|
A Family Study of Juvenile Obsessive-Compulsive Disorder P Srinivas Reddy, DPM, DNB1, YC Janardhan Reddy, MD, DPM2, S Srinath, MD, DPM3, S Khanna MD, DPM, PhD, MRCPsych3, SP Sheshadri, MD, DPM3, SR Girimaji, MD3 |
|||
|
Canadian Journal of Psychiatry, May 2001 |
|||
|
Objective: To determine
whether juvenile obsessive-compulsive disorder (OCD) is familial
and whether the rate of Tourette syndrome (TS) and tic disorders is higher
among relatives of patients with OCD than among relatives of controls
subjects. |
|||
|
The familial nature of obsessive-compulsive disorder (OCD) has been described since the 1930s (1-14), but the subject remains controversial. While some early family studies reported findings that support the familial nature of OCD (1-4,8), others found no increase in the rates of OCD among first-degree relatives (5-7). Many earlier studies are, however, difficult to interpret because of differences in diagnostic criteria and methods of assessment. Most studies did not directly interview relatives and lacked control groups. In some recent studies, the methodological shortcomings of the earlier studies have been addressed by directly interviewing relatives and employing standard diagnostic criteria (9-16). Three of the studies (9-11) reported a morbidity risk of 13% to 36% in families having children with OCD. A Manuscript received April 2000, revised, and accepted
March 2001. |
major methodological weakness of all 3 studies was the lack of a comparison group. The study by Pauls and others (14), which included a comparison group, reported a higher risk for OCD (20%) among the relatives of probands with early-onset OCD (less than age 18 years) than among those with late-onset OCD (11%), whilein relatives of the psychiatrically unaffected comparison group, the rate was 1.9%. Interestingly, 2 studies found evidence for familiality only in relatives of early-onset probands (12 ). All these studies strongly support the possible familial component for the expression of OCD, particularly childhood- and adolescent-onset OCD. Additional support for the role of familial factors in the expression of CD comes from the finding of high rates of OCD among relatives of probands with Tourette syndrome (TS) (15,17-19). Similarly, high rates of tic disorders and TS have been reported among relatives of OCD probands (11 ), suggesting that some forms of OCD may be etiologically (and perhaps genetically) related to TS. While many studies report increased rates of OCD in relatives of OCD probands (9-12,14-16), findings of some studies do not support the familial nature of OCD (13,20). The study by Black and others (13) found no evidence that OCD was familial, and they concluded that an anxiety diathesis rather than than OCD per se may be inherited. A family study from our centre (20), which examined 135 adult first-degree relatives of 35 adults with OCD and 148 adults from the general population, found no evidence for familiality. Next |
||
|
May 2001 |
A Family Study of Juvenile Obsessive-Compulsive Disorder |
||
|
Given the controversy regarding the contribution of familial and genetic factors in the expression of OCD as well as its relation to tic disorders, together with the findings that early-onset OCD may be more familial than adult-onset OCD, we undertook this study to determine whether childhood and adolescent OCD is familial and whether the rate of TS and tic disorders was higher among the relatives of persons with early-onset OCD, compared with those of control subjects. Material and Methods Thirty-five childhood- and adolescent-onset OCD probands
(aged 16 years or less) were ascertained through the Child and Adolescent
Psychiatric (CAP) services of the National Institute of Mental Health and Neuro
Sciences (NIMHANS), |
Procedure for Assessment of Relatives Informed consent was obtained from all the relatives who were assessed. All 250 relatives were contacted, but only 134 relatives (103 parents, 31 siblings) turned up for direct personal interviews. Information on 66 siblings was obtained by interviewing parents. For children, informed consent was obtained from parents. No relative refused to give consent, but, owing to practical problems, not all the relatives could be assessed. These problems included financial and other constraints that prevented relatives from travelling to the hospital, relatives' not getting leave from school and office, relatives' distance from the hospital, and failure to turn up for appointments. All the first-degree relatives aged over 17 years were evaluated using the Schedule for Clinical Assessment in Neuropsychiatry (SCAN) (27), Leyton's Obsessional Inventory (LOI) (28), Y-BOCS (29,30), and the Questionnaire for tic disorders (13). The diagnostic information about siblings aged 6 to17 years was obtained by administering DICA-R, CV-LOI, CY-BOCS, and the Questionnaire for tic disorders. For those siblings aged 17 years or less who could not be interviewed directly, the information was obtained by interviewing parents, using the DICA-R-Parent Version. No interviews were conducted by telephone. All the diagnostic information from the parents was obtained through a direct personal