behavioral disorders in pediatric epilepsy: unmet psychiatric need

Behavioral Disorders in Pediatric Epilepsy: UnmetPsychiatric Need

*Derek Ott, *Prabha Siddarth, ‡Suresh Gurbani, †Susan Koh, †Anne Tournay,†W. Donald Shields, and *Rochelle Caplan

Departments of *Psychiatry and †Neurology and Pediatrics, University of California at Los Angeles, Los Angeles; and‡Department of Pediatrics, University of California at Irvine, Irvine, California, U.S.A.

Summary: Purpose: This study examined the relation be-tween psychiatric diagnosis and mental health services in chil-dren with epilepsy and the associated demographic, cognitive,linguistic, behavioral, and seizure-related variables.

Methods: The Kiddie Schedule for Affective Disorders andSchizophrenia (K-SADS), the Child Behavior Checklist, theTest of Language Development, and the Wechsler IntelligenceScale for Children–Revised (WISC-R) were administered to114 children, aged 5 to 16 years, with either complex partialseizures (CPS) or primary generalized with absence (PGE,petit mal). A Diagnostic and Statistical Manual of Mental Dis-orders (DSM-IV) diagnosis and information regarding mentalhealth services were derived from the K-SADS.

Results: Although ∼60% of the subjects had a DSM-IV psy-

chiatric diagnosis, >60% received no mental health treatment.Absence of mental health care was associated with youngerage, less parental education, limited number of antiepilepticdrugs (AEDs; i.e., one or none), and higher verbal IQ. In ad-dition, children with PGE and a single psychiatric diagnosiswere less likely to have a history of mental health treatment.

Conclusions: This is the first study to demonstrate unmetmental health need in a large sample of children with CPS andPGE. The study’s findings suggest that parents and cliniciansshould be aware of the mental health needs of children withepilepsy, particularly if they have one or more of the identifiedrisk factors. Key Words: Psychopathology—Mental healthtreatment—Epilepsy—Child.

For more than three decades, investigators have docu-mented psychopathology associated with pediatric epi-lepsy. In one of the original, community-based studies,29% of children with uncomplicated epilepsy had ahigher incidence of psychiatric disturbances relative to12% of children with chronic, nonneurologic illnesses,and to 6.6% in the general population (1,2). More re-cently, in studies with children recruited from commu-nity and university-based pediatric neurology clinics, thepresence of behavioral disturbances was demonstrated in21–32% of children with epilepsy by using the ChildBehavior Checklist (3–5), in 23–26% with the Child De-pression Inventory (5,6), in 48% with the Rutter scale(7), and in 55–60% with the Kiddie Schedule for Affec-tive Disorders and Schizophrenia (K-SADS) (8,9). Al-though the wide range of psychopathology might reflectthe use of different rating and diagnostic instruments, anoverall rate of 21 to 60% in these studies represents an

increased risk of �3–6 times, as compared with the gen-eral population.

The pervasive and enduring impact of epilepsy on be-havior in children is further emphasized by the work ofAustin et al. (3) and Dunn et al. (4). Identification ofbehavioral problems in children 6 months before the firstrecognized seizure, especially in those with subsequentseizures, suggests that psychopathology may be inte-grally linked to epilepsy. Considering rates of psychopa-thology in adults with epilepsy as high as 80% (10–12),these findings imply that behavioral difficulties mightdevelop in early childhood and persist into adulthood inpatients with epilepsy.

In terms of the type of psychopathology, several in-vestigations have identified a wide variety of psychiatricsymptoms including anxious (6,8), depressive (5,6,8,13),disruptive (8,14,15), and psychotic symptom (16). In ad-dition, children and adolescents with epilepsy are at in-creased risk for suicidal ideation and suicidal attempts(6,9,17,18). These findings underscore the severity of thepsychopathology and the mental health needs of childrenwith epilepsy.

Nevertheless, only a few studies have examined

Accepted November 10, 2002.Address correspondence and reprint requests to Dr. D. Ott at Neu-

ropsychiatric Institute, Rm. 58-242B, 760 Westwood Plaza, Los An-geles, CA 90024, U.S.A. E-mail: [email protected]

Epilepsia, 44(4):591–597, 2003Blackwell Publishing, Inc.© 2003 International League Against Epilepsy

591

whether children with epilepsy receive mental health ser-vices. Based on a review of records and parental inter-views of 44 children with epilepsy, Ettinger et al. (6)found rates of 26% and 16% of depression and anxiety,respectively. However, none of these children had beenpreviously identified or treated for these psychiatricsymptoms. Similar findings were demonstrated by Dunnet al. (5), who suggested that misinterpretation of psy-chiatric symptoms as a manifestation of seizures or a sideeffect of antiepileptic drugs (AEDs) might be responsiblefor inadequate psychiatric assessment and treatment inchildhood epilepsy (19). These findings highlight the im-portance of determining whether children with epilepsyreceive appropriate mental health care.

