Sex-genetic model. Additive genetic factors are correlated at 0.5in dizygotic (DZ) twins, because DZ twins share on average half of their genes,and at 1.0 in monozygotic (MZ) twins. The parameters a, c, and e are loadings of the observed phenotype (P) onthe latent additive genetic factors (A), common or shared environment factors(C), and nonshared environment effects (E) and indicate the degree of relationsbetween the latent factors and the observed P. The proportion of the varianceaccounted for by genetic and environmental influences is calculated by squaringthe parameters a, c, and e and dividingthem by the total variance (a2+c2+e2). In addition, in the univariate model,the effects of sibling interaction (path s) are also considered.Genotype×sex interaction effects (illustrated for the case of unlike-sexsibling pairs) may take the form of sex differences in the magnitude of geneticor environmental influences (paths am, cm, and em vs af, cf, and ef) or the form of an additionalgenetic or common environmental influence on only 1 of the unlike-sex siblingpairs (paths a′ and c′). OCS indicatesObsessive-Compulsive Scale.
Hudziak JJ, van Beijsterveldt CEM, Althoff RR, Stanger C, Rettew DC, Nelson EC, Todd RD, Bartels M, Boomsma DI. Genetic and Environmental Contributions to the Child Behavior ChecklistObsessive-Compulsive ScaleA Cross-cultural Twin Study. Arch Gen Psychiatry. 2004;61(6):608-616. doi:10.1001/archpsyc.61.6.608
We have reported elsewhere on the development of an 8-item Obsessive-Compulsive
Scale (OCS) contained in the Child Behavior Checklist (CBCL) to identify children
who meet criteria for DSM-IV obsessive-compulsive
disorder. Twin studies of obsessive-compulsive disorder have indicated a significant
genetic component to its expression.
To determine the relative contributions of genetic and environmental
influences on childhood obsessive-compulsive behavior using the CBCL OCS in
The CBCL data were received by survey of twins in the Netherlands Twin
Registry (NTR) and the Missouri Twin Study (USA/MOTWIN).
General community twin samples.
Participants were 4246 twin pairs aged 7 years, 2841 aged 10 years,
and 1562 aged 12 years (who also participated in the study at 7 and 10 years
of age) from the NTR and 1461 mixed-age twin pairs (average age, approximately
9 years) from the USA/MOTWIN.
Main Outcome Measures
Model fitting to test for genetic and environmental influences, sex
differences, and sibling interaction/rater contrast effects on the CBCL OCS.
In each case, the best-fitting model was one that indicated significant
additive genetic influences (range, 45%-58%; 95% confidence interval [CI],
45%-61%), and unique environmental influences (range, 42%-55%; 95% CI, 39%-55%),
with shared environmental influences in the NTR sample aged 12 years (16%).
Sex differences were seen in the mixed-age USA/MOTWIN model, but not in the
NTR samples. No evidence of dominance, sibling interaction, or rater-contrast
effects was seen. These data were relatively consistent across age and cultures.
The CBCL OCS is influenced by genetic factors (approximately 55%) and
unique environmental factors (approximately 45%) in the younger sample, with
common environmental influences only at 12 years of age. These effects do
not vary with differences in sex or sibling interaction/rater contrast effects.
Our data reveal higher genetic influences for obsessive-compulsive behavior
and do not demonstrate genetic differences across sex.
