32,673 (26.7) 1,047 (0.9) 2,050 (1.7) 86,220 (70.7)
0.90 (0.88—0.92) 0.92 (0.86—0.99) 0.80 (0.74—0.88) 1.00
25,206 (28.1) 1,188 (1.3) 2,023 (2.3) 61,391 (68.3)
1.00
(1.11—1.15)
25,300 (20.7) 96,690 (79.3)
0.28 (0.28—0.29) 1.00
44,440 (49.5) 43,368 (50.5)
1.00
(0.73—0.78) (1.74—1.85) (1.71—1.81)
19,212 15,720 26,128 60,930
0.77 (0.75—0.79) 0.67 (0.65—0.69) 0.94 (0.92—0.96) 1.00
16,280 18,647 16,641 38,240
1.00
(2.25—2.52) (1.10—1.24) (1.37—1.53) (1.62—1.77) (1.27—1.38)
(15.7) (12.9) (21.4) (50.0)
(18.1) (20.8) (18.5) (42.6)
1.00
A.W. Selassie et al.
a
Migraine (n = 121,990) Frequency (%) ORa (95% CI)
ORa (95% CI)
Epilepsy beyond seizure Table 1 shows the distribution of the various comorbid conditions among PWE, PWM and PWLF. Of 35 conditions initially identified, four were dropped for very low counts (cysticercosis, onchocerciasis/toxocariasis, celiac disease, and somatoform disorders). The remaining conditions were grouped into two major categories: 18 somatic and 13 psychiatric/neurodevelopmental conditions. The most common somatic comorbidity was cardiovascular disease (CVD), a group that includes hypertension, ischemic heart disease and diseases of pulmonary circulation. Depression and anxiety were the most prevalent psychiatric comorbidities. PWE had significantly higher proportions of somatic and psychiatric/neurodevelopmental conditions than PWM; PWM had significantly higher proportions than PWLF. Table 2 shows the frequency and multivariable adjusted odds ratio (OR) of clinical and demographic risk characteristics in PWE and PWM compared with PWLF. The odds of both somatic and psychiatric/neurodevelopmental comorbidities was significantly higher in PWE than PWLF (OR = 5.44; 95% Confidence Interval [CI] = 5.25—5.63). The corresponding OR in PWM compared to PWLF was 2.49 (95% CI = 2.42—2.55). The absolute risk increase (ARI) of having both somatic and psychiatric/neurodevelopmental comorbidities in PWE compared to PWM is 18.5% [((5.44 − 2.49)/2.49) × 100]. The odds of psychiatric/neurodevelopmental comorbidities without a somatic comorbidity were 3.70 (95% CI = 3.51—3.90) and 1.51 (95% CI = 1.44—1.58) in PWE and PWM respectively, with an ARI of 145.0% in PWE. Although somatic comorbidities without psychiatric/neurodevelopmental comorbidity among PWE were significantly higher than in PWM, the gap was narrower (ARI = 11.6%). Compared to PWLF, the odds of mortality from any cause in PWE was 1.80 (95% CI = 1.74—1.86) vs 0.54 (95% CI = 0.51—0.56) among PWM, resulting in an ARI of 233.3% in PWE compared to PWM. There were important differences regarding the association of epilepsy and migraine with demographic characteristics, specifically in age, race/ethnicity, and payer status. PWE were 1.47 (95% CI = 1.43—1.51) times more likely to be black, and 1.79 and 1.76 times more likely to have coverage by Medicare and Medicaid, respectively, compared to PWLF. The corresponding ratios were inversely associated in PWM. Compared with PWLF, PWE were 2.4 times more likely to be between ages 0 and 5 compared to PWLF and PWM were 5.8 times more likely to be between ages 19 and 34. Among the 18 conditions listed as somatic comorbidities, the adjusted odds ratios were significantly higher in PWE than PWM except for intestinal problems, gastric reflux and peptic ulcer (Table 3). Stroke showed the strongest association with epilepsy, with PWE 4.20 (95% CI = 4.06—4.34) times more likely than PWLF to have been diagnosed with stroke. The odds of stroke in PWM were 1.65 (95% CI = 1.59—1.71) resulting in an ARI in PWE of 154.6% relative to PWM. Other neurological diseases with higher odds of occurrence included multiple sclerosis (MS) (OR 2.25; 95% CI = 1.98—2.54) and Parkinson’s disease (PD) (OR 2.45; 95% CI = 1.98—2.54). The corresponding odds in PWM were 1.68 (95% CI = 1.50—1.89) and 1.01 (95% CI = 0.88—1.16) for MS and PD respectively. The ARI of MS and PD among PWE compared to PWM is 33.9 and 142.6% for, respectively. Stark differences between PWE and PWM occurred in intellectual disability (ID), cognitive dysfunction, and autism
309 spectrum disorders (ASD). Compared to PWM, the ARI in PWE was 2791.7% higher for ID, 1487% higher for cognitive dysfunction, and 1730.6% higher for ASD. In all of the remaining psychiatric/neurodevelopmental comorbidities, the odds ratios were significantly higher in PWE compared to PWM, most notably in Alzheimer’s and schizophrenia with ARI of 388.3% and 259.0% in PWE, respectively. The strengths of the observed associations of epilepsy and migraine with the specific comorbid conditions are summarized in Table 4. The adjusted association of epilepsy with ID, cognitive dysfunction, and ASD was profoundly strong (OR ≥ 10). The association of epilepsy with psychiatric comorbidities and seven somatic comorbidities (stroke, MS, Parkinson’s, vision loss, HIV/AIDS, migraine, and nutritional deficiencies) was very strong (OR 2.0—9.99). The association of epilepsy with diabetes remained in the weak range while osteoporosis was inversely associated. The remaining conditions were strongly associated with epilepsy (OR 1.5—1.99). In regard to migraine, five somatic (diabetes, TBI, nutritional deficiencies, osteoporosis, and Parkinson’s disease) and four psychiatric/neurodevelopmental comorbidities (schizophrenia, alcoholism, Alzheimer’s dementia, and ID) showed neutral or inverse association. Six somatic (intestinal problems, asthma, gastric reflux, stroke, peptic ulcer and MS) and six psychiatric/neurodevelopmental comorbidities (depression, anxiety, personality disorder, psychosis, cognitive dysfunction and ADHD) were strongly associated with migraine while the rest showed weak associations.