There also have been no studies on the relation be-tween demographic factors and mental health treatmentin childhood epilepsy. Demographic factors, however,have generally been inconsistent predictors of psychopa-thology in this population. Some studies have identifiedmore boys with behavioral problems (20–22), particu-larly disruptive disorders (21,23). Other studies report nogender differences (5,6,24–27) or greater risk in girls(13,28). Similar conflicting findings have been demon-strated for chronologic age (5,6,25,26,29). Other demo-graphic factors associated with increased risk forbehavioral problems include lower socioeconomic status(SES) (5,13,21,30–32), less parental education (33), andincreased family stress (25,28).

The association of seizure-related variables (i.e., sei-zure type, seizure control, EEG lateralization, AEDs, ageat onset, duration of epilepsy) with psychopathology alsohas been extensively examined in childhood epilepsy(for review, see 34). In terms of seizure type, childrenwith complex partial seizures (CPS) historically weredescribed as antisocial and aggressive (1,35), whereasthose with primary generalized seizures with absence(PGE) were characterized as neurotic (35). However,more recent investigations have demonstrated that chil-dren with PGE and CPS have similar rates and types ofpsychopathology (i.e., disruptive, affective, and anxietydisorders) (8,9,36).

Regarding other seizure-related variables, poor seizurecontrol (3,4,7,8,21,23,29,37–39), a left temporal focus(16,20,23,40), and treatment with AEDs, especially inhigh doses or in combination (34,41), are associated withbehavioral disturbances. Additional predictors of psy-chopathology in this population include both early age atonset and longer duration of epilepsy (21,38,40,42,43).

As highlighted by several studies, cognitive impair-ment is another important correlate of psychopathologyin children with epilepsy (1,2,23). Camfield et al. (24)demonstrated that the severity of neuropsychologicaldeficits in children with epilepsy is related to increasedpsychopathology in these children. More recently, IQ hasbeen consistently associated with the presence or ab-

sence of a psychiatric diagnosis (8), the severity of be-havioral deviance (9), and the severity of thoughtdisorder (42,44–46). Last, cognitive impairment is asso-ciated with poor seizure control (47–49), which, as dis-cussed previously, is linked to increased risk forpsychopathology (21,24,34,50).

The study presented here determined whether childrenwith CPS and PGE with psychopathology received men-tal health services. It also examined the demographic,cognitive, linguistic, behavioral, and seizure-related vari-ables associated with both mental health services andpsychopathology in these two groups of children withepilepsy. We predicted that most children who meet cri-teria for a psychiatric diagnosis would not be receivingmental health interventions. Based on the previously re-viewed studies, we posited that male gender, lower SES,presence of various seizure-related variables (i.e., type ofseizure disorder, poor seizure control, AED polytherapy,early age at onset, increased duration of illness), the pres-ence of cognitive and linguistic deficits, as well as in-creased psychiatric morbidity, would be associated withmental health care.

METHODS

SubjectsThis study is part of a series of studies comparing

social, communication and psychopathologic factors ina large sample of children (N � 114) aged 5 to 16 yearswith average IQ scores and either CPS or PGE (petitmal). Table 1 presents demographic, cognitive, seizure-related (i.e., seizure control, age at onset, duration ofepilepsy), and psychiatric variables of the children in thestudy. We determined SES by using the Hollingshead IIfactor index (51), based on parental occupational andeducational status. Information regarding seizure historywas obtained from the parents, as well as from the neu-rologic records. Seizure control was defined as the ab-sence of seizures within the last year, and duration ofillness, as the time from onset of seizures to the child’sparticipation in the study.

To be included in the study, a child had to have adiagnosis of CPS or PGE based on clinical history andEEG findings, as defined by the International Classifi-cation of Epilepsy (52). As defined by the Commission,we also included patients with a clinical history of CPS,but no EEG evidence for focal epileptic activity. In ad-dition to a clinical history suggestive of PGE, all PGEpatients had EEG evidence for three-per-second spikeand wave. Three patients with PGE had a history ofgeneralized tonic–clonic seizures in addition to their ab-sence seizures. A pediatric neurologist (S.G., A.T.,W.D.S.) reviewed the neurologic diagnosis and EEG ofeach child. Whenever a discrepancy emerged regardingeither diagnosis or location of epileptic foci, the childwas excluded from the study.

