Human Genetics in the Social Sciences, Department of Psychology, Hebrew. University, Jerusalem. This study was supportedin part by NIMH grants MH49351, ...
Obsessive-Compulsive Characteristics: From Symptoms to Syndrome ALAN APTER, M.D., THEODORE J. FALLON, JR., M.D., M.P.H., ROBERT A. KING, M.D., GIDI RATZONI, M.D., ADA H. ZOHAR, PH.D., MONICA BINDER, M.D., AVI WEIZMAN, M.D., JAMES F. LECKMAN, M.D., DAVID L. PAULS, M.D., SHMUEL KRON, M.D., AND DONALD J. COHEN, M.D.
ABSTRACT Objective: To assess the distribution and severity of obsessions and compulsions in a nonclinical adolescent population. Method: During preinduction military screening, 861 sixteen-year-old Israelis completed a questionnaire regarding the lifetime presence of eight obsessive-compulsive (OG) symptoms and three severity measures. The presence or absence of obsessive-compulsive disorder (OGD) or subclinical OGD was ascertained by an independent interview. Results: Although only 8.0% and 6.3% of respondents reported disturbing and intrusive thoughts, respectively, 27% to 72% of subjects endorsed the six remaining OGD symptoms. Twenty percent of subjects regarded the symptoms they endorsed as senseless and 3.5% found them disturbing; 8% reported spending more than an hour daily on symptoms. OGD and subclinical OGD cases differed significantly from non-OGD cases, but not from each other, in distress and mean number of symptoms. Although the distribution of nine of the items differed for noncases, compared with OGD and subclinical OGD cases, the distributions for all items overlapped markedly across the three groups. Conclusions: OG phenomena appear to be on a continuum with few symptoms and minimal severity at one end and many symptoms and severe impairment on the other. Defining optimal cutoff points for distinguishing between psychiatric disorder and OG phenomena that are common in the general population remains an open question. J. Am. Acad. Child Ado/esc. Psychiatry,
1996,35(7):907-912. Key Words: obsessive-compulsive disorder, adolescence.
There are few systematic studies of obsessions and compulsions in nonclinical child and adolescent populations. At one time, obsessive-compulsive disorder (OCD) was believed to be relatively uncommon in this age group, with an overall prevalence of about 0.05% (Elkins et al., 1980). In the Isle of Wight study, Rutter et al. (1970) found that only 0.3% of 10- to
Accepted August 18, 1995. Drs. Apter, Ratzoni, Binder, and Weizman are with Geha Psychiatric Hospital and the Division of Child and AdolescentPsychiatry, Seckler School ofMedicine, University of Tel Aviv. Dr. Kron is Head of the Mental Health Branch, Medical Corps, IsraelDefenseForce. Drs. Fallon,King, Pauls,Lechman, and Cohen are with the Yale Child Study Center, Yale University School of Medicine, New Haven, CT. Dr. Zohar is with the Scheinfeld Center for Human Genetics in the Social Sciences, Department of Psychology, Hebrew University, Jerusalem. This study was supportedin part by NIMH grants MH49351, MH00508, and NS16648. Reprint requests to Dr. King, Yale Child Study Center, 230 S. Frontage Road, P.O. Box 207900, New Haven, CT 06520-7900. 0890-8567/96/3507-0907$03.00/0© I 996 by the American Academy of Child and Adolescent Psychiatry.
12-year-olds had "prominent obsessive features" and none met the criteria for OCD. More recent studies, however, suggest higher prevalence rates. Flament et al. (1988), for example, found that although only 0.35% of high school students met the full criteria for OCD, as many as 2% of students reported obsessive preoccupations or behaviors that were numerous or interfered with their functioning. Few of these affected subjects had ever sought treatment. At follow-up 2 years later, these symptoms were found still to be present in most cases (Flament et al., 1990). In a previous study of 16- to 17-year-old Israeli adolescents, we found a point prevalence of 3.6% for OCD and an additional 1.25% of adolescents who reported obsessions and compulsions without substantial impairment or distress (Zohar et al., 1992). Similarly, Valleni- Basile et al. (1994) have recently reported a 3% point prevalence for OCD in a community sample of young adolescents. Studies of adult community samples also have found higher than previously anticipated prevalence rates for
J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 35:7, JULY 1996
907
APTER ET AL.