Discussion This large, population-based, retrospective cohort study examined the association of somatic and psychiatric/neurodevelopmental comorbidities in PWE and PWM compared with PWLF. Somatic comorbidities occurred in 84.6% of PWE and 77.5% of PWM; psychiatric/neurodevelopmental comorbidities occurred in 62.1% of PWE and 45% of PWM. PWE had strongly elevated odds of having both somatic and psychiatric/neurodevelopmental comorbidities compared to PWM and PWLF. Furthermore, PWE had 80% higher mortality from all causes than PWLF and 126% higher than PWM. There was 18-fold increase in ASD, 16-fold increase in cognitive dysfunction, and 27-fold increase in ID among PWE compared to PWM. Epilepsy was significantly positively associated with all 31 specific comorbid conditions while migraine was significantly positively associated with 20 of 30 conditions. Unlike studies that reported the associations of epilepsy with comorbid conditions from self-report or subsets of population groups, our study evaluated multivariable adjusted associations of specific comorbidities in PWE compared with PWM and PWLF using clinically ascertained diagnoses. The study findings are consonant with the research priorities of the IOM report (England et al., 2012). To our knowledge, this is the first study of a statewide population with large minority and socioeconomically disadvantaged groups to examine the relationships of specific comorbidities with epilepsy compared with another neurological disorder and healthy controls.
310
A.W. Selassie et al.
Table 3 Association of common comorbidities in persons with epilepsy and migraine compared to lower extremity fracture controls. Comorbid condition
Epilepsy Adjusteda OR
Somatic comorbidities Cardiovascular disease Intestinal problems Asthma/pulmonary Gastric reflux Anemia Stroke Diabetes Peptic ulcer Traumatic brain injury Nutritional deficiency GI bleed Osteoporosis Vision loss Hearing loss Parkinson’s disease HIV/AIDS Multiple sclerosis Migraine
1.97 1.54 1.61 1.59 1.91 4.20 1.29 1.57 1.59 2.27 1.80 0.87 2.47 1.79 2.45 2.17 2.25 3.37
Psychiatric and neurodevelopmental Depression 2.12 Anxiety 2.29 Psychoses 3.17 Schizophrenia 3.15 Personality disorder 3.61 Alcoholism 1.77 Drug abuse 2.37 Suicidal ideation/attempt 2.95 Alzheimer’s dementia 2.19 Intellectual disability 12.88 Cognitive dysfunction 28.09 ADHD 2.31 Autism spectrum disorder 22.15 a
Migraine Adjusteda (95% CI)
OR
(95% CI)
(1.92—2.03) (1.50—1.58) (1.57—1.65) (1.55—1.63) (1.86—1.96) (4.06—4.34) (1.26—1.33) (1.52—1.61) (1.54—1.67) (2.18—2.37) (1.72—1.87) (0.83—0.91) (2.28—2.68) (1.67—1.93) (2.20—2.73) (1.91—2.45) (1.98—2.54) (3.23—3.52)
1.45 1.59 1.53 1.78 1.18 1.65 0.89 1.78 0.81 0.83 1.42 0.65 1.33 1.42 1.01 1.13 1.68
(1.42—1.49) (1.56—1.63) (1.50—1.56) (1.74—1.81) (1.15—1.21) (1.59—1.71) (0.86—0.91) (1.73—1.82) (0.78—0.83) (0.79—0.88) (1.36—1.48) (0.62—0.69) (1.22—1.45) (1.31—1.53) (0.88—1.16) (0.98—1.29) (1.50—1.89) —
(2.06—2.17) (2.23—2.35) (3.06—3.28) (2.94—3.37) (3.34—3.89) (1.71—1.82) (2.29—2.45) (2.81—3.10) (2.05—2.33) (11.59—14.30) (23.33—33.82) (2.16—2.47) (16.77—29.26)
1.62 1.87 1.51 0.85 1.67 0.60 1.19 1.44 0.61 0.48 1.77 1.56 1.21
(1.59—1.66) (1.82—1.91) (1.46—1.56) (0.78—0.92) (1.55—1.80) (0.58—0.62) (1.15—1.23) (1.38—1.52) (0.56—0.68) (0.41—0.57) (1.38—2.26) (1.46—1.67) (0.81—1.80)
Adjusted for age, race, gender, insurance status, and mortality status and number of comorbid conditions.