Obsessive-compulsive disorder (OCD) is common in children and adults.It is associated with serious impairment and, in many cases, has a lifelongcourse. Studies on prevalence indicate that the lifetime rates of OCD in adolescentsrange from 1.9% to 4.1%,1- 6 witha higher prevalence in girls than in boys.7 Thewide variability in prevalence rates may be due to differences in the populationssampled and the methods used for assessment. Nonetheless, the prevalence ofOCD in children and adolescents may be higher than originally thought. Otherreports8 (including unpublished data March2003: J.J.H., R.R.A., C.S., C.E.M.v B., E.C.N., G. L. Hanna, MD, D.I.B., andR.D.T.) have described our development of an 8- item Obsessive-CompulsiveScale (OCS) contained in the Child Behavior Checklist (CBCL) that is usefulin identifying children aged 7 to 12 years who meet criteria for DSM-IV OCD (Table 1). Usinga summed score of 5 (the borderline clinical range, >95th percentile basedon the CBCL normative sample), the CBCL OCS demonstrated high sensitivity(92%), moderate specificity (67%), high negative predictive value (90%), andmoderate positive predictive value (73%) in subjects who had been diagnosedas having DSM-IV OCD by board-certified psychiatrists.Using a higher 98% clinical range, which corresponds to a score of 6, thesensitivity of the test was 79%; specificity was 78.1%; positive predictivevalue was 77.8%; and negative predictive value was 79.4%. In 4 large twinsamples (1 in the United States and 3 in the Netherlands), the percentageof participants with CBCL OCS scores in the borderline clinical range was2.4% to 4.3%, with 1.4% to 3.8% in the clinical range (Table 2).
The purpose of the present study was to extend our previous work anddetermine the contributions of environment, genes, sex, and age to the expressionof CBCL OCS and, by extension, childhood OCD.
The estimation of genetic and environmental influences on a disordertypically includes twin, family, adoption, and molecular genetic approaches.Tsuang et al9 argued that although our moleculargenetic techniques have advanced to the point that identifying genetic variantsthat contribute to the development of a phenotype is now a trivial laboratoryexercise, the development of phenotypic identification strategies that refinediagnoses for molecular genetic investigations remain problematic. This isespecially true in childhood OCD, where relatively few genetic studies havebeen performed. Family and twin studies of OCD are rare relative to the studiesof other psychiatric disorders. Of the OCD studies, many have used small samplesthat were generally derived from highly comorbid clinical populations, witha resultant reduction in the generalizability of the results.10 Despitethese limitations, the belief that OCD is influenced by genetic factors iswidely held; in fact, initial reports on the genetic contributions to OCDare more than 60 years old.11
Family studies to date have revealed conflicting results. The percentageof affected first-degree relatives of patients with OCD has ranged from beingindistinguishable from control subjects12 to30%,13 with several estimates between.14- 18 ManyOCD studies include patients with other disorders such as Tourette syndromeand other tic disorders. These studies have reported rates of OCD as highas 26% among first-degree relatives of patients with Tourette syndrome, afamilial condition known to have significant genetic influences.19,20 Arecent meta-analysis21 demonstrated an unadjustedaggregate risk of 8.2% in first-degree relatives of patients with OCD vs 2%in relatives of controls. In summary, published family studies support thecontention that OCD, alone or comorbid, is a condition that is influencedby genetic factors.
Twin studies have also demonstrated genetic influences on OCD. Monozygotic(MZ) twins have been shown to have a concordance rate for OCD as high as 70%to 80%, compared with 22% to 47% among dizygotic (DZ) twins.22,23 Heritabilityestimates have been calculated in the range of 26% to 33%.10,24 However,Hettema et al21 were unable to find any twinstudies of adequate size without ascertainment bias to meet their criterionfor inclusion, although they noted that published twin studies had found consistentevidence of genetic contributions to obsessions and compulsions.
Most molecular genetic studies of OCD have focused on the monoaminepathway genes. Numerous molecular genetic studies have aimed to determinethe relative contributions of different candidate genes to the pathophysiologyof OCD. The catechol O-methyltransferase (COMT) gene,25- 28 theserotonin 2β receptor (5HT2B) gene,29,30 the serotonin transporter gene,31,32 the serotonin 2α receptor (5HT2A) gene,33,34 andthe 7 repeat of the dopamine D4 receptor (DRD4) gene,35 among others, have all beenimplicated in OCD or OCD-like phenotypes, some perhaps in a sex-specific36 or a population-specific27,37 fashion.Other reports have shown the contrary.38- 40 Ineach of these studies, the authors considered the results preliminary andcalled for studies on much larger populations and more refined samples toclearly understand the contribution of individual gene variations to the etiologyof OCD.
