KISEP

Original Article

Clinical Psychopharmacology and Neuroscience 2005; 3: 38-42

Dimensional Analyses of the Yale-Brown Obsessive-Compulsive Scale and Yale-Brown Obsessive-Compulsive Scale Checklist Se Joo Kim1, Hyun Joo Hong2, Hong Shick Lee1, Chan-Hyung Kim1 1

Department of Psychiatry, Yonsei University College of Medicine, Seoul, 2Department of Psychiatry, Hallym University College of Medicine, Chuncheon, Korea

Although obsessive-compulsive disorder (OCD) has long been a unitary diagnosis, recently there has been much interest in its potential heterogeneity, as manifested by symptom subgroups. Although the Yale-Brown Obsessive-Compulsive Scale (YBOCS) is a widely used instrument to assess obsessive-compulsive symptomatology, the variables reflecting the pathogenesis of OCD and Y-BOCS subscores were not usually significantly associated. The aims of this study are to clarify the identification of the dimensional structure of the symptoms included in the Y-BOCS checklist and to explore the factor structure of the YBOCS. Ninety five OCD patients participated in this study and performed the Y-BOCS and Y-BOCS checklist. The 13 main symptom categories included in the Y-BOCS checklist and 10 items included in the Y-BOCS were factor analyzed using principal components analysis. Using principal component analysis, we derived 4 factors from the 13 main contents of the YBOCS checklist. Four factors, viz. hoarding/repeating, contamination/cleaning, aggressive/sexual, and religious/somatic, accounted for more than 60% of the variance. We also derived 3 factors from the 10 items of the Y-BOCS, viz. severity of obsession, severity of compulsion, and resistance to symptoms, which accounted for more than 70% of the variance. Four symptom dimensions from the Y-BOCS checklist and three symptom dimensions from the Y-BOCS were identified as significant factors accounting for the variance. These factors may be of value in future genetic, neurobiological, and treatment response studies. ：Dimensional analyses; Yale-brown obsessive-compulsive scale; OCD. KEY WORDS：

INTRODUCTION

bidity with other conditions, such as tic-related disorders5 and other demographic and clinical variables. Others have proposed subtypes based on phenomenological or thematic consistencies.6 Although several of these subtype models show promise, supporting empirical evidence, to date, has been lacking or inconsistent.3 Recently, the most common method of identifying subgroups of patients is to use the symptoms manifested by OCD patients as classifying variables, and such subgroups can be hypothesized to be related to treatment response, biological markers, or genetic transmission in OCD. In clinical practice, different prognoses and treatment outcomes could be associated with different obsessive-compulsive (OC) contents. Erzegovesi et al.7 reported that OCD patients with somatic obsession were less responsive to medications, while Mataix-Cols et al.8 reported that patients with hoarding obsession and compulsion were less responsive to medications. This suggests that specific OC contents may characterize distinct OCD subtypes with different biological backgrounds.9 Recently, there have been several studies which used statistical methods to reformulate the structure of OC symptoms as ascertained

Obsessive-compulsive disorder (OCD) has long been regarded as a coherent constellation of symptoms, that is a unitary disorder.1 However, the symptoms used to define OCD are diverse and include various intrusive thoughts, preoccupations, rituals, and compulsions. Two individuals with OCD may have totally different symptoms and non-overlapping symptom patterns.2 Therefore, allusions to the possibility that OCD is not a homogenous diagnostic entity have become increasingly common in recent years and many studies have attempted to classify the complex structures of OCD symptoms into more simple subtypes.3 Some investigators have identified subtypes of OCD based on the age of onset, gender,4 comorAddress for correspondence: Chan-Hyung Kim, MD, PhD. Department of Psychiatry, College of Medicine, Yongdong Severance Hospital, Yonsei University, Dogok-dong, Gangnam-gu, Seoul 135-720, Korea Tel: +82-2-2019-3340, Fax: +82-2-3462-4304 E-mail: [email protected] This article is an English version of the one published in Korean J Psychopharmacol 2004:15:339-345