Somatic comorbidities in epilepsy Cardiovascular diseases (CVD) were the most prevalent somatic comorbidity overall (61.7%, 95% CI = 61.3—62.1) and strongly associated with epilepsy (OR = 1.97; 95% CI = 1.92—2.03), similar to the association seen in a US study of adults with seizures (Strine et al., 2005). Hypertension was the dominant comorbidity in the CVD category in our dataset contributing for the high prevalence (data not shown). Severe hypertension has been implicated in new onset seizure (Hesdorffer et al., 1996). Of the somatic comorbidities, stroke (27.1% of PWE) showed the strongest association with epilepsy (OR = 4.20; 95% CI = 4.06—4.34). This finding concurs with prior studies (Gaitatzis et al., 2012; Nuyen et al., 2006; Tellez-Zenteno et al., 2005). Antecedent stroke has been well-established as a cause of epilepsy (Benbir et al., 2006; Hussain et al., 2006; Mousali et al., 2009). However, epilepsy as a
antecedent risk factor of stroke remains tenuous because the common risk factors of stroke such as hypertension, diabetes, and dyslipidemia are prevalent in PWE making the association with pre-existing epilepsy uncertain (Cleary et al., 2004; Seidenberg et al., 2009; Shinton et al., 1987). Migraine (12.1% of PWE) was the second most strongly associated comorbidity with epilepsy (OR = 3.37; 95% CI = 3.23—3.52). Our study showed stronger association between migraine and epilepsy than other population-based studies (Ottman et al., 2011; Strine et al., 2005). Migraine shares many of the features of epilepsy, which obscures discrimination between the two conditions and indicates possible shared genetic and physiological mechanisms and bidirectional associations between the two (Kasteleijn-Nolst Trenite and Parisi, 2012; Seidenberg et al., 2009). PD and MS were less prevalent but strongly associated with epilepsy. The association of PD (OR = 2.45; 95% CI = 1.98—2.54) was similar to that seen in a
Epilepsy beyond seizure
311
Table 4 Summary of associations of epilepsy and migraine with specific comorbid conditions by their the strengths of association. Adjustment
Profoundly strong OR ≥ 10
Very strong 2.0 ≤ OR < 10.0
Strong 1.5 ≤ OR < 2.0
Weak 1.0 < OR < 1.5
Neutral or inverse OR ≤ 1.0
Epilepsy
Age, race, gender, insurance status, mortality status
Intellectual disability, cognitive dysfunction, ASD
Stroke, nutritional deficiency, vision loss, Parkinson’s disease, HIV/AIDS, MS, migraine, depression, anxiety, psychoses, schizophrenia, personality disorder, drug abuse, suicidal ideation/attempt, Alzheimer’s, ADHD
Cardiovascular, intestinal, asthma, gastric reflux, anemia, peptic ulcer, TBI, GI bleed, hearing loss, alcoholism
Diabetes
Osteoporosis
Migraine
Age, race, gender, insurance status, mortality status
Intestinal, asthma, gastric reflux, stroke, peptic ulcer, ms, depression, anxiety, personality disorder psychoses, cognitive dysfunction, ADHD
Cardiovascular, anemia, vision loss, hearing loss, GI bleed, HIV/AIDS, drug abuse, suicidal ideation, ASD
Diabetes, TBI, nutritional deficiency, osteoporosis, Parkinson’s, schizophrenia, alcoholism, Alzheimer’s, intellectual disability
population-based UK study (Gaitatzis et al., 2004a, 2012). However, the association of PD may be confounded by stroke and dementia that both share common underlying risk factors. We found a significant positive association of MS with epilepsy (OR = 2.25; 95% CI = 1.91—2.45) while others reported associations ranging from non-significant to a 3-fold increase in reference to the general population (Nicoletti et al., 2003; Nyquist et al., 2002). The underlying pathophysiological link between epilepsy and MS is speculated to involve cortical and subcortical demyelination and inflammation (Kelley and Rodriguez, 2009). The prevalence of diabetes in PWE (24.8%; 95% CI = 24.4—25.1) was significantly higher than the prevalence in the comparison groups. The adjusted association between epilepsy and diabetes in this study (OR = 1.29; 95% CI = 1.26—1.33) approximates the findings reported by Gaitatzis et al. (2004a).