D. OTT ET AL.592

Epilepsia, Vol. 44, No. 4, 2003

We also excluded children with a mixed seizure dis-order, an underlying neurologic disorder, a metabolicdisorder, a hearing disorder, or epilepsy surgery (past orproposed). In addition, only children with native fluencyin American English or mental age greater than 6 wereincluded. We recruited 114 children with epilepsy, 55%from tertiary (i.e., university-based pediatric neurologyservices) and 45% from community sources (public andprivate pediatric neurology clinics, the Los Angeles andSan Diego branches of the Epilepsy Foundation ofAmerica). The primary pediatric neurologist identifiedchildren who might meet the study’s inclusion criteriaand referred the parents to the project irrespective of thepresence or absence of a psychiatric history.

ProceduresThis study was performed in accordance with the poli-

cies of the Human Subjects Protection Committees of theUniversity of California, Los Angeles, and of the South-ern California Kaiser Permanente. We obtained informedconsent from the parents and assents from the children.

Kiddie Schedule for Affective Disordersand Schizophrenia

The Schedule for Affective Disorders and Schizophre-nia for School-Age Children–Epidemiologic version (K-

SADS) (53) was administered separately to each childand parent by R.C. or a research assistant trained in theadministration of the interview. Because the child or par-ent often talks about the child’s seizures during the in-terview, these interviewers were not blinded with regardto the child’s seizure disorder (i.e., presence or absence,type). The second clinician reviewed videotapes of thechild interviews and audiotapes of the parent interviews,and a consensus DSM-IV (54) diagnosis was reached. Ifa diagnostic consensus was not reached, the child wasexcluded from the study.

Given the large number of diagnoses relative to thenumber of subjects in each diagnostic group, we groupedthe diagnoses as follows: “affective/anxiety” disordersincluded any mood or anxiety disorder, and “disruptive”disorders included attention-deficit disorder, opposi-tional defiant disorder, and conduct disorder. Childrenwith a “comorbid” diagnosis had both “affective/anxiety” and “disruptive” disorders. We obtained infor-mation on mental health treatment (i.e., any contact witha mental health professional, including therapy and psy-chotropic medication treatment) from the parent and/orchild K-SADS interview. In this article, mental healthtreatment refers to both past and current mental healthtreatment.

TABLE 1. Demographic, seizure-related, and cognitive characteristicsof patients

All CPS PGE

DemographicN 114 62 52

Age average (yrs) (SD) 10.5 (3.0) 11.0 (3.1) 9.9 (2.8)Gender (%)

Male 47 48 40Female 53 52 60

EthnicityAsian 5.3 1.6 9.6Anglo 52.6 56.5 48.1Afro-American 7.9 8.1 7.7Hispanic 20.2 25.8 13.5Other 14.0 8.1 21.2

Socioeconomic status (%)Low (III IV, V) 82.5 77.4 88.5High (I, II) 17.5 22.6 11.5

Parental education max < high school (%) 46.0 43.5 49.0Referral source (%)

Tertiary 55.3 69.3 38.5Community 44.7 30.6 61.5

Seizure relatedPoor seizure control (%) 28.9 35.5 21.2Age at onset (yrs) (SD) 5.4 (3.5) 5.2 (4.0) 5.7 (3.0)Duration of epilepsy (yrs) (SD) 5.0 (3.5) 5.7 (3.8) 4.1 (3.1)

AED treatment (%)None 9.6 6.5 13.5Monotherapy 67.5 59.7 76.9Polytherapy 22.8 33.9 9.6

CognitiveFull Scale IQ (SD) 95.6 (16.6) 92.8 (17.4) 99.0 (15.1)Verbal IQ (SD) 95.0 (17.8) 92.5 (18.5) 98.1 (16.6)

LinguisticLanguage age (yrs) (SD) 9.2 (2.8) 9.5 (2.6) 8.8 (3.1)

CPS, complex partial seizure; PGE, primary generalized with absence.

UNMET PSYCHIATRIC NEED IN PEDIATRIC EPILEPSY 593


Childhood Behavioral Checklist (CBCL)Parents completed the CBCL (55), which consists of

20 social competence and 113 behavioral problem items.Although the CBCL generates broad-band (i.e., external-izing, internalizing) and narrow-band behavioral scales(i.e., aggression, depression, hyperactivity), only broad-band scores were used in the current study. The selectedcut point for clinically significant pathology in this studywas 65 (93rd percentile) (54).