38

Dimensional Analyses of Yale-Brown Obsessive-Compulsive Scale·39

by the widely recognized and exhaustive symptom checklist of the Yale-Brown Obsessive-Compulsive Scale (YBOCS).10-12 However, most of these studies were published in other countries and there has been no such study of Korean OCD patients. Therefore, identifying a specific dimensional factor structure of OC symptoms in Korean OCD patients might be useful for future studies and clinical practice, and might help us to put aside the current conventional set of diagnostic criteria that still classifies OCD as a unitary nosographic entity. The Y-BOCS checklist, which corrected the selection bias of the Mausley Obsessive Compulsive Inventory, contains more than 50 items of obsessions and compulsions designed to assist in the measure of the Y-BOCS.13 It is obvious that the Y-BOCS checklist is a reliable instrument for evaluating the contents of OC symptoms, but is less valid to assess the severity of OC symptoms. The Y-BOCS is a gold standard assessment scale used to evaluate the severity of OC symptoms. The use of both instruments would produce not only qualitative but also quantitative data. Although the Y-BOCS is considered valid to grade the severity of OCD, investigators remain divided in their opinions about the validity of subscales consisting of the severity of obsessive thoughts (1-5 items) and the severity of compulsive behaviors (6-10 items) in the model of Goodman et al.14 Indeed, many studies reported that the independent measures of the factors representing the pathophysiology of OCD were not significantly correlated with Y-BOCS subscales.15,16 Also, the validity of items 4 and 9 has been a subject of debate among investigators.14 To minimize the hassles involved in interpreting the validity of the Y-BOCS, it is essential to verify its factorial structure. To date, there have been several studies that verified the factor structure of Y-BOCS, but the results were inconclusive. Moreover, there have been no such studies of Korean OCD patients. Therefore, the purpose of this study was to demonstrate the structural validity of the Y-BOCS checklist and the YBOCS by analyzing the factors included in the models, because these are the two most useful instruments for assessing OCD symptoms, both qualitatively and quantitatively.

went a face-to-face Structured Clinical Interview17 for the diagnosis of OCD and comorbidities. Of the 95 subjects, 43 had previously been treated for OCD and 38 were currently taking medications and/or undergoing psychotherapy for OCD. However, the remaining 57 patients had either never received any kind of treatment for their OCD symptoms or had not received any treatment for the last six months. Clinical Assessment Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) The 10-item Yale-Brown scale is designed to measure time spent on, interference from, distress over, the amount of resistance to and control over obsessions in items 1 to 5, and to assess the same dimensions in relation to compulsions in items 6 to 10. The questionnaire is supposed to be administered by an expert clinician in a semistructured interview. The severity of obsessions and compulsions is rated on a scale of 1 to 4, where 0 indicates “no symptoms” and 4 indicates “extreme symptoms”.18

MATERIALS AND METHODS

Y-BOCS Checklist The Y-BOCS checklist provides a comprehensive list of obsessions and compulsions that comprises more than 60 items divided into 15 categories based on mental and behavioral phenomena. Among them, eight categories include aggressive, contamination, sexual, hoarding, religious, symmetry and somatic content that correspond to obsessive thoughts. The other seven categories include cleaning, checking, repeating, counting, ordering and hoarding variables related to compulsive behaviors. For each item, the patient can give a yes or no answer by ascertaining current and past obsessions and compulsions.13 Each item was rated on a 0-2 score in accordance with the method described by Leckman et al.12 as follows: 1 was assigned to the category if any subset within the category not representing a principal or major problem was answered by “yes”, irrespective of whether or not it is a current or past obsession or compulsion. A score of zero (0) was assigned to the category if every subset was answered by “no”. A score of 2 was assigned to the category if any subset that indicates a principal or major problem was answered by “yes”.