OCD. Levav et al. (1993) estimated a prevalence of 1.16% of probable OCD for young Israeli adults, aged 23 to 33 years. The multisite Epidemiologic Catchment Area study reported a lifetime prevalence for OCD of 1.2% to 2.4% (Karno et al., 1988). Methodological studies of the Epidemiologic Catchment Area samples, however, suggested lack of agreement between clinical and lay raters with respect to "caseness" (Helzer et al., 1985) . One potential reason that might account for this difficulty in defining "caseness" is that repetitive, intrusive, and senseless thoughts and urges are common experiences in adult nonclinical subjects (King and Noshpitz, 1991; Rachman and de Silva, 1978), suggesting the possible merits of a dimensional approach. Similarly, the frequent presence of mild obsessivecompulsive phenomena in nonclinical subjects suggests that these conditions might be best understood from a dimensional as well as categorical perspective (Thomsen and Jensen, 1991) . If a categorical approach is employed, the dilemma as to where to draw the cutoff points for the various severity criteria (resistance, distress, functional impairment, and/or time burden) becomes a major issue. DSM-IV revises the diagnostic criteria for OCD specified in DSM-III-R One major substantive change is the inclusion of subjects , under the specification "W ith Poor Insight," who lack insight into the excessiveness or unreasonableness of their symptoms for most of the current episode. This change was adopted to permit the diagnosis of severely disturbed, perhaps quasi-delusional patients whose symptomatology and pharmacological treatment might best be addressed as part of the OCD spectrum. This change, however, complicates the classification of subjects at the other end of the severity spectrum, such as those found in community or family genetic studies . These individuals may have obsessions or compulsions that are timeconsuming, or even impairing, but they do not view their symptoms as clearly senseless. Because one goal of this study was to examine the distribution of symptoms and severity (including perceived senselessness) in community populations, the decision was made to use the DSM-IV criteria for OCD, modified to exclude those subjects who would be classified in DSM-IV as "With Poor Insight." All of these considerations point to the difficulty of developing valid instruments for studying obsessivecompulsive phenomena in community samples. To
908
delineate these issues more clearly, the current study was designed to determine the distribution of obsessive and compulsive phenomena in a nonclinical population-based adolescent sample and to determine the relationship of these phenomena to the diagnosis of OCD. METHOD Sample The sample consisted of 861 consecutive inductees (436 males and 425 females) to the Israeli Defense Force, drawn from a central induction center. All Jewish males and most Jewish females between the ages of 16 and 17 are screened by the Israeli Defense Force induction centers in preparation for conscription (Apter et al., 1993). Fewer than 5% of girls are exempt on religious grounds. Thus, the induction centers screen more than 95% of a complete national cohort of 16- to 17-year-old adolescents.
Procedure All inductees completed an l l-Itern self-report questionnaire consisting of the screening questions for OCD of the Hebrew version of the Schedule for Touretre's Syndrome and Other Behavioral Syndromes (H-STSOBS) (Pauls and Zohar, 1991). (Further description of the questionnaire and its psychometrics is available from the authors.) Eight of the 11 items were yes-orono questions about the lifetime occurrence of the following obsessions and compulsions: disturbing thoughts or images, intrusive thoughts or images, repetitiveactions, urge to repeat, ritualized routines, extreme neatness, orderliness, and hoarding. Three additional items related to the severity of endorsed items: time spent per day on all of the symptoms , perceived senselessness of the symptoms , and distress caused by the symptoms. After completion of the questionnaire, each subject was interviewed by an experienced child and adolescent psychiatrist (G.R. or M.B.) who was blind to the individual's responses on the questionnaire. This 20-minute interview used selected questions from the H-STSOBS, a structured psychiatric interview with a detailed section on OCD (see Pauls et al., 1995, for details). On the basis of this interview the presence or absence of OCD was diagnosed using DSM-N criteria, modified to exclude the category "With Poor Insight." A diagnosis of subclinical OCD was made for individuals who met the DSM-N criteria for the presence of obsessions or compulsions but who failed to meet one of the severity criteria of distress, perceived senselessness, or duration of more than 1 hour. Reliability was assessed in a subsample of 50 subjects whose interviews were observed by the noninterviewing psychiatrist (G.R. or M.B.). On the basis of this interview, the interviewing and non interviewing psychiatrists independently rated each subject for the presence or absence of OCD or subclinical OCD. The agreement between raters was excellent (lC ~ .9.)
Data Analysis The OCD, subclinical OCD, and non-OCD cases were compared using a one-way analysis of variance. Where significant differences were found , post hoc pairwise comparisons of the three groups were performed using Bonferroni correction to account for multiple comparisons.
J. AM . ACAD. CHILD ADOLESC. PSYCHIATRY, 35 :7, JULY 1996
OBSE SSIONS AND COMPULSIONS IN ADOLESCENCE
TABLE 1 OCD Symptoms in a General Adolescent Population Pairwise Comparisons P < .05 (Bonferroni Correction)
Proportion of Population Endorsing Each Symptom Total Non -OCD Population Cases (N = 861) (n = 807)
Question 1. 2. 3. 4. 5. 6. 7. 8.
Disturbing thoughts Intrus ive images Repetitive actions Urge to repeat Ritualized routines Extreme nearness Orderliness Hoarding
Subclinical Cases (n = 34)
OCD Non-OCD Non-OCD Cases Probability vs. vs. (n = 20) (FTest) Subclinical OCD
0.080 0.063 0.27 0.30 0.34 0.72 0.49 0.29
0.063 0.056 0.23 0.28 0.33 0.70 0.46 0.27
0.32 0.15 0.71 0.41 0.53 0.97 0.74 0.56
0.35 0.20 0.75 0.79 0.60 0.90 0.80 0.35