In aggregate, family, twin, and molecular genetics studies support thepremise that OCD or features of OCD are influenced by genetic factors. Mostof these studies did not control for sex, age, referral bias, or comorbidity.Thus, confusion remains about the best way to conceptualize and refine theOCD phenotype for genetic analysis.
There were multiple aims of this study. The first was to determine thegenetic and environmental contributions to CBCL OCS scores. The second wasto determine whether evidence of sex-genetic interactions existed. The thirdwas to determine whether the age of the child contributed to the genetic/environmentalinfluences on CBCL OCS scores by analyzing these data in samples of twinsaged 7, 10, and 12 years and a mixed sample of twins aged 8 to 12 years. Finally,the study assessed cultural differences by determining whether the genetic/environmentalcontributions differed by country (the Netherlands vs the United States).
The study was part of an ongoing twin-family study of health-relatedcharacteristics, personality, and behavior in the Netherlands. The subjectswere all part of the Netherlands Twin Registry (NTR).41 Atpresent, the NTR has data on more than 25 000 twin pairs from ages 3to 30 years. For this study, we assessed a sample of Dutch twin pairs whoseparents reported on their behavior when they were 7, 10, and 12 years of age(NTR-7, NTR-10, and NTR-12 samples, respectively). The total sample with CBCLdata available in the NTR at these age groups at the time of the study included4484, 2905, and 1664 twin pairs aged 7, 10, and 12 years, respectively. Ofthese, 238, 64, and 102 twin pairs, respectively, were excluded owing to missingdata, leaving a total of 4246, 2841, and 1562 twin pairs available for analysisat ages 7, 10, and 12 years, respectively. The socioeconomic status of theparents of the twins was somewhat higher than the level in the general Dutchpopulation.42
The assessment procedures at ages 7, 10, and 12 years have been describedelsewhere.41 Parents received a CBCL by mail.Parents who did not return the forms within 2 months received a reminder.Those who did not respond after 4 months were called by the NTR research assistant.This procedure resulted in an average continued participation rate of 80%from 7 to 10 and 10 to 12 years of age.
Zygosity were determined by means of DNA analyses of blood group polymorphismsfor 634 same-sex twin pairs. For the remaining twins, zygosity was determinedby questionnaire items about physical similarity and frequency of confusionof the twins by family and strangers. The classification of zygosity was basedon a discriminant analysis, relating the questionnaire items to zygosity basedon blood/DNA typing. According to this analysis, the zygosity was correctlyclassified by questionnaire in nearly 95% of the cases.43
To apply this screening tool to a large sample of twins with differentages and mixed ethnicity, we selected twins from an ongoing project, the MissouriTwin Study (USA/MOTWIN). This sample has been described previously.44 Briefly, an attempt was made to contact parents ofall twins born in Missouri from 1975 to 1991 to invite them to participate.They were paid $5 for completing survey materials. At the time that the geneticanalyses were performed, data on 1461 of 1565 twin pairs who were sent CBCLswere used. One hundred four pairs were excluded because 1 or more of the 8items of the CBCL OCS were missing. The CBCL OCS scores from each of theseremaining twin pairs were computed.
Zygosity was determined by means of questionnaire items as for the NTRsamples, with assignment based on a latent class approach. In a comparisonof genotypic determination of 121 twin pairs, only 1 pair was misassignedusing this method.45
Table 3 provides a descriptionof the numbers of twin pairs by sex and zygosity for the 4 samples. Therewas a small but statistically significant difference in age, with girls slightlyyounger; however, there was no statistical age difference by zygosity group.Although the USA/MOTWIN sample was ethnically mixed (85% European American,13.4% African American, and <1% Hispanic, Asian, or Native American), theNTR samples consisted primarily of children of European descent. The NTR-10sample was a subset of the NTR-7 sample studied 3 years later, and the NTR-12sample was a subset of the NTR-10 sample.