Cognitive testingThe Wechsler Intelligence Scale for Children–Revised

(WISC-R) (56), administered to each child by a clinicalpsychologist, generated Full Scale, Verbal, and Perfor-mance IQ scores.

The Test of Language DevelopmentThe Test of Language Development (TOLD) (57) has

three forms: the TOLD-2 Primary, normed for childrenaged 4 to 8 years; the TOLD-2 Intermediate, normed forchildren aged 8 to 12 years; and the TOAL, normed foradolescents 12 to 18 years. Each form of the TOLD-2consists of a series of subtests through which it assessesboth vocabulary and grammar. Language age derivedfrom each of these tests was used as an independentvariable in the study’s data analysis.

Data analysisBecause the CPS and PGE groups did not differ based

on demographic, cognitive, or seizure-control variables,the CPS and PGE groups were pooled for this study.Logistic regression was used to determine which factorsare the best predictors of use of mental health treatment.Presence or absence of mental health treatment was thedependent variable, and demographic (age, gender, eth-nicity, SES, maximal parental education, referral source)seizure-related (type of seizure disorder, seizure control,age at onset, duration, AED treatment), cognitive (Verbaland Full Scale IQ), linguistic (language age), and behav-ioral variables (psychiatric diagnosis, comorbid psychi-atric diagnosis, CBCL scores) were used as thepredictors in the model. Variables that did not signifi-cantly contribute to the model were trimmed, preservingmodel hierarchy, until only a significant model re-mained. The tests were two-tailed, and results were con-sidered significant if the significance level was <0.05.

RESULTS

Unmet mental health needTable 2 presents rates of mental health treatment,

K-SADS psychiatric diagnosis, type of psychiatric diag-nosis, and CBCL measures (mean scores and percentageof subjects with scores in the clinical range) for thesample. Although 61% of the patients had a psychiatricdiagnosis, only 33% received mental health services;

therefore, nearly two thirds (67%) did not receive mentalhealth treatment.

Associated factorsThe logistic regression model yielded age of the child,

maximal educational level of the parents, verbal IQ ofthe child, AED polytherapy, and the interaction term oftype of seizure disorder, and psychiatric comorbidity(i.e., more than one psychiatric diagnosis) as significantpredictors of use of mental health services. We nowelaborate on the findings for each of these factors.

Younger children were less likely to have a history ofmental health treatment than were older children [Table3; odds ratio (OR), 1.3; 95% confidence interval (CI),1.1–1.5; p � 0.003]. Thus for each 5-year increase inage, the child was 3.75 times more likely to have receivedtreatment. In terms of other demographic factors, if theparents’ maximal educational level is less than highschool, a child was 5 times less likely to have a history ofmental health services (OR, 5.3; 95% CI, 1.7–16.5; p �0.004).

Regarding cognitive variables, higher verbal IQ wasassociated with absence of mental health treatment (OR,0.96; 95% CI, 0.93–0.98; p � 0.008). In other words, foreach 10-point decrease, the child was 1.5 times more likelyto have received some form of mental health services.

Of those seizure-related factors examined, only AEDpolytherapy (as compared with no drug or monotherapy)was associated with the presence of mental health treat-ment (OR, 2.8; 95% CI, 1.1–6.8; p � 0.02). Last, al-though type of seizure disorder (i.e., CPS vs. PGE) itselfwas not significant, a child with CPS and the presence ofpsychiatric comorbidity (i.e., more than one psychiatricdiagnosis) was nearly twice as likely to have received treat-ment (OR, 1.9; 95% CI, 1.0–3.6; p � 0.05). In otherwords, those children with PGE and a single psychiatricdiagnosis were less likely to have a history of mentalhealth treatment.

TABLE 2. Mental health treatment and psychopathologyin patients

All CPS PGE

N. 114 62 52Treatment (%) 32.7 40.3 23.5Diagnosis (%) 60.5 61.3 59.6

Affective/Anxiety 15.9 12.9 19.6Disruptive 23.0 22.6 23.5Comorbid 21.2 25.8 15.7

CBCL-Mean score (SD)Total 54.7 (13.3) 55.1 (13.8) 54.2 (13.4)Internal 53.0 (11.8) 53.5 (11.3) 52.5 (12.6)External 49.2 (12.4) 49.4 (13.2) 48.9 (11.5)

CBCL-T >65 (%)Total 23.9 22.9 25.0Internal 14.8 14.6 15.0External 13.6 14.6 12.5

CPS, complex partial seizure; PGE, primary generalized with ab-sence.