Subjects A total of 95 patients who presented with OCD symptoms at the department of psychiatry of Yonsei University Yongdong Severance Hospital and Hallym University Sacred Heart Hospital were recruited after obtaining informed consent from each patient. All subjects under-

Statistical Analysis A principal component analysis was performed on 13 categories from the Y-BOCS checklist, excluding the two categories of miscellaneous obsessions and compulsions. And another principal component analysis was done on items of the Y-BOCS using the observed scores. Those

40·CLINICAL PSYCHOPHARMACOLOGY AND NEUROSCIENCE 2005; 3: 38-42

factors with an eigenvalue of greater than 1 were extracted following Kaiser’s criterion, and varimax rotation was employed to preserve the independence of the factors and make the interpretation easier.

RESULTS Social, Demographic and Clinical Data of Subjects A total of 95 patients (65 males and 30 females) with OCD with a mean age of 29.46±9.25 years participated in the study. The mean age at the onset of OCD was 19.93±8.61 years. The average duration of illness was 9.92±8.45 years. The mean Y-BOCS score for all patients was 27.32±5.63 points Factor Analysis on the Y-BOCS Symptom Checklist The principal components analysis of the Y-BOCS symptom checklist suggested a four-factor solution. The Keyser-Meyer-Olkin measure of sampling adequacy was .78, which indicates an acceptable pattern of correlations among variables.19 Varimax rotation was applied to obtain unrelated factors. Only those items which have factor loadings greater than 0.5 comprising the factors were retained. Factor 1 was named hoarding/repeating. This factor, which included items of obsession with hoarding and the compulsions of counting, repeating, hoarding and ordering, explained 33.99% of the total variance. Factor 2 was named contamination/cleaning. Factor 2, which included items of obsession with contamination and the compulsion of cleaning, explained 11.12% of the total variance. Factor 3 was labeled aggressive/sexual. Factor 3, which covered aggressive and sexual obsessions, explained 9.96% of the total variance. Factor 4 was labeled religious/somatic. This factor, which covered religious and somatic obsessions, explained 7.70% of the total variance. Altogether, the four factors accounted for 62.77% of the total variance (Table 1). Factor Analysis of Y-BOCS Principal component analysis was applied to the YBOCS on the item level and yielded a three-factor solution. The Keyser-Meyer-Olkin test of sampling adequacy gave a result of 0.84. The three-dimensional model was comprised of severity of obsession (factor 1), severity of compulsion (factor 2) and resistance to symptoms (factor 3). The obsession factor included Y-BOCS items 1, 2, 3 and 5, and explained 13.25% of the total variance. The compulsion factor included Y-BOCS items 6, 7, 8 and 10, and explained 52.02% of the total variance. The resistance to symptom factor included Y-BOCS items 4 and 9, and explained 10.02% of the total variance. Alto-

Table 1. Four factors solution for Y-BOCS checklist Factor 2 Factor 1 (Contami(Hoarding/ nation/ repeating) cleaning)

Factor 3 (Aggressive/ sexual)

Factor 4 (Religious/ somatic)

Obsessions Aggressive

0.72

Contamination

0.71

Sexual Hoarding

0.84 0.55

Religious

0.77

Symmetry Somatic

0.83

Compulsions Cleaning

0.81

Checking Repeating

0.76

Counting

0.83

Ordering

0.54

Hoarding

0.62

% of explained 34.00 11.12 9.96 variance Factor loadings>0.5. Principal components analysis

7.70

Table 2. Three factors solution for Y-BOCS Factor 1 Factor 2 Factor 3 (Severity of (Severity of (Resistance to obsessions) compulsions) symptoms) 1. Time on obsessions 0.81 2. Interference from 0.74 obsessions 3. Distress from 0.84 obsessions 4. Resistance to obsessions 5. Control over 0.72 obsessions 6. Time on 0.75 compulsions 7. Interference from 0.79 obsessions 8. Distress from 0.73 obsessions 9. Resistance to obsessions 10. Control over 0.91 obsessions % of explained 52.02 13.25 variance Factor loadings>0.5, Principal components analysis

0.90

0.76

10.02

gether, the three-factor solution accounted for 75.29% of the total variance (Table 2).