The CBCL is a widely used questionnaire for parents to respond to 118problem behaviors exhibited by their child during the previous 6 months. Theparent responds along a 3-point scale with the code of 0 if the item is nottrue of the child, 1 for sometimes true, and 2 for often true. The characteristicsand psychometric stability of the CBCL have been well established.46 The analyses performed herein used the 1991 versionof the CBCL, but the same items can be scored on the more recent 2001 version.47
The CBCL OCS was developed using factor analysis on 11 CBCL items thatwere thought to likely predict OCD.8 Usinga 1-factor model, 8 items were retained and were shown to have good internalconsistency (Cronbach α = .84). Those items retained are shown in Table 1, along with their CBCL item number.A numerical value for the CBCL OCS is created by adding the scores on these8 items (0, 1, or 2 for each), thus limiting the scale to a range of 0 to16. The CBCL OCS was tested to determine prevalence, specificity, and sensitivity.9
Means, variances, and twin correlations were calculated using Mx.48 Differences in mean scores and variances betweensex and zygosity were tested by means of likelihood ratio c2 tests.These tests were performed taking into account the dependency that existsbetween scores of the twins. Because the CBCL OCS score was not normally distributed,the data were square root transformed to approximate a normal distribution.
Genetic and environmental influences on CBCL OCS scores were computedusing structural equation modeling. The relative contributions of geneticand environmental factors to individual differences in CBCL OCS scores canbe inferred from the different level of genetic relatedness of MZ and DZ twins.49Figure 1 summarizesthe fundamental univariate genetic model that underlies our analyses. Thevariance may be due to additive genetic factors (A), common or shared environmentfactors (C), or nonshared environment effects (E). We also tested for dominancegenetic effects (D), which correlate at 1.0 in MZ twins and 0.25 in DZ twins.Estimating D and C at the same time is not possible in a design using onlyMZ and DZ twins reared together. Using D instead of C in the models did notcontribute to a better fit; thus, D was not examined further. The geneticfactors are correlated at 1.0 in MZ twins, as they are genetically identical.For DZ twins, the additive genetic factors are correlated at 0.5, becauseDZ twins share on average half of their genes. The environment shared by atwin pair is assumed not to depend on zygosity, and thus shared environmentalfactors correlate at 1.0 in both MZ and DZ twins. The E term is by definitionuncorrelated. All uncorrelated error is also absorbed in the E term. The parameters a, c, and e are loadings of the observed phenotype on the latent factors A, C,and E and indicate the degree of relations between the latent factors andthe observed phenotype. The proportion of the variance accounted for by geneticand environmental influences is calculated by squaring the parameters a, c, and e and dividing them by the total variance (a2+c2+e2). In addition, in the univariate model, the effects of sibling interaction(path s) are also considered. Sibling interactionreflects the effect of the behavior of one twin on the behavior of the othertwin. The interaction effect may also be due to bias in parental reports whenparents rate their children's behavior in comparison with each other. Whetherthe sibling interaction effects are a function of rater contrast or of realsibling interaction cannot be tested with the current data, but would needinformation from more than 1 informant. The AE model with sibling interactionwas tested and did not lead to a better fit than the AE model without siblinginteraction, and it was not examined further.
Figure 1 extends the latentvariable component of the model by allowing for genotype×sex interactioneffects (illustrated for the case of unlike-sex sibling pairs). This may takethe form of sex differences in the magnitude of genetic or environmental influences(paths am, cm, and em vs af, cf, and ef) or the form of an additional genetic or common environmentalinfluence on only 1 of the unlike-sex sibling pairs (paths a′ and c′). These analyses allowedus to test for sex differences on CBCL OCS scores.