D. OTT ET AL.594


DISCUSSION

This is the first study to demonstrate unmet mentalhealth need in a large sample of children with CPS andPGE. Although ∼60% of the subjects were diagnosedwith a psychiatric diagnosis, >60% did not have a historyof mental health services. Absence of mental health carewas associated with younger age, less parental education,AED monotherapy (or none), and higher verbal IQ. Inaddition, children with PGE and a single psychiatric di-agnosis were less likely to have a history of mentalhealth treatment.

In the current study, the rates of children with a psy-chiatric diagnosis and percentages with CBCL scores inthe clinical range were comparable with other investiga-tions in children with epilepsy (3,4,8,9,29). Rates of be-havioral disturbance in neurologic conditions, such asepilepsy, are consistently higher as compared with otherchronic medical conditions (nearly 3 times) (58,59) andwith the general youth population (nearly 5 times) (1,13).However, to date, other than the current investigation,little information exists on how many of these childrenreceive mental health care and the associated demo-graphic, seizure-related, cognitive, and linguistic factors.

The discrepancy between the high rate of psychiatricdiagnosis (60%) and low rate of mental health services(33%) is, however, concerning. Because other studieshave demonstrated severe psychopathology includingsuicide (6,9,17,18), early identification and treatment ofpsychiatric problems in this population is particularlyrelevant.

Regarding demographic factors, as in studies involv-ing children with chronic illness (60) or those seen forroutine medical care (61), we also identified an associa-

tion of lower parental educational status with absence ofmental health care in children with CPS or PGE. Thisfinding suggests that greater education may facilitate un-derstanding or awareness of the child’s illness, his or heremotional status, and the link between the brain and be-havior.

The mixed referral source (i.e., patients from both thecommunity and university-based pediatric neurologyclinics) in the current study is similar to that in many ofthose previously described studies, which examined ratesof psychopathology in pediatric epilepsy (3–5,8,9). Al-though the number of children from university-basedclinical sources (55%) was slightly greater than from thecommunity source (45%), this difference was not relatedto the presence or absence of mental health treatment.Therefore absence of psychiatric care in those childrenwith epilepsy and psychopathology cannot be attributedto their referral source.

From the developmental perspective, the associationof mental health services with older chronologic age sug-gests that sufficient time from the onset of the epilepsy isrequired before psychiatric problems are identified andtreated. For those with a newly diagnosed seizure disor-der, the time needed by the family and child to acceptthis diagnosis may delay psychiatric interventions, whichmay be necessary even at this early stage (3,4). Similarly,the initial focus of both clinicians and parents on seizurecontrol also could obscure recognition of possible mentalhealth issues. Finally, many parents may lack the aware-ness of the potential impact of epilepsy on the child’sdevelopment and behavior (8). Additional studies are re-quired to examine if these factors play a role in this timelag until children with CPS and PGE receive mentalhealth care. Such studies are important because during

TABLE 3. Factors associated with mental health treatment in the logistic regression

Associated factor

Mental health treatment

Oddsratio CI p

Yes(n � 37)

No(n � 77)

DemographicAge (yr) (SD) 11.6 (3.1) 10.0 (2.8) 1.3 1.1–1.5 0.003Max parental education > high school (%) 59.5 51.3 5.3 1.7–16.5 0.004

CognitiveVerbal IQ (SD) 88.8 (15.8) 98.0 (18.0) 0.96 0.93–0.98 0.008

Seizure relatedAEDs 2.78 1.1–6.8 0.02None (%) 5.4 11.7Monotherapy (%) 56.8 72.7Polytherapy (%) 37.8 15.6CPS/PGE (%) 67.6/32.4 48.1/51.9

BehavioralComorbidity (%) 32.4 15.6Comorbidity × seizure diagnosis 1.9 1.0–3.6 0.05Comorbid and CPS (%) 27.0 7.8Comorbid and PGE (%) 5.4 7.8Single psychiatric diagnosis and CPS (%) 40.6 40.3Single psychiatric diagnosis and PGE (%) 27.0 44.2

NOTE: Interchange rows comorbidity % and comorbidity × seizure diagnosis.



this time, the child may experience fairly severe psycho-pathology, including suicidal ideation, as a result of thecumulative impact of epilepsy on the development of thechildren’s behavior (8), cognition (62), communication(42), and language (42,44).

Like psychopathology (1,2,8,9,24), mental health carealso is associated with lower Verbal IQ in children withCPS or PGE. Our earlier findings suggest that childrenwith average IQ scores have significant, but unrecog-nized, psychopathology (8,9). Perhaps children with CPSor PGE with good verbal skills are viewed as functioningwell psychologically; thus despite the presence of psy-chiatric problems, these children are underdiagnosed andundertreated.