Psychiatric and mental health comorbidities in epilepsy Depression was the most prevalent psychiatric comorbidity (31.3%) and was strongly associated with epilepsy (OR = 2.12;
95% CI = 2.06—2.17). Other studies reported similar prevalence and association (Hesdorffer et al., 2006; Rai et al., 2012). Although social stigma and unemployment may contribute to the high prevalence of depression among PWE, recent evidence suggest shared underlying neurochemical abnormalities may contribute to the bidirectional association between epilepsy and depression (Jensen, 2011; Seidenberg et al., 2009). Anxiety was the second most prevalent psychiatric comorbidity (29.1% in PWE) and was strongly associated with epilepsy (OR = 2.29; 95% CI = 2.23—2.35). The lifetime prevalence of anxiety disorders among PWE varies between 10 and 25% (Gaitatzis et al., 2004b) although higher rates have been reported with refractory epilepsy. The OR reported in our study is comparable to other population-based studies (Hesdorffer et al., 2012). The prevalence in PWM was similar (29.0%) despite significantly weaker association compared with epilepsy. Alcoholism and drug abuse have been identified as common mental health problems in PWE (Hillbom et al., 2003) as was evident in our cohort. Alcoholism was identified in 18.1% of PWE and was strongly associated with epilepsy (OR = 1.77; 95% CI = 1.71—1.82) consistent with prior reports
312 (Chan, 1985; Hauser et al., 1988). However, alcoholism often precedes epilepsy; the risk of seizure with daily alcohol consumption is reported to increase from 3-fold to 20-fold (Ng et al., 1988). In our study, drug abuse was identified in 16.9% of PWE and was strongly associated with epilepsy (OR = 2.37; 95% CI = 2.29—2.45). There is limited information in the literature regarding the prevalence of drug abuse in PWE. The very strong association of schizophrenia with epilepsy (OR = 3.61; 95% CI = 3.34—3.89) despite its low prevalence (5.6% in PWE) indicate the importance of schizophrenia in PWE, suggesting the need for coordinated patient care with a neuropsychiatry team. Previously reported prevalence rates of schizophrenia range from 4.3 to 18% (Gaitatzis et al., 2004b). A comparable longitudinal population-based study from Denmark reported a relative risk of 2.48 in PWE (95% CI = 2.20—2.80) (Qin et al., 2005). The mechanism by which epilepsy and schizophrenia are associated may relate to cortical dysgenesis or diffuse brain lesions that underlie both conditions or confounding sociobehavioral factors (Sachdev, 1998). Similarly, despite the low prevalence (4.3% in PWE), personality disorder was very strongly associated with epilepsy (OR = 3.61; 95% CI = 3.34—3.89) in our study. The prevalence of personality disorder among PWE is reported to range from 5 to 18% depending on cortical lesion and type of epilepsy with higher rates in temporal lobe epilepsy (Devinsky, 2003). Suicidal ideation and attempt among PWE has been evaluated extensively. In our study, the prevalence of suicidal ideation in PWE was 8.8% and the observed association with epilepsy was strong (OR = 2.95; 95% CI = 2.81—3.10). This result is comparable with a meta-analysis of 23 studies (Jones et al., 2003) and a study that examined suicidal ideation as function of onset of epilepsy (Hesdorffer et al., 2012). Risk of suicidal ideation among PWE is significantly increased in the presence of other psychiatric comorbidities, especially major depression and anxiety disorders (Gandy et al., 2013). Suicidal ideation may also be associated with unprovoked seizures in children independent of depression (Hesdorffer et al., 2006). The prevalence of Alzheimer’s dementia (AD) was 5.0% in PWE with strong association (OR = 2.19; 95% CI = 2.05—2.33). A health interview survey found an unadjusted prevalence ratios ranging from of 4.3 to 8.1 times the prevalence in the general population, although a prospective study showed no significant increased risk in PWE (Gaitatzis et al., 2004a; Scarmeas et al., 2009; Tellez-Zenteno et al., 2005). The attributed underlying pathophysiological phenomenon between epilepsy and AD is a progressive neurodegenerative process that begins with cognitive decline (Amatniek et al., 2006; Palop and Mucke, 2009).
Neurodevelopmental comorbidities in epilepsy While the prevalence of cognitive dysfunction was only 3.6% in PWE, the association was profoundly strong (OR = 28.09; 95% CI = 23.33—33.82). Cognitive dysfunction is predominately diagnosed in young children with epilepsy, although memory deficits are also common in temporal lobe epilepsy in adults (Jensen, 2011). In our cohort of PWE, 55% of cognitive dysfunction was diagnosed in age ≤5, 26% in ages 6—18, and 9% in ages 19—34. In a community-based study,
A.W. Selassie et al. 26.4% of children had evidence of cognitive decline, mostly attributable to epileptic encephalopathies (Berg et al., 2008). Similarly, our study noted a prevalence rate of 7.3% in ID among PWE and a profoundly strong association with epilepsy (OR = 12.88; 95% CI = 11.59—1.40). The prevalence of ID in SC is reported to be 1.13% in the general population (Massey and McDermott, 1996). The prevalence of autism spectrum disorders (ASD) in PWE was low (1.3%) but the association was profoundly strong (OR = 22.15; 95% CI = 16.77-29.26). ASD are a common comorbidity of epilepsy in children and approximately 75% of those with ASD were children and adolescents giving a prevalence of 4.3% for those 18 and under. The prevalence of ASD in the general US population of children with epilepsy is undetermined but a community-based study of childhood-onset epilepsy reported 5% prevalence of ASD (Berg et al., 2011). A UK study found a prevalence of 8.1% (95% CI = 2.2—25.9) in non-institutionalized adults with epilepsy (Rai et al., 2012). The prevalence of attention deficit/hyperactivity disorder (ADHD) was 3.4% and the association of ADHD with epilepsy was strong (OR = 2.31; 95% CI 2.16—2.47). ADHD has been described as one of the commonly co-occurring comorbidities of epilepsy in children (Parisi et al., 2010). The prevalence in our study was 12.8% (95% CI = 11.2—14.4) in children 18 and younger, similar to that seen in descriptive study of 175 children and adolescents with epilepsy (Dunn et al., 2003). ADHD has been associated with increased risk for developing unprovoked seizure (Hesdorffer et al., 2004). Factors that may contribute to the association between epilepsy and ADHD are chronic seizures, repetitive epileptiform discharges, and therapeutic medications (Parisi et al., 2010).