DISCUSSION The principal component analysis of the data collected from the 95 patients using the Y-BOCS checklist suggested a four-factor solution comprising hoarding/repea-

Dimensional Analyses of Yale-Brown Obsessive-Compulsive Scale·41

ting, contamination/cleaning, aggressive/sexual and religious/somatic factors. These findings are consistent with the potential multidimensional models suggested in other studies.8,10 Factor 1 included items of obsession with hoarding and the compulsions of counting, repeating, hoarding and ordering. This result was similar to that of a previous report by Baer et al., in which obsessions with hoarding and symmetry and compulsions with hoarding, ordering, repeating, and counting were classified into one factor.10 There have been other studies which reported similar results, but some of them classified obsession with hoarding and the compulsion of hoarding as another separate factor.8,12 Bear et al.10 stated that the symptoms loading on the hoarding/repeating factor are reminiscent of Janet’s description of patients who were tormented by an inner sense of imperfection and felt that their actions were never completely achieved to their satisfaction.20 Baer et al.10 and Leckman et al.12 reported that OCD patients with tic disorders showed a higher score for the hoarding/repeating factor and suggested that this factor is associated with the feeling of incompleteness shown in patients with tic disorder or trichotillomania. However, unlike in the study of Baer et al., our study revealed that obsession with symmetry was not included in any of the four factors. The contamination/cleaning factor (factor 2) included items of obsession with contamination and the compulsion of cleaning. The combination of these two items is commonly supported not only by clinically determined categories, but also by evidence suggested in earlier studies.8-12 The aggressive/sexual factor (factor 3) covered items corresponding to aggressive and sexual obsessions. The religious/somatic factor (factor 4) included items corresponding to religious and somatic obsessions. The last two factors are composed of pure obsessions. Leckman et al.12 and Cavallini et al.9 put aggressive, sexual, religious and somatic obsessions into a single factor. There have been other studies in which sexual and religious obsessions were classified into a single factor8,11 or in which aggressive, sexual and religious obsessions were combined into a single factor.10 This factor is similar to the final stage of psychastenia described by Janet, in which patients have obsessions involving forbidden thoughts and acts of a sacrilegious, violent, or sexual nature.20 The items of obsessions and compulsions included in factors vary slightly among studies, reflecting the differences associated with the sample size, social and cultural background and research method. Among the recent studies in which a factor analysis of OCD symptoms was conducted, Mataix-Cols et al. said that some obsessions and compulsions in OCD patients remained relatively unchanged and steady when their symp-

toms were closely tracked for two years. Even symptoms that may change over time showed waxing and waning within the symptom dimensions of the Y-BOCS checklist, rarely involving shifts between dimensions.21 These findings highlighted the importance of the multidimensional assessment of OCD. At the same time, the multidimensional approach to assessing OCD symptoms is likely to give a boost to neurobiological research and treatment. A factor analysis performed on the individual items of the Y-BOCS produced a three-dimensional model comprising severity of obsession, severity of compulsion and resistance to symptoms. Among the existing studies in which a factor analysis of the Y-BOCS was conducted, Fals-Stewart22 put all 10 items into a single factor along with six additional items whose scores did not add up to the Y-BOCS total score. McKay et al. came up with a two-factor solution covering obsessive thoughts and compulsive behaviors.23 Amir et al. also suggested a two-factor solution involving the disturbance factor (items 2, 3, 7 and 8) and severity of symptoms (items 1, 4, 5, 6, 9, 10).24 Kim et al. reported the same three-factor solution as that identified in our study. Moritz et al.14 classified items 1, 2 and 3 into severity of obsessions, items 6, 7 and 8 into severity of compulsions and items 4 and 9 into resistance to symptoms. They stated that items 5 and 10 showed mixed loadings on the corresponding severity factor and the common resistance factor, and that it appears that symptom control reflects a joint function of symptom severity and symptom resistance, thus decreasing the homogeneity of the subscales when they are incorporated. However, in our study, we identified items 5 and 10 as being included in severity of obsessions and the compulsions dimension, respectively, rather than in resistance to symptoms. Thus, the results of our study are in agreement with the findings of the previous studies conducted by Kim et al.25 and Moritz et al.,14 providing evidence that the Y-BOCS is best represented by a multidimensional model. In detail, resistance to symptoms is classified as a separate factor in addition to the two distinct dimensions-obsessions and compulsions. As the three-factor model showed better fit, it is possible that assessing OCD symptoms with the total Y-BOCS score or the two-factor models specifying obsessions and compulsions initially proposed by Goodman et al. would lead to a misinterpretation of the severity of symptoms. When an OCD patient is characterized by severe obsessions, mild compulsions and strong resistance to OCD symptoms, the assessment based only on the total YBOCS score would raise the possibility of this pattern being overlooked and may give the false impression that the symptoms are not severe. Therefore, it is important