interview. No data were obtained about the unavailable adult relatives (parents), because the SCAN does not have the option to obtain family-history data. By contrast, we could get information about younger relatives (siblings) by using the Parent Version of DICA. All the interviews were conducted by a qualified psychiatrist. The information obtained from the relatives was analyzed, and consensual diagnoses were determined by 2 qualified psychiatrists according to DSM-III-R criteria (21). The raters were not, however, blind to the diagnosis of the probands. The diagnostic criteria for subthreshold OCD were as follows: the individual 1) met all criteria for OCD, except that symptoms were reported to occur for less than 1 hour per day; or 2) met all the criteria for OCD, except for ego dystonicity and insight; or 3) met all criteria for OCD, except for interference and distress. Data Analyses For estimation of morbid risk, Weinberg's method (31) was used. Morbid risk is defined as an estimate of the probability that a person will develop a given disease at some time during life if he or she survives the period of risk for the disease. Weinberg's calculation is as follows: Morbid risk (%) = [A / (A + U)] 100, where A = number of affected relatives, and U = age of adjusted well relatives. U = (U1 0) + (U2 1/2) + (U3 U1= Number of first-degree relatives who had not reached age of risk Next |
||
|
May 2001 |
A Family Study of Juvenile Obsessive-Compulsive Disorder |
||
|
U2 = Number of first-degree relatives who were in the period of risk U3 = Number of first-degree relatives who had covered the period of risk The denominator A + U is called Bezugsziffer (BZ) or lifetimes at risk. The period of risk employed for OCD was age 5 to 35 years (9). Statistical comparison of the risk between the 2 age-corrected groups of relatives was estimated using a proportion test (Z test). Results The sample included 35 OCD probands, 34 matched, unaffected control subjects, and 250 first-degree relatives. The demographic characteristics of the probands and their relatives are given in Table 1. Thirty-two OCD probands (91.42%) had both obsessions and compulsions, while 2 subjects (5.71%) had only obsessions, and the remaining 1 subject (2.86%) had predominantly compulsions. The OCD probands and unaffected control subjects did not differ significantly with regard to sex ratio, age, and years of education. The clinical characteristics of the OCD probands are given in Table 2. The information about psychiatric morbidity was obtained from 200 (80%) relatives out of a total of 250 first-degree relatives of both the probands and the comparison subjects. Of the 200 relatives interviewed, 134 (67%) were assessed in person. The details of the assessment of relatives are provided in Table 1. The psychiatric morbidity profiles of the first-degree relatives are summarized in Table 3. Four parents from 4 families of OCD probands had OCD, while none of the siblings received a diagnosis of OCD. OCD was not diagnosed in any of the relatives of unaffected control subjects. None of the first-degree relatives received a diagnosis of subthreshold OCD. Using the Weinberg's age-correction method, the BZ (number of relatives after age correction) for relatives of probands and control subjects was 80.5 and 74.5 respectively. |
The age-corrected morbid risk for OCD among the relatives of OCD probands was 4.96%. The relatives of OCD 22422p154w probands had significantly more risk for OCD, compared with those of control subjects (Z = 2.4; P = 0.05). None of the relatives of either probands or control subjects received a diagnosis of TS, although 20% of the probands had a comorbid tic disorder. Only 1 parent of an OCD proband had chronic motor tic disorder. Relatives of OCD probands also had high rates of anxiety and mood disorders, compared with those of control subjects (Table 3), who did not receive any diagnosis of anxiety or mood disorder. Discussion Our study had 3 main findings. First, an elevated risk for OCD (4.96%) existed in the relatives of the 35 probands with OCD. This diagnosis was not made for any relatives of the control subjects, which suggests that OCD is familial in at least some children and adolescents. Second, TS was absent in the relatives of probands. Third, subthreshold OCD was absent among the relatives of both probands and control subjects. A morbid risk of 4.96% in the relatives of OCD 22422p154w probands is higher than the estimated prevalence of 2% to 3% in the general population (32,33), and much higher than the estimated prevalence of approximately 1% reported in a recent review (34) that critically examined all the available data on the epidemiology of OCD. Our finding of elevated morbid risk in the relatives of juvenile OCD probands also assumes importance in the light of the finding of a previous study from our centre (20), which reported no evidence to support familiality in cases of adult-onset OCD. The obvious inference is that OCD in juveniles is perhaps a more severe form of illness, at least in terms of risk to relatives. The findings from our centre are also consistent with those of Bellodi and others (12) and Nestadt and others (16), who found evidence to support familiality only in cases of early-onset OCD. While the findings of our study do suggest that juvenile OCD is familial in at least some cases, the rates reported in previous studies are much higher (13% to 36%) (9-11,14,16), compared with the rate found in our study. There could be several reasons for such a difference. First, it is possible that some Next |