Among the seizure-related variables examined in thecurrent investigation (i.e., type of seizure disorder, sei-zure control, age at onset, duration of epilepsy), onlyAED polytherapy was associated with a history of men-tal health services. Although past studies demonstratedan association of AED polytherapy (34,41) with behav-ioral disturbances, this is the first time that an associationwith mental health treatment has been demonstrated.Clearly, one interpretation of these findings is that thoseindividuals requiring multiple AED agents are morelikely to have behavioral or cognitive side effects, whichmay prompt psychiatric referral. Alternatively, the needfor AED polytherapy also may suggest to parents and/orclinicians that the child is “more ill” and thus facilitatepsychiatric interventions.

Although seizure type per se was not a significantfactor in the model, the presence of CPS, in combinationwith comorbid psychiatric diagnoses, was significantlyassociated with a history of mental health treatment (seeTable 3). Thus it seems that to receive mental healthservices, a child must have a seizure disorder with moreovert clinical manifestations (i.e., CPS), in addition tohaving severe (i.e., comorbid) psychiatric problems. Per-haps the parents’ and neurologists’ focus mainly on sei-zures enables them to assume that some of the children’sbehavioral difficulties are merely manifestations of theseizure disorder (19,60). Similarly, the lack of apparentictal manifestations in PGE, as well as the common be-lief that these children have no psychological difficulties,might prevent these children from receiving appropriateservices.

In terms of other behavioral factors, the CBCL, awidely used behavioral rating scale, did not identifythose children who received mental health interventions.Similarly, in an earlier study, we found that CBCL scoreswere not good predictors of the presence of a psychiatricdiagnosis in children with CPS and PGE (9). Thereforethe CBCL might have only limited value in identifyingpsychopathology in children with epilepsy who clearlywarrant mental health interventions.

In terms of the study’s limitations, we also recognize

that the use of language age scores derived from differentinstruments (TOLD primary, TOAL Intermediate,TOAL) does not rule out the possible role of undiag-nosed linguistic deficits as an associated factor. In addi-tion, because this study was not designed as anepidemiologic investigation, the study’s findings are lim-ited by the lack of detailed information about mentalhealth treatment (i.e., what type, provided by whom, ef-ficacy, duration, barriers to care). The findings imply theneed for well-designed mental health services studiesthat address these issues in this population. Last, thefocus on children with either CPS or PGE and averageIQ scores limits the generalizability of the study’s find-ings to children with epilepsy.

In conclusion, despite relatively high rates of psycho-pathology in children with CPS and PGE, few childrenreceive mental health treatment. Therefore cliniciansshould carefully evaluate children with these seizure dis-orders, who are young, have higher Verbal IQ scores,and parents with less than a high school education andrequire a limited number of AEDs (i.e., one or none). Inaddition, all children with PGE warrant careful assess-ment, including those without severe behavioral prob-lems.

Acknowledgment: This study was supported by NINDSgrant 1 RO1 NS 32070 (R.C.). We thank Shawn Zink, NatashaWheeler, Psy.D., and Amy Mo for their technical assistance.

REFERENCES

1. Rutter M, Graham P, Yule W. A neuropsychiatric study in child-hood. London: S.I.M.P./William Heineman Medical Books, 1970.

2. Graham P, Rutter M. Organic brain dysfunction and child psychi-atric disorder. Br Med J 1968;3:695–700.

3. Austin J, Harezlak J, Dunn D, et al. Behavior problems in childrenbefore first recognized seizures. Pediatrics 2001;107:115–22.

4. Dunn D, Austin J, Huster G. Behaviour problems in children withnew onset epilepsy. Seizure 1997;6:283–7.

5. Dunn D, Austin J, Huster G. Symptoms of depression in adoles-cents with epilepsy. J Am Acad Child Adolesc Psychiatry 1999;38:1132–8.

6. Ettinger A, Wesbrot D, Nolan E, et al. Symptoms of depressionand anxiety in pediatric epilepsy patients. Epilepsia 1998;39:595–9.

7. Hoare P. The development of psychiatric disorders among school-children with epilepsy. Dev Med Child Neurol 1984;26:3–13.

8. Caplan R, Arbelle S, Magharious W, et al. Psychopathology inpediatric complex partial and generalized epilepsy. Dev Med ChildNeurol 1998;40:805–11.

9. Ott D, Caplan R, Guthrie D, et al. Measures of psychopathology inchildren with complex partial seizures and primary generalizedepilepsy with absence. J Am Acad Child Adolesc Psychiatry 2001;40:907–14.