Strengths and limitations Our study has several strengths. First, data come from a legally mandated, multifaceted medical database of all healthcare encounters spanning 12 years. Use of a medical database overcomes the drawback of relying on self-report used in many epidemiological studies of epilepsy. Second, the findings are generalizable because cases and controls are representative of the referent population. Third, case ascertainment relied on ICD-9-CM codes following the ILAE guidelines and listing of routinely associated comorbidities for epidemiological studies (International League Against Epilepsy Commission on Epidemiology and Prognosis, 1993). Fourth, the methodological and analytical approaches of comparing epilepsy, migraine, and LEF as polytomous discrete responses and the very large sample size provided robust estimates and within group comparisons. Fifth, comparison with a similar chronic neurological disease (migraine) allowed estimation of ARI attributable to epilepsy. Despite the aforementioned strengths, there are important limitations worth noting. First, the data do not include PWE who receive healthcare in military and veteran affairs hospitals that may have higher rates of epilepsy. This limitation, however, applies to all public health data systems in the US. Second, analysis relied on administrative data designed for billing third-party providers and a
Epilepsy beyond seizure
313
coding practice responsive to the policies of providers, possibly influencing estimation of the comorbid conditions. Third, with wide variability in skillset and diagnostic resources among the hospitals, the accuracy of the 4th and 5th digit of the diagnosis codes that provide much of the desired diagnostic specificity might be unreliable. Nonetheless, previous validation studies involving medical chart review of randomly selected records conducted for various prior projects using the SC data system found that coding error is very low and unlikely to sway the observed results (data not shown). Fourth, we have not accounted for venue of care. Inpatient hospital encounters may emphasize specific comorbidities more than others, and identify more comorbidities overall due to more extensive workup. However, this bias is likely to affect all population groups equally; those with epilepsy, migraine or LEF who require hospitalization are likely to have more comorbidities compared with those treated and released in the ED. Fifth, we are unable to obtain data on encounters in physician offices and other outpatient settings resulting in a bias towards more complex cases. In summary, this study provides population-based distribution and associations of a set of somatic and psychiatric/neurodevelopmental comorbidities among PWE in comparison with PWM and PWLF. Epilepsy was positively associated with all of the comorbidities examined except for osteoporosis. Further, comorbidities, particularly neurodevelopmental comorbidities, were more strongly associated with epilepsy than with migraine or LEF.
Acknowledgements This study was funded by cooperative agreement DP003251 from the National Center for Chronic Disease Prevention and Health Promotion, the Centers for Disease Control and Prevention. The findings of the study and opinions presented in the manuscript are those of the authors and do not reflect the opinion of the funding agencies. The authors recognize the support of Mr. Chris Finney from the Office of Research and Statistics, South Carolina Budget and control Board, for selecting and organizing the data from multiple administrative data sources.
Appendix A. ICD-9-CM codes for case and control group and comorbid condition definitions ICD-9-CM code(s) Cohort status Case: Epilepsy Control 1: Migraine Control 2: Tibia, Fibula, Ankle fractures Comorbidity ADHD Alcoholism
Alzheimer’s dementia
345.1, 345.4—345.9 346 823, 824
314.0 291.1, 291.2, 291.5, 291.8, 291.9, 303.90, 303.93, 305.00, 305.03, V11.3 331.0
Appendix (Continued ) ICD-9-CM code(s) Anemia Anxiety Asthma/pulmonary disease Autism spectrum disorder Cardiovascular disease Celiac disease Cognitive dysfunction Cysticercoids Depression Diabetes
Drug abuse
Gastric reflux GI bleed Hearing loss HIV/AIDS Intestinal problems Intellectual disability Migraine Multiple sclerosis Nutritional deficiencies Onchocerciasis/toxocariasis Osteoporosis Parkinson’s disease Peptic ulcer Personality disorder Psychoses Schizophrenia Somatoform disorder Stroke Suicidal ideation/attempt Traumatic brain injury Vision loss
280.1—281.9, 285.9 300.0, 300.7 490—496 299.0 401—405, 410—417, 420—429 579.0 315, V40.0 123.1 300.4, 309.0, 309.1, 311 250.00—250.33, 250.40—250.73, 250.90—250.93 292.0, 292.82—292.89, 292.9, 304.00—304.93, 305.20—305.93 530.81 578 389 042, 044 560—569 317—319 346 340 260—269 125.3, 128.0 733.0, V82.81 332 531—535 301 293.8, 297—298.9, 299.10, 299.11 295 300.8 430—438 300.9, V62.84 800, 801, 803, 804, 850—854, 959.01 369
References Amatniek, J.C., Hauser, W.A., DelCastillo-Castaneda, C., Jacobs, D.M., Marder, K., Bell, K., et al., 2006. Incidence and predictors of seizures in patients with Alzheimer’s disease. Epilepsia 47 (5), 867—872, http://dx.doi.org/10.1111/ j. 1528-1167.2006.00554.x. American Medical Association, 2009. International Classification of Diseases, 9th revision, Clinical Modification. Saunders, Chicago, IL. Baca, C.B., Vickrey, B.G., Caplan, R., Vassar, S.D., Berg, A.T., 2011. Psychiatric and medical comorbidity and quality of life outcomes in childhood-onset epilepsy. Pediatrics 128 (6), e1532—e1543, http://dx.doi.org/10.1542/peds.2011-0245. Balottin, U., Chiappedi, M., Rossi, M., Termine, C., Nappi, G., 2011. Childhood and adolescent migraine: a neuropsychiatric disorder? Med. Hypotheses 76 (6), 778—781, http://dx.doi.org/ 10.1016/j.mehy.2011.02.016.