To estimate the genetic and environmental contributions, the data fortwins 1 and 2 were summarized into 2 × 2 covariance matrices, computedby PRELIS scientific software.50 All modelfitting was performed with Mx,48 a statisticalsoftware package designed for conducting genetic analyses with an approachthat is standard in structural equation modeling.51 Thebasic model tested was an ACE model. The significance of the A and C factorswas tested by dropping each of these variance components one at a time andusing the c2 difference test. The c2 statistic was computedby subtracting the c2 statistic for the full model from that fora reduced model. The degrees of freedom for this test are equal to the differencesbetween the degrees of freedom for the full and the reduced model. If thec2 statistic is significant, this means that the variance componentmakes a significant contribution to the fit of the full model, because removingit significantly worsens the fit of the model. In addition, the Akaike informationcriterion, a goodness-of-fit index that considers the rule of parsimony, wascalculated. A smaller Akaike information criterion indicates a better fit.We also computed likelihood-based 95% confidence intervals (CIs) for eachparameter.48,52 More technicaldetails of genetic model-fitting analyses are reviewed elsewhere.49
The square root–transformed mean CBCL OCS scores and variancesacross sex and zygosity are presented in Table 4. Raw CBCL OCS scores were quite similar across age and country.The homogeneity of the variance across sex was tested with Mx48 andrevealed significant sex differences for CBCL OCS scores in all 3 samples.In the USA/MOTWIN sample, boys had higher scores; in both Dutch samples, girlshad higher scores. The variance and covariance matrices for all zygosity groupsare given in Table 5. There wereno differences in the means, variances, and covariances across the 5 zygositygroups for the USA/MOTWIN sample and the NTR-10 or NTR-12 samples. In theNTR-7 sample, the means and the variances of the DZ twins were larger thanfor MZ twins (Δc28 = 21.18). As shown in Table 5, although significant, these differenceswere very small.
The twin correlations for the CBCL OCS score are shown in Table 6. Analysis of twin correlations yielded evidence of the influenceof genetic and environmental factors. In all 3 samples, MZ correlations werelarger than DZ correlations, indicating the influences of genes. The MZ andDZ correlations were not different across sex, with 1 exception. The femaleDZ correlation was lower than the male DZ correlation in all samples, butonly significantly so in the USA/MOTWIN sample. In the remaining samples,the magnitude of genetic and environmental effects was equal across sex. Inaddition, the DZ opposite-sex correlations equaled the same-sex male correlationsin all 3 samples, suggesting that the same genes and environmental influencesplay a role for boys and girls.
A summary of the model-fitting results is given in Table 7. The c2 statistic indicates the goodness of fit,and smaller c2 statistics indicate better agreement of the observeddata with the model. First, we computed a model for each sample that allowedthe variance components to differ between boys and girls. In the second setof models, A, C, and E parameters were constrained to be equal across sex.These constrained models were compared with the unconstrained models, andthe best-fitting models were selected. These results showed no deteriorationin fit in the 2 Dutch samples when the parameters were constrained to be thesame across sex. In the USA/MOTWIN sample, this resulted in a worsening ofthe fit (Δc23 = 16.85). This sex difference wasprobably due to differences in total variance between boys and girls seenin this sample only. Next, the significance of the C factor was tested bydropping it from the models and calculating change in the goodness of fit.Dropping C from the model did not lead to a deterioration of the fit in anyof the samples except for the NTR-12, meaning that the AE was the best modelfor all 3 of the other samples, with an ACE model being the best fit for theNTR-12 sample only.
The best-fitting model for 3 samples included A and E contributions(Table 7), with C contributionsevident in the NTR-12 sample. There were no sex effects in the Dutch samples,although there were minor sex effects in the USA/MOTWIN sample, likely dueto underlying differences in the means and variances of the CBCL OCS scorein this mixed-aged sample. Across age groups and cultures, the additive geneticinfluence of the CBCL OCS varied from 45% to 58% (95% CI, 45%-61%). The Efactors (which conspire to make members of a twin pair different) ranged from42% to 55% (95% CI, 39%-55%). In the NTR-12 only, the magnitude of the sharedenvironmental influences was about 16%.
With these results, we extend our previous work on the CBCL OCS by revealingthat scores on this proposed scale for assessing childhood OCD are highlyheritable and influenced by additive genetic and unique environmental factorsin younger children, with common environmental influences appearing to playa role beginning at 12 years of age. The magnitude and type of the geneticand environmental influences were surprisingly stable across age, sex, andculture within the younger group, but may differ as the children enter puberty,given the differences reported herein at 12 years of age.