||
|
May 2001 |
A Family Study of Juvenile Obsessive-Compulsive Disorder |
||
|
forms of OCD may be genetically transmitted, while others are not. This heterogeneity of OCD may explain the high prevalence among relatives reported in previous studies (9,11,14). All 3 studies investigated only families of probands with severe primary OCD. Moreover, in the studies by Pauls and others (14) and Nestadt and others (16), 82% and 90% of the adult probands, respectively, had onset of illness prior to age 18 years. Possibly, the chronicity of OCD among probands in these studies was a significant determinant of the rate of reported familiality. If severity and chronicity are related to familiality, the resulting sample of families could be biased. By contrast, the OCD probands in our study were self-referred, moderately ill (mean CY-BOCS, 20.9), had a relatively shorter duration of illness (mean 22 months), and were ascertained from the general child psychiatric services. It has also been shown recently that phenotypic variability in OCD might be contributing to the variability in rates across studies (35 ). Second, it is also possible that families with multiple affected members are more likely to present to specialized clinics, which would have further affected the findings (14). Third, the low rate could be related to the method of interviewing. In the study by Pauls and others (14), only 39% of relatives were interviewed. In our study, the information on psychiatric morbidity was available for 80% of the relatives. Of the relatives interviewed, 67% were assessed in person. Many obsessive-compulsive traits and unaffected childhood rituals may be diagnosed as OCD when the history is obtained through relatives, resulting in an overestimation of OCD. Fourth, the low rate in our study could be due to the fact that OCD was diagnosed only in parents and not in siblings. It is evident that siblings are less likely than parents to have gone through the risk period for developing OCD. |
The relation between OCD and TS is far from clear. The findings of Leonard and others (11) and Pauls and others (14) suggest that some forms of OCD, particularly early-onset OCD, are related to TS. In addition, in the study by Pauls and others (14), a higher rate of tic disorders and TS among relatives of OCD probands was associated with a positive family history of OCD and a diagnosis of comorbid tic disorder. In our study, we did not find TS in relatives of either the OCD probands or the control group of children, suggesting lack of a genetic relation between OCD and TS. Our sample had 4 families with a positive family history of OCD, but none of them had any tic disorder. It is also interesting to note that the sample had 7 probands with comorbid tic disorders, but none of their relatives had a tic disorder. A family study from our centre of adults with OCD (20) also did not find evidence to support a familial relation between OCD and TS. Our finding is, however, consistent with that of Bellodi and others, who also did not find evidence to support an association between OCD and TS (12). The absence of association between OCD and TS assumes importance in the light of findings from other studies which suggest that OCD is a variant expression of the same factors that are important for the expression of TS and some chronic tic disorders (15,17,18). Several factors may account for the failure to find evidence to support a familial relation between OCD and TS. First, previous samples had subjects with severe OCD, whereas our sample was moderately ill and self-referred. If severity were to predict familiality and vice versa, the absence of TS in relatives of OCD probands is perhaps understandable, because the rate of OCD in relatives of OCD probands was itself very low in our study, compared with the findings of previous studies. It may be recalled here that Pauls and others (14) found a higher rate of tic disorders and TS among relatives of probands who had a family history of OCD than among those without such a family history. Second, is this lack of familial relation between OCD and TS a Next |
||
|
May 2001 |
A Family Study of Juvenile Obsessive-Compulsive Disorder |
||
|