314 Benbir, G., Ince, B., Bozluolcay, M., 2006. The epidemiology of post-stroke epilepsy according to stroke subtypes. Acta Neurol. Scand. 114 (1), 8—12, http://dx.doi.org/10.1111/j. 1600-0404.2006.00642.x. Berg, A.T., Caplan, R., Hesdorffer, D.C., 2011. Psychiatric and neurodevelopmental disorders in childhood-onset epilepsy. Epilepsy Behav. 20 (3), 550—555, http://dx.doi.org/10.1016/ j.yebeh.2010.12.038. Berg, A.T., Langfitt, J.T., Testa, F.M., Levy, S.R., DiMario, F., Westerveld, M., et al., 2008. Global cognitive function in children with epilepsy: a community-based study. Epilepsia 49 (4), 608—614, http://dx.doi.org/10.1111/j. 1528-1167.2007.01461.x. Biesheuvel, C.J., Vergouwe, Y., Steyerberg, E.W., Grobbee, D.E., Moons, K.G., 2008. Polytomous logistic regression analysis could be applied more often in diagnostic research. J. Clin. Epidemiol. 61 (2), 125—134, http://dx.doi.org/10.1016/ j.jclinepi.2007.03.002. Chan, A.W., 1985. Alcoholism and epilepsy. Epilepsia 26 (4), 323—333. Chiappedi, M., Beghi, E., Ferrari-Ginevra, O., Ghezzo, A., Maggioni, E., Mattana, F., et al., 2011. Response to rehabilitation of children and adolescents with epilepsy. Epilepsy Behav. 20 (1), 79—82, http://dx.doi.org/10.1016/j.yebeh.2010.10.029. Cleary, P., Shorvon, S., Tallis, R., 2004. Late-onset seizures as a predictor of subsequent stroke. Lancet 363 (9416), 1184—1186, http://dx.doi.org/10.1016/S0140-6736(04)15946-1. Darlington, G.A., Armtage, P., Colton, T., 1998. Collinearity Encyclopedia of Biostatistics. John Wiley & Sons, Chichester, West Sussex, UK, pp. 788—789. Devinsky, O., 2003. Psychiatric comorbidity in patients with epilepsy: implications for diagnosis and treatment. Epilepsy Behav. 4 (Suppl. 4), S2—S10. Dunn, D.W., Austin, J.K., Harezlak, J., Ambrosius, W.T., 2003. ADHD and epilepsy in childhood. Dev. Med. Child Neurol. 45 (1), 50—54. England, M.J., Liverman, C.T., Schultz, A.M., Strawbridge, L.M., 2012. Epilepsy across the spectrum: promoting health and understanding. A summary of the Institute of Medicine report. Epilepsy Behav. 25 (2), 266—276, http://dx.doi.org/10.1016/ j.yebeh.2012.06.016. Fienberg, S.E., 1989. Sampling Models. In: Press, T.M. (Ed.), The Analysis of Cross-Classified Categorical Data. The Massachusetts Institute of Technology, Cambridge, MA, pp. 29—32. Gaitatzis, A., Carroll, K., Majeed, A., J., W.S., 2004a. The epidemiology of the comorbidity of epilepsy in the general population. Epilepsia 45 (12), 1613—1622, http://dx.doi.org/10.1111/j. 0013-9580.2004.17504.x. Gaitatzis, A., Sisodiya, S.M., Sander, J.W., 2012. The somatic comorbidity of epilepsy: a weighty but often unrecognized burden. Epilepsia 53 (8), 1282—1293, http://dx.doi.org/10.1111/j. 1528-1167.2012.03528.x. Gaitatzis, A., Trimble, M.R., Sander, J.W., 2004b. The psychiatric comorbidity of epilepsy. Acta Neurol. Scand. 110 (4), 207—220, http://dx.doi.org/10.1111/j. 1600-0404.2004.00324.x. Gandy, M., Sharpe, L., Perry, K.N., Miller, L., Thayer, Z., Boserio, J., et al., 2013. Rates of DSM-IV mood, anxiety disorders, and suicidality in Australian adult epilepsy outpatients: a comparison of well-controlled versus refractory epilepsy. Epilepsy Behav. 26 (1), 29—35, http://dx.doi.org/10.1016/j.yebeh.2012.10.023. Hauser, W.A., Ng, S.K., Brust, J.C., 1988. Alcohol, seizures, and epilepsy. Epilepsia 29 (Suppl. 2), S66—S78. Hesdorffer, D.C., Caplan, R., Berg, A.T., 2012. Familial clustering of epilepsy and behavioral disorders: evidence for a shared genetic basis. Epilepsia 53 (2), 301—307, http://dx.doi.org/10.1111/j. 1528-1167.2011.03351.x. Hesdorffer, D.C., Hauser, W.A., Annegers, J.F., Rocca, W.A., 1996. Severe, uncontrolled hypertension and adult-onset seizures: a case-control study in Rochester, Minnesota. Epilepsia 37 (8), 736—741.