42·CLINICAL PSYCHOPHARMACOLOGY AND NEUROSCIENCE 2005; 3: 38-42

to assess OCD symptoms using the three-factor model comprising severity of obsessive thoughts, severity of compulsive behaviors and resistance to symptoms, as well as the total score. This multidimensional (severity of obsessions, severity of compulsions, and resistance to symptoms) approach should prove its effectiveness in research studies aimed at identifying the biological causes of OCD and treatment responses to various therapies. This study has several limitations. First, the sample size is smaller than those of other studies. Secondly, the English version of Y-BOCS and Y-BOCS checklist used in this study were translated into Korean by the authors, because there are no standardized Korean versions. Therefore, there could be some problems concerning the validity and reliability of the Y-BOCS and Y-BOCS checklist used in our study. To ensure more reliable results, further studies are needed with a standardized instrument and a larger sample size. Future studies in the areas of genetics, neuroimaging, and neuropsychology are expected to reveal the neural elements corresponding to the symptoms identified as factors from the Y-BOCS checklist. When these neural elements are identified, OCD would be seen as a heterogeneous group of disorders rather than a single disease. Thus, the present study is intended to serve as a cornerstone for guiding further studies in this area of research. The three-factor structure, identified by the Y-BOCS factor analysis in this study, should aid other investigators to understand the severity of OCD symptoms and the differences between the responses of these factors following treatment. REFERENCES 1. Berrios GE. Obsessive-compulsive disorder. Its conceptual hi-

story in France during the 19th century. Compr Psychiatry 1989; 30:283-95. 2. Rasmussen SA, Eisen JL. Clinical and epidemiological findings of significance to neruopharmacological trials. Psychopharmacol Bull 1988;24:466-70. 3. Summerfeldt LJ, Richter MA, Antony MM, Swinson RP. Symptom structure in obsessive-compulsive disorder: a confirmatory factor-analytic study Behav Res Ther 1999;37:297-311. 4. Lensi P, Cassano GB, Correddu G, Ravagli S. Obsessive-compulsive disorder: familial-developmental history, symptomatology, comorbidity and course with special reference to genderrelated differences. Br J Psychiatry 1996;169:101-7. 5. Zohar AH, Pauls DL, Ratzoni G, Apter A, Dycian A, Binder M, et al. Obsessive-compulsive disorder with and without tics in an epidemiological sample of adolescents. Am J Psychiatry 1997; 154:274-6. 6. Rasmussen SA, Eisen JL. The epidemiology and clinical features of obsessive-compulsive disorder. Psychiatr Clin North Am 1992;15:742-58.