true cross-cultural difference in the expression of the common factors that are important for manifestation of both OCD and TS? In other words, is the phenotypic expression in the form of TS uncommon in our setting? What factors determine the differential phenotypic expression of a possible common genotype is, however, a matter of great speculation. A higher rate of subthreshold OCD among the first-degree relatives of persons with OCD has been considered as further evidence in support of the hypothesis that OCD is familial (9,11,13,14). The findings of our study, however, are not consistent with those of previous studies, because none of the relatives had subthreshold OCD. If high familial rates of subthreshold OCD in the previous studies are also related to severity and chronicity of OCD in probands, then the absence of subthreshold OCD in the relatives of OCD probands in our study is understandable. For example, in the study by Pauls and others (14), 82 of the 100 adult OCD probands (mean age 33 years) had early-onset OCD and also more chronic illness. The relatives of these early-onset probands had a much higher rate of subthreshold OCD, compared with those of late-onset probands. Our study also found high rates of anxiety and mood disorders in the relatives of OCD 22422p154w probands. The finding is consistent with that of McKeon and Murray (37), who found high rates of anxiety disorders and depression in relatives of OCD probands, and with that of Black and others (13), who reported high rates of anxiety disorders in relatives of obsessional subjects. The increased rate of anxiety disorders in relatives of obsessional subjects and the findings of twin studies (38 ) prompted Black and others to suggest that a nonspecific anxiety disorder diathesis is transmitted in families with OCD (13). Our study, however, unlike those of McKeon and Murray (37) and Black and others (13), also reports elevated morbid risk for OCD in families with OCD, in addition to high rates of anxiety and mood disorders. OCD is considered a heterogenous condition (14): some cases are familial and unrelated to tic disorders, some familial and related to tic disorders, and others nonfamilial and unrelated to tic disorders. The findings of our study suggest that most juvenile forms of OCD are perhaps nonfamilial and unrelated to tic disorders, while only a few are familial. Our findings, however, need to be considered in the context of some limitations. First, the sample was small and recruited from a psychiatric hospital. Second, the raters were not blind to the proband status. Third, the relatives of psychiatrically unaffected control subjects may not have been representative of the general population. They may have been less likely to have mental disorders (as is the case in this study), enhancing the likelihood of finding differences between the case and control subjects. There is clearly a need to re-examine the issue of familiality in cases of OCD and its relation to TS, using larger samples recruited through community surveys. Samples collected from the community would be less likely to be affected by bias and thus provide better data for examining hypotheses about familial transmission and comorbidity. The findings of segregation analyses in a recent study (36) suggest that OCD is a genetically heterogenous condition. Thus, future studies need to examine the relation between phenotypic variability in OCD and its relation to familial risk, because this approach could facilitate the identification of the genetic mechanisms involved in the manifestation of OCD. |
Clinical Implications Most juvenile forms of OCD are perhaps nonfamilial and unrelated to tic disorders, while only a few are familial. The low rate of OCD among relatives of OCD probands in this study may be related to moderate symptomatology and relatively short duration of illness. The familiality
of OCD and its relation to tic disorders needs examination in larger
community samples, because such samples are less likely to be affected by
bias toward severe and chronic forms of illness. The sample size of OCD probands was relatively small. Relatives of psychiatrically unaffected control subjects may not be representative of the general population. The raters were not blind to the proband status. In addition, larger studies comparing families of OCD probands with and without tic disorders with families of tic-disorder probands and with control subjects would help further to clarify the relation between OCD and tic disorders. The issue that needs closer examination is the relation between the severity and chronicity of illness and its familiality, because the subjects in all the previous samples reporting higher rates were severely and chronically ill. It is possible that early-onset OCD with moderate symptomatology and relatively short duration of illness is less familial than the more severe and chronic form of early-onset OCD. To conclude, chronicity and severity may be important variables in the relation between phenotypic variability in OCD and familial risk and, further, in the relation between OCD and TS. References 1. Lewis A. Problems of obsessional illness. Proceedings
of the Royal Society of Medicine 1935;29:325-36. Next |
||
|
May 2001 |
A Family Study of Juvenile Obsessive-Compulsive Disorder |
||
|
9. Lenane MC, Swedo SE, Leonard H, Pauls DL, Sceery W, Rapoport JL.
Psychiatric disorders in first degree relatives of children and adolescents
with obsessive-compulsive disorder. J Am Acad Child
Adolesc Psychiatry 1990 -12.
|
24. Scahill L, Riddle MA,
Hardin M, Ort SI, King RA, Goodman WK, and others. Children's Yale-brown
Obsessive Compulsive Scale: reliability and validity. J Am Acad Child Adolesc Psychiatry
1997 -52. Back |
|