A.W. Selassie et al. Hesdorffer, D.C., Hauser, W.A., Olafsson, E., Ludvigsson, P., Kjartansson, O., 2006. Depression and suicide attempt as risk factors for incident unprovoked seizures. Ann. Neurol. 59 (1), 35—41, http://dx.doi.org/10.1002/ana.20685. Hesdorffer, D.C., Ludvigsson, P., Olafsson, E., Gudmundsson, G., Kjartansson, O., Hauser, W.A., 2004. ADHD as a risk factor for incident unprovoked seizures and epilepsy in children. Arch. Gen. Psychiatry 61 (7), 731—736, http://dx.doi.org/10.1001/ archpsyc.61.7.731. Hillbom, M., Pieninkeroinen, I., Leone, M., 2003. Seizures in alcohol-dependent patients: epidemiology, pathophysiology and management. CNS Drugs 17 (14), 1013—1030. Hussain, S.A., Haut, S.R., Lipton, R.B., Derby, C., Markowitz, S.Y., Shinnar, S., 2006. Incidence of epilepsy in a racially diverse, community-dwelling, elderly cohort: results from the Einstein aging study. Epilepsy Res. 71 (2-3), 195—205, http://dx.doi.org/10.1016/j.eplepsyres.2006.06.018. International League Against Epilepsy Commission on Epidemiology and Prognosis, 1993. Guidelines for epidemiologic studies on epilepsy. Commission on epidemiology and prognosis, international league against epilepsy. Epilepsia 34 (4), 592—596. Jensen, F.E., 2011. Epilepsy as a spectrum disorder: implications from novel clinical and basic neuroscience. Epilepsia 52 (Suppl. 1), 1—6, http://dx.doi.org/10.1111/j.1528-1167.2010.02904.x. Jones, J.E., Hermann, B.P., Barry, J.J., Gilliam, F.G., Kanner, A.M., Meador, K.J., 2003. Rates and risk factors for suicide, suicidal ideation, and suicide attempts in chronic epilepsy. Epilepsy Behav. 4 (Suppl. 3), S31—S38. Kasteleijn-Nolst Trenite, D., Parisi, P., 2012. Migraine in the borderland of epilepsy: migralepsy an overlapping syndrome of children and adults? Epilepsia 53 (Suppl. 7), 20—25, http://dx.doi.org/10.1111/j. 1528-1167.2012.03711.x. Kelley, B.J., Rodriguez, M., 2009. Seizures in patients with multiple sclerosis: epidemiology, pathophysiology and management. CNS Drugs 23 (10), 805—815, http://dx.doi.org/ 10.2165/11310900-000000000-00000. Kobau, R., Zahran, H., Thurman, D.J., Zack, M.M., Henry, T.R., Schachter, S.C., et al., 2008. Epilepsy surveillance among adults — 19 States, Behavioral Risk Factor Surveillance System, 2005. MMWR Surveill. Summ. 57 (6), 1—20. Linehan, C., Tellez-Zenteno, J.F., Burneo, J.G., Berg, A.T., 2011. Future directions for epidemiology in epilepsy. Epilepsy Behav. 22 (1), 112—117, http://dx.doi.org/ 10.1016/j.yebeh.2011.06.006. Massey, P.S., McDermott, S., 1996. State-specific rates of mental retardation — United States, 1993. MMWR 45 (3), 61—65. Mousali, Y.M., Obeid, T.H., Al-Gahtani, H.A., Al-Janadi, M.A., Attar, A.A., Dammas, H.O., 2009. Epilepsy in patients 60 years and above. Neurosciences (Riyadh) 14 (1), 53—55. National Institutes of Health, 2012. Epilepsy Research Benchmarks: Progress Report 2007—2012. National Institute of Neurological Disorders and Stroke, Bethesda, MD. Ng, S.K., Hauser, W.A., Brust, J.C., Susser, M., 1988. Alcohol consumption and withdrawal in new-onset seizures. N. Engl. J. Med. 319 (11), 666—673, http://dx.doi.org/10.1056/ NEJM198809153191102. Nicoletti, A., Sofia, V., Biondi, R., Lo Fermo, S., Reggio, E., Patti, F., et al., 2003. Epilepsy and multiple sclerosis in Sicily: a population-based study. Epilepsia 44 (11), 1445—1448. Nuyen, J., Schellevis, F.G., Satariano, W.A., Spreeuwenberg, P.M., Birkner, M.D., van den Bos, G.A., et al., 2006. Comorbidity was associated with neurologic and psychiatric diseases: a general practice-based controlled study. J. Clin. Epidemiol. 59 (12), 1274—1284, http://dx.doi.org/10.1016/j.jclinepi.2006.01.005. Nyquist, P.A., Cascino, G.D., McClelland, R.L., Annegers, J.F., Rodriguez, M., 2002. Incidence of seizures in patients with multiple sclerosis: a population-based study. Mayo Clin. Proc. 77 (9), 910—912, http://dx.doi.org/10.4065/77.9.910.