10. Blumer D, Montouris G, Hermann B. Psychiatric morbidity inseizure patients on a neurodiagnostic monitoring unit. J Neuropsy-chiatry Clin Neurosci 1995;7:445–56.

11. Mendez M, Cummings J, Benson D. Depression in epilepsy: sig-nificance and phenomenology. Arch Neurol 1986;43:766–70.

12. Perini G, Tosin C, Carraro C, et al. Interictal mood and personalitydisorders in temporal lobe epilepsy and juvenile myoclonic epi-lepsy. J Neurol Neurosurg Psychiatry 1996;61:601–5.

13. McDermott S, Mani S, Krishnaswami S. A population-based

D. OTT ET AL.596


analysis of specific behavior problems associated with childhoodseizures. J Epilepsy 1995;8:110–8.

14. Holdsworth L, Whitmore K. A study of children with epilepsyattending ordinary schools, I: their seizure patterns, progress andbehaviour in school. Dev Med Child Neurol 1974;16:746–58.

15. Ounsted C. The hyperkinetic syndrome in epileptic children. Lan-cet 1955;2:303–11.

16. Caplan R, Shields W, Mori L, et al. Middle childhood onset ofinterictal psychoses: case studies. J Am Acad Child Adolesc Psy-chiatry 1991;30:893–6.

17. Brent D. Overrepresentation of epileptics in a consecutive series ofsuicide attempts seen at Children’s Hospital, 1978–1983. J AmAcad Child Adolesc Psychiatry 1986;25:242–6.

18. Mackay A. Self-poisoning: a complication of epilepsy. Br J Psy-chiatry 1979;134:277–82.

19. Dunn D, Austin J. Behaviour issues in pediatric epilepsy. Neurol-ogy 1999;53:S96–100.

20. Stores G. School children with epilepsy at risk for learning andbehavioral problems. Dev Med Child Neurol 1978;20:502–8.

21. Hermann B, Whitman S, Hughes J, et al. Multietiological deter-minants of psychopathology and social competence in childrenwith epilepsy. Epilepsy Res 1988;2:51–60.

22. Whitman S, Hermann B, Black R, et al. Psychopathology andseizure type in children with epilepsy. Psychol Med 1982;12:843–53.

23. Lindsay J, Ounsted C, Richards P. Long-term outcome in childrenwith temporal lobe seizures, III: psychiatric aspects in childhoodand adult life. Dev Med Child Neurol 1979;21:630–6.

24. Camfield P, Gates R, Ronen G, et al. Comparison of cognitiveability, personality profile, and school success in epileptic childrenwith pure right versus left temporal lobe EEG foci. Ann Neurol1984;15:122–6.

25. Hoare P, Kerley S. Psychosocial adjustment of children withchronic epilepsy and their families. Dev Med Child Neurol 1991;33:201–15.

26. Kaminer Y, Apter A, Lerman P, et al. Psychopathology and tem-poral epilepsy in adolescents. Acta Psychiatry Scand 1988;77:640–64.

27. Mitchell W, Scheier L, Baker S. Psychosocial, behavioral, andmedical outcome in children with epilepsy: a developmental riskfactor model using longitudinal data. Pediatrics 1994;94:471–7.

28. Austin J, Risinger M, Beckett L. Correlates of behavioral problemsin children with epilepsy. Epilepsia 1992;33:1115–22.

29. Austin J, Dunn D, Huster G. Childhood epilepsy and asthma:changes in behavior problems related to gender and change incondition severity. Epilepsia 2000;41:615–23.

30. Austin J. Childhood epilepsy: child adaptation and family re-sources. J Child Adolesc Psychiatry Ment Health Nurs 1988;1:18–24.

31. Hermann B, Whitman S. Neurobiological, psychosocial, and phar-macological factors underlying interictal psychopathology in epi-lepsy. Adv Neurol 1991;55:439–52.

32. Hermann B, Whitman S, Wyler A, et al. Psychosocial predictors ofpsychopathology in epilepsy. Br J Psychiatry 1990;156:98–105.

33. Carlton-Ford S, Miller R, Brown M, et al. Epilepsy and children’ssocial and psychological adjustment. J Health Soc Behav 1995;36:285–301.

34. Caplan R, Gillberg C. Psychiatric and social aspects of epilepsy inchildhood, In: Engel J Jr, Pedley T, eds. Epilepsy: a comprehensivetextbook. Vol II. New York: Raven Press, 1998:2125–39.