The magnitude of the genetic contribution for each sample is largerthan previous estimates,10,24 whichfell in the range of 26% to 33%. There are several reasons for this difference.First, restriction of range introduced by examining genetic effects on OCDin clinical samples may have attenuated previous estimated genetic contributions.Clinical cases are often more severe, with higher rates of comorbid illnessthat may be influenced by environmental factors that, in concert with geneticrisks, may lead to the expression of OCD (eg, pediatric autoimmune neuropsychiatricdisorders associated with streptococcal infections53).We also used a continuous CBCL OCS score rather than the dichotomous DSM-IV diagnoses used by others. Previous studies haveobserved high rates of subclinical OCD symptoms in family members of OCD probands.By using CBCL OCS scores, we may have better characterized the underlyingtrait.
Second, the environmental contributions for each of the younger samplesare of the unique or unshared type. There are a variety of propositions aboutwhich unique environmental contributions may lead to the expression of OCD,including the presence of autoimmune processes. The unique environmental contributionfound in all 3 samples may explain earlier reports of genetic-environmentalinteraction leading to the expression of OCD. We did not directly test forpediatric autoimmune neuropsychiatric disorders associated with streptococcalinfections in our samples. Other possibilities include differences in parenting,school, activities, etc, as well as error that is part of the E term in structuralequation modeling.
Third, although we did not test for developmental factors (these datawill be reported later once our samples are large enough to follow up childrenacross key developmental periods), it is important to note the similarityof the genetic/environmental contributions to CBCL OCS scores within and acrossthe young age groups. Collapsing twin pairs aged 8 to 12 years into the sameanalyses (ie, the USA/MOTWIN sample), and therefore introducing a possibledevelopmental bias into the analyses, had no effect on the magnitude of thegenetic or environmental contributions, with the possible exception of inflatingthe contribution of sex. However, looking only at the cross-sectional analysisat 12 years of age, shared environmental influences first appear. This maymean that investigation of these same twins in adolescence (a topic for furtherstudy) may help reveal why sex differences change after puberty and why othergenetic studies have shown lower estimates of OCD.
Fourth, no sex-genetic differences were apparent in most of these models,except in the USA/MOTWIN group. This may be due to the inclusion of childrenwho are mostly younger than 12 years. Previous work on the epidemiology ofOCD in children has shown that the prevalence of OCD in children increasesmarkedly after 13 years of age54 and that ashift in prevalence from affecting boys more frequently at a young age toaffecting women more frequently in adulthood likely occurs after 18 yearsof age.55 In our study, we see little, if any,sex difference affecting the heritability of the CBCL OCS scores up to 12years of age. Future research will be directed at the adolescent period todetermine whether sex effects increase during this time of life.
Finally, in studies of psychopathology, there have been notable differencesbetween the European and US psychiatric communities. These include, but arenot limited to, differences in how schizophrenia vs bipolar affective disorderswere conceptualized in the 1950s and 1960s, leading to markedly differentrates of both disorders across continents.56 Morerecently, there have been differences in how children with symptoms of inattention,hyperactivity, and impulsivity were characterized across the two continents.In England, the International Classification of Diseases,Eighth Revision, described hyperkinetic conduct disorder, whereas inthe United States during the same time period, the era of attention-deficit/hyperactivitydisorder was born.57 These across-continentdifferences led to different diagnostic paradigms, different prevalences,and different treatment approaches. Our data on the CBCL OCS provide a nicecontrast. In our previous work, using normative data with the same instrumentin the Netherlands and the United States, we were able to compute rates ofthe prevalence of CBCL OCS in both cultures independent of bias that may emergeby imposing standards of one culture on another (unpublished data March 2003:J.J.H., R.R.A., C.S.,C.E.M.v B., E.C.N., G. L. Hanna, MD, D.I.B., and R.D.T.)We then tested for genetic contributions to a continuous distribution of CBCLOCS scores across both cultures, and essentially found the same results. Whetherthese results apply to clinically diagnosed OCD can only be determined byfuture studies; however, the similarity across age, sex, and culture supportsthe premise that deviance on the CBCL OCS represents a prevalent syndromeand that scores on the CBCL OCS are stable across ages and cultures with asignificant genetic component.