Epilepsy beyond seizure Ottman, R., Lipton, R.B., Ettinger, A.B., Cramer, J.A., Reed, M.L., Morrison, A., et al., 2011. Comorbidities of epilepsy: results from the Epilepsy Comorbidities and Health (EPIC) survey. Epilepsia 52 (2), 308—315, http://dx.doi.org/10.1111/j. 1528-1167.2010.02927.x. Palop, J.J., Mucke, L., 2009. Epilepsy and cognitive impairments in Alzheimer disease. Arch. Neurol. 66 (4), 435—440, http://dx.doi.org/10.1001/archneurol.2009.15. Parisi, P., Moavero, R., Verrotti, A., Curatolo, P., 2010. Attention deficit hyperactivity disorder in children with epilepsy. Brain Dev. 32 (1), 10—16, http://dx.doi.org/10.1016/ j.braindev.2009.03.005. Qin, P., Xu, H., Laursen, T.M., Vestergaard, M., Mortensen, P.B., 2005. Risk for schizophrenia and schizophrenia-like psychosis among patients with epilepsy: population based cohort study. BMJ 331 (7507), 23, http://dx.doi.org/10.1136/ bmj.38488.462037.8F. Rai, D., Kerr, M.P., McManus, S., Jordanova, V., Lewis, G., Brugha, T.S., 2012. Epilepsy and psychiatric comorbidity: a nationally representative population-based study. Epilepsia 53 (6), 1095—1103, http://dx.doi.org/10.1111/ j. 1528-1167.2012.03500.x. Sachdev, P., 1998. Schizophrenia-like psychosis and epilepsy: the status of the association. Am. J. Psychiatry 155 (3), 325—336. SAS Institute, 2011. Statistical Analytical Software (Version 9.3.1). SAS Institute Inc., Cary, North Carolina. Scarmeas, N., Honig, L.S., Choi, H., Cantero, J., Brandt, J., Blacker, D., et al., 2009. Seizures in Alzheimer disease: who, when, and how common? Arch. Neurol. 66 (8), 992—997, http://dx.doi.org/ 10.1001/archneurol.2009.130. Seidenberg, M., Pulsipher, D.T., Hermann, B., 2009. Association of epilepsy and comorbid conditions. Future Neurol. 4 (5), 663—668, http://dx.doi.org/10.2217/fnl.09.32.
315 Shinton, R.A., Gill, J.S., Zezulka, A.V., Beevers, D.G., 1987. The frequency of epilepsy preceding stroke. Case-control study in 230 patients. Lancet 1 (8523), 11—13. Stokes, M.E., Davis, C.S., Koch, G.G., 1997. Logistic regression II: polytomous response. In: SAS Institute (Ed.), Categorical Data Analysis using the SAS System. 3rd ed. SAS Institute, Cary, NC. Strine, T.W., Kobau, R., Chapman, D.P., Thurman, D.J., Price, P., Balluz, L.S., 2005. Psychological distress, comorbidities, and health behaviors among U.S. adults with seizures: results from the 2002 National Health Interview Survey. Epilepsia 46 (7), 1133—1139, http://dx.doi.org/ 10.1111/j.1528-1167.2005.01605.x. Swartz, K.L., Pratt, L.A., Armenian, H.K., Lee, L.C., Eaton, W.W., 2000. Mental disorders and the incidence of migraine headaches in a community sample: results from the Baltimore Epidemiologic Catchment area follow-up study. Arch. Gen. Psychiatry 57 (10), 945—950. Tellez-Zenteno, J.F., Matijevic, S., Wiebe, S., 2005. Somatic comorbidity of epilepsy in the general population in Canada. Epilepsia 46 (12), 1955—1962, http://dx.doi.org/10.1111/j. 1528-1167.2005.00344.x. Thurman, D.J., Beghi, E., Begley, C.E., Berg, A.T., Buchhalter, J.R., Ding, D., et al., 2011. Standards for epidemiologic studies and surveillance of epilepsy. Epilepsia 52 (Suppl. 7), 2—26, http://dx.doi.org/10.1111/j.1528-1167.2011.03121.x. Weis, M.A., Bradberry, C., Carter, L.P., Ferguson, J., Kozareva, D., 2006. An exploration of human services system contacts prior to suicide in South Carolina: an expansion of the South Carolina Violent Death Reporting System. Inj. Prev. 12 (Suppl. 2), ii17—ii21, http://dx.doi.org/10.1136/ip.2006.012427.