7. Erzegovesi S, Cavallini MC, Cavedini P, Diaferia G, Locatelli

M, Bellodi L. Clinical predictors of drug response in obsessive-compulsive disorder. J Clin Psychopharmacol 2001;21: 488-92. 8. Mataix-Cols D, Rauch SL, Manzo PA, Jenike MA, Baer L. Use of factor-analyzed symptom dimensions to predict outcome with serotonin reuptake inhibitors and placebo in the treatment of obsessive-compulsive disorder. Am J Psychiatry 1999;156: 1409-16. 9. Cavallini MC, Di Bella D, Siliprandi F, Malchiodi F, Bellodi L. Exploratory factor analysis of obsessive-compulsive patients and association with 5-HTTLPR polymorphism. Am J Med Genet 2002;114:347-53. 10. Baer L. Factor analysis of symptom subtypes of obsessive compulsive disorder and their relation to personality and tic disorders. J Clin Psychiatry 1994;55(Suppl):18-23. 11. Feinstein SB, Fallon BA, Petkova E, Liebowitz MR. Item-byitem factor analysis of the Yale-Brown Obsessive Compulsive Scale Symptom Checklist. J Neuropsychiatry Clin Neurosci 2003; 15:187-93. 12. Leckman JF, Grice DE, Boardman J, Zhang H, Vitale A, Bondi C, et al. Symptoms of obsessive-compulsive disorder. Am J Psychiatry 1997;154:911-7. 13. Goodman WK, Price LH, Rasmussen SA, Mazure C, Delgado P, Heninger GR, et al. The Yale-Brown Obsessive Compulsive Scale. II. Validity. Arch Gen Psychiatry 1989;46:1012-6. 14. Moritz S, Meier B, Kloss M, Jacobsen D, Wein C, Fricke S, et al. Dimensional structure of the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS). Psychiatry Res 2002;109:193-9. 15. Moritz S, Birkner C, Kloss M, Jacobsen D, Fricke S, Bothern A, et al. Impact of comorbid depressive symptoms on neuropsychological performance in obsessive-compulsive disorder. J Abnorm Psychol 2001;110:653-7. 16. Moritz S, Fricke S, Wagner M, Hand I. Further evidence for delayed alternation deficits in obsessive-compulsive disorder. J Nerv Ment Dis 2001;189:562-4. 17. Spitzer RL, Williams JB, Gibbon M, First MB. The Structured Clinical Interview for DSM-III-R (SCID). I: History, rationale, and description. Arch Gen Psychiatry 1992;49:624-9. 18. Goodman WK, Price LH, Rasmussen SA, Mazure C, Fleischmann RL, Hill CL, et al. The Yale-Brown Obsessive Compulsive Scale. I. Development, use, and reliability. Arch Gen Psychiatry 1989;46:1006-11. 19. Kaiser HF. An index of factorial simplicity. Psychometrika 1974; 39:31-6. 20. Pitman RK. Pierre Janet on obsessive-compulsive disorder (1903). Review and commentary. Arch Gen Psychiatry 1987; 44:226-32. 21. Mataix-Cols D, Rauch SL, Baer L, Eisen JL, Shera DM, Goodman WK, et al. Symptom stability in adult obsessive-compulsive disorder: data from a naturalistic two-year follow-up study. Am J Psychiatry 2002;159:263-8. 22. Fals-Stewart W. A dimensional analysis of the Yale-Brown Obsessive Compulsive Scale. Psychol Assess 1992;70:239-40. 23. McKay D, Danyko S, Neziroglu F, Yaryura-Tobias JA. Factor structure of the Yale-Brown Obsessive-Compulsive Scale: a two dimensional measure. Behav Res Ther 1995;33:865-9. 24. Amir N, Foa EB, Coles ME. Factor structure of the Yale-Brown Obsessive-Compulsive Scale. Psychol Assess 1997;9:312-6. 25. Kim SW, Dysken MW, Pheley AM, Hoover KM. The YaleBrown Obsessive-Compulsive Scale: measures of internal consistency. Psychiatry Res 1994;51:203-11.