These similar genetic and prevalence data raise another dilemma. Estimatesof the prevalence of attention-deficit/hyperactivity disorder in the UnitedStates range from 3% to 5% in general population studies and are somewhathigher in twin studies.58 Attention-deficit/hyperactivitydisorder is the second most common disorder seen in US child psychiatry clinicsand the most common psychiatric disorder treated by pediatricians.59 If OCD is nearly as common as attention-deficit/hyperactivitydisorder, with prevalence rates in the range of 2% to 4% according to ourstudies, and is highly stable across age and culture, why are so few childrenidentified and treated for this diagnosis? Although our data cannot directlyanswer this question, one possible explanation has to do with the difficultyin screening for and thus diagnosing OCD when using existing OCD diagnosticinstruments. Furthermore, the psychopathology measured by deviance on theOCS may not be impairing enough to parents or teachers to lead to early identificationand referral. As March and colleagues60 havedemonstrated, the obsessional and compulsive characteristics of children withOCD are often not viewed as pathologic by parents or teachers. In fact, manyparents become so familiar with their childrens' symptom complex that theylose the ability to discriminate what is normal vs pathologic. Finally, withthe high and stable heritability estimates that have emerged in this study,together with the family study data that indicate OCD is highly familial,12- 18 itis also possible that children with OCD are not being identified as havingan illness, because their parents have a similar or the same malady. Thesequestions can be answered only by extending this research to a twin/familydesign. In such a study, twins with deviant scores on the OCS could be enrolledin a twin/family design to test for endophenotypic, genetic, and environmentalcontributions to this disorder. The prevalence and genetic data that haveemerged from our studies suggest that such research should be performed soon,as it is likely that most children with this illness are not being identified,are not receiving treatment, and are suffering in private.
The genetic and environmental contributions presented in this reportreflect CBCL OCS scores, not clinical measures of DSM-IV OCD. Although we have performed prior studies to demonstrate the validity,specificity, sensitivity, and predictive power of the CBCL OCS in relationto DSM-IV OCD, it remains possible that the CBCLOCS may overidentify or underidentify cases in general population samples.One specific set of cases that may be underrepresented is the population ofchildren who may have an alternative manifestation of OCD associated withtics. No item in the CBCL OCS assesses tics. Prospective studies of the CBCLOCS or similar measures with more traditional end-point clinical assessmentsare needed to address this issue. The cutoffs used for the CBCL OCS were 92%sensitive but only 67% specific, resulting in many false-positive findings.The CBCL OCS cut points could be changed, (eg, a score of 7 instead of 6 for10-year-old children), and the scale will become less sensitive and more specific.Higher cut points may be needed for gene-finding expeditions where false-positivefindings are less acceptable.
A further limitation is the fact that parent ratings of the same twinswere included at 7, 10, and 12 years of age. Although this provides us a windowon the genetic and environmental contributions to CBCL OCS at specific ages,it could also introduce an ascertainment bias (ie, why do specific parentsparticipate at each wave and others do not?). A long-term aim of this workis to test developmental stability and change when our sample sizes are largeenough to allow for such analyses. An additional limitation is the relianceon parental report, given the secrecy that is inherent to children's OCD symptoms.61 Although youth self-report may be of questionablevalue, as these children move into adolescence a reassessment of the CBCLOCS using youth-self report will be important.
These data support the contention that childhood obsessive compulsivebehavior is prevalent, influenced by both genetic and environmental factors,and affects children of both genders across the 7- to 12-year age range. Thefindings provide a strategy for using quantitative, gender, and developmentallysensitive screening approaches to identify children at risk for this commonand impairing disorder. Future studies using self-reports from children andadolescents may reveal more sensitive and specific ways to screen for OCDacross the lifespan of an individual.
Corresponding author and reprints: James J. Hudziak, MD, Departmentof Psychiatry, University of Vermont, Given Bldg, Room B229, Burlington, VT05405 (e-mail: email@example.com).
Submitted for publication May 1, 2003; final revision received August20, 2003; accepted December 18, 2003.
This study was supported by grants MH58799 and MH52813 from the NationalInstitute of Mental Health, Rockville, Md.
We thank Jeri Ogle for her assistance.