35. Nuffield E. Neurophysiology and behavioral disorders in epilepticchildren. J Ment Sci 1961;107:43–58.

36. Caplan R, Arbelle S, Guthrie D, et al. Formal thought disorder andpsychopathology in pediatric primary generalized and complexpartial epilepsy. J Am Acad Child Adolesc Psychiatry 1997;36:1286–94.

37. Austin J, Huster G, Dunn D. Adolescents with active and inactive

epilepsy or asthma: a comparison of quality of life. Epilepsia 1996;37:1228–38.

38. Lindsay J, Ounsted C, Richards P. Long-term outcome in childrenwith temporal lobe seizures, I: social outcome and childhood fac-tors. Dev Med Child Neurol 1979;21:285–98.

39. Williams J, Sharp G, Bates S, et al. Academic achievement and thebehavioral ratings in children with absence and complex partialepilepsy. Educ Treat Child 1996;19:143–52.

40. Pritchard P, Lombroso C, McIntyre M. Psychological complica-tions of temporal lobe epilepsy. Neurology 1980;30:227–32.

41. Harbord M. Significant anticonvulsant side-effects in children andadolescents. J Clin Neurosci 2000;7:213–6.

42. Caplan R, Guthrie D, Komo S, et al. Social communication inchildren with epilepsy. J Child Psychol Psychiatry 2002;43:245–53.

43. Ounsted C. Aggression and epilepsy: rage in temporal lobe epi-lepsy. J Psychosom Res 1969;13:237–42.

44. Caplan R, Guthrie D, Komo S, et al. Conversational repair inpediatric epilepsy. Brain Lang 2001;78:82–93.

45. Caplan R, Guthrie D, Shields W, et al. Communication deficits inchildren undergoing temporal lobectomy. J Am Acad Child Ado-lesc Psychiatry 1993;32:604–11.

46. Caplan R, Guthrie D, Shields W, et al. Formal thought disorder inpediatric complex partial seizure disorder. J Child Psychol Psy-chiatry 1992;33:1399–412.

47. Farwell J, Batzel L. Neuropsychological abilities of children withepilepsy. Epilepsia 1985;26:395–400.

48. Funakoshi A, Morikawa T, Maramatsu A. A prospective WISC-Rstudy in children with epilepsy. Jpn J Psychiatry Neurol 1988;42:562–4.

49. Rodin E, Schmaltz S, Twitty G. Intellectual functions of patientswith childhood-onset epilepsy. Dev Med Child Neurol 1986;28:25–33.

50. Stanicolo M, Galetti F. Idiopathic epilepsy and school achieve-ment. Arch Dis Child 1994;70:424–8.

51. Hollingshead A. Medical sociology: a brief review: Milbank Me-morial Fund Quarterly. Health Soc 1973;51:531–42.

52. World Health Organization. The ICD-10 classification of mentaland behavioral disorders: diagnostic criteria for research.Geneva: WHO, 1993.

53. Orvaschel H, Puig-Antich J. Schedule for affective disorders andschizophrenia for school-age children: epidemiologic version.Florida: Nova University Center for Psychological Services, 1987.

54. American Psychiatric Association. Diagnostic and statisticalmanual of mental disorders. 4th ed. (DSM-IV). Washington, DC:American Psychiatric Association, 1994.

55. Achenbach T. Manual for the Child Behavior Checklist and re-vised Child Behavior Profile. Vermont: Department of Psychiatry,University of Vermont, 1991.

56. Wechsler D. Wechsler Intelligence Scale for Children. New York:The Psychological Corporation, 1955.

57. Newcomer P, Hammil D. Test of Language Development-2 Pri-mary. Austin, TX: Pro-Ed, 1988.

58. Howe G, Feinstein C, Reiss D, et al. Adolescent adjustment tochronic physical disorders, I: comparing neurological and non-neurological conditions. Child Psychol Psychiatry 1993;34:1153–71.

59. Austin J. Comparison of child adaptation to epilepsy and asthma.J Child Adolescent Ment Health Nurs 1992;33:1115–22.

60. Gortmaker S, Walker D, Weitzman M, et al. Chronic conditions,socioeconomic risks, and behavioral problems in children and ado-lescents. Pediatrics 1990;85:267–76.

61. Brugman E, Reijneveld SA, Verhulst FC, et al. Identification andmanagement of psychosocial problems by preventive child healthcare. Arch Pediatr Adolesc Med 2001;155:4629.

62. Strauss E, Loring D, Chelune G, et al. Predicting cognitive im-pairment in epilepsy: findings from the Bozeman Epilepsy Con-sortium. J Clin Exp Neuropsychol 1995;17:909–17.



behavioral disorders in pediatric epilepsy: unmet psychiatric need

Documents