- Email: [email protected]
The clinical utility of symptom dimensions in obsessive–compulsive disorder
Psychiatry Research 180 (2010) 25–29
Contents lists available at ScienceDirect
Psychiatry Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s
The clinical utility of symptom dimensions in obsessive–compulsive disorder Hisato Matsunaga a,⁎, Kazuhisa Hayashida a, Nobuo Kiriike a, Kensei Maebayashi a, Dan J Stein b,c a b c
Department of Neuropsychiatry, Osaka City University Medical School, Osaka, Japan MRC Unit on Anxiety Disorders, University of Cape Town, Cape Town, South Africa Department of Psychiatry, Mt. Sinai Medical School, New York, USA
a r t i c l e
i n f o
Article history: Received 27 April 2009 Received in revised form 3 September 2009 Accepted 11 September 2009 Keywords: Obsessive–compulsive disorder Symptom dimension Yale–Brown Obsessive–Compulsive Scale (Y-BOCS) Clinical utility
a b s t r a c t Factor analyses in obsessive–compulsive disorder (OCD) have consistently identified several different symptom dimensions. Nevertheless the clinical utility of identifying such symptom dimensions remains somewhat unclear. On the basis of their principal symptoms, 343 OCD patients were divided into four symptom dimension subgroups; 1) contamination/washing, 2) hoarding, 3) symmetry/repeating and ordering, and 4) forbidden thoughts/checking. Clinical variables including 1-year treatment outcome were compared across these patient subgroups. Most patients (74%) could distinctively be categorized as falling into a particular symptom subgroup. The groups were differentially characterized by some demographic and clinical features. For instance, both the symmetry and hoarding groups were significantly associated with decreased global functioning and greater OCD severity. Moreover the hoarding group was significantly more likely than the others to show longer duration of illness, lower rate of marriage, poor insight, and poorer outcome. However, about a quarter of the participants could not be classified definitively into a particular group. Our findings provide partial support for the clinical utility of a simple measure of symptom dimensions in OCD. In clinical settings, however, the limitations of such a simple measure of predominant symptom dimensions should be borne in mind and further work on their validity and utility is needed. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction There is increasing evidence that obsessive–compulsive disorder (OCD) is a heterogeneous condition (Lochner and Stein, 2003; MataixCols et al., 2005; Mathis et al., 2006; Bloch et al., 2008). If OCD subtypes were characterized by a distinct psychobiology, then this would account for a variance in studies on biological markers, and would potentially impact on treatment strategies. For example, it has been suggested that some forms of OCD are etiologically related to tic disorders (Eichstedt and Arnold, 2001; Mathis et al., 2006), and OCD with comorbid tics has been characterized by specific phenomenological, genetic, and neuro-imaging features and a differential treatment response (McDougle et al., 1994; Eichstedt and Arnold, 2001; Grados et al., 2001; Mathis et al., 2006). Recent factor and cluster analyses have provided consistent evidence that distinct obsessive–compulsive symptom dimensions exist, including forbidden thoughts, contamination/washing, hoarding and symmetry/repeating rituals and ordering (Baer, 1994; Leckman et al., 1997; Calamari et al., 1999; Mataix-Cols et al., 1999; Summerfeldt et al., 1999; Bloch et al., 2008). These symptom dimensions have been differentially related to clinical features such ⁎ Corresponding author. Department of Neuropsychiatry, Osaka City University Medical School, 1-4-3 Asahi-machi, Abeno-ku, Osaka, 545-8585, Japan. Tel.: +81 6 6645 3821; fax: +81 6 6636 0439. E-mail address: [email protected] (H. Matsunaga). 0165-1781/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2009.09.005
as comorbidity (Baer, 1994; Leckman et al., 1997; Hasler et al., 2005; Mataix-Cols et al., 2005). Early-onset or tic-related OCD is a unique subtype characterized by more symmetry/ordering compulsions (Mathis et al., 2006), and OCD patients with hoarding tend to exhibit specific and disabling clinical features such as a lower level of global functioning and greater overall illness severity as well as poorer response to serotonin reuptake inhibitors (SRIs) (Mataix-Cols et al., 1999; Steketee and Frost, 2003; Mataix-Cols et al., 2005; Saxena, 2007; Matsunaga et al., 2008). It has also been hypothesized that each symptom dimension may be underpinned by a distinctive set of biobehavioral mechanisms (Mataix-Cols et al., 2004; Saxena et al., 2004). Data that OCD subtypes and symptom dimensions are stable across different cultures and ethnic groups (Bloch et al., 2008; Matsunaga et al., 2008), support the hypothesis that OCD is mediated by universal psychobiological mechanisms. Nevertheless, it remains unclear whether classifying patients as suffering predominantly from one or other symptom dimension has clinical utility. In research settings, it is possible to use instruments such as the Dimensional Yale–Brown Obsessive–Compulsive Scale (DYBOCS) (Rosario-Campos et al., 2006). In clinical settings, only a simple assessment of the predominant symptom dimension such as “OCD with prominent contamination obsessions and cleaning compulsions” may be practical (Leckman et al., 2007). It is unclear as to whether such a measure is always possible, and whether it would usefully inform clinical practice. Based on the previous findings (Mataix-Cols et al., 1999; Mataix-Cols et al., 2002b; Steketee and Frost, 2003; Mataix-Cols et
26
H. Matsunaga et al. / Psychiatry Research 180 (2010) 25–29
al., 2005; Saxena, 2007; Landeros-Weisenberger et al., 2010), it can be hypothesized that the predominance of hoarding symptoms would be associated with poorer treatment response and social disability, and forbidden thoughts might be associated with positive response to SRIs. In this study, we sought to examine 1) whether a simple clinical measure of predominant symptoms was consistent with the data from a more complex principle component factor analysis, and 2) whether such a measure is useful in informing practice, insofar as it is associated differentially with particular clinical characteristics of OCD. 2. Methods Subjects were 343 outpatients who met Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV; APA, 1994) criteria for OCD, consecutively admitted to our hospital between April, 2005 and December, 2007 and gave us informed consent to participate in this study. Each subject was diagnosed with the Structured Clinical Interview for DSM-IV Patient version (SCID-P) (First et al., 1997), received the standardized combination treatments for at least 1 year, and had participated in our recent study (Matsunaga et al., 2008). Information was obtained regarding demographic profile, family and medical history, and clinical features and course. Global functioning was assessed using the DSM-IV Axis V Global Assessment of Functioning Scale (GAFS) (APA, 1994). The involvement of others – i.e. whether OCD subjects involved others in their symptoms by forcing them to help with their rituals, or asking them for reassurance, or to perform ritualistic behaviors similar to their own – (Matsunaga et al., 2000) was also assessed in each subject. The procedure to assess the lifetime incidence of impulsive behaviors such as substance abuse, stealing, selfmutilation, suicide attempts, and destructive behaviors has been described in our previous study (Matsunaga et al., 2005). Ratings of OCD symptoms were obtained with the Japanese version of the Y-BOCS Symptom Checklist (Goodman et al., 1989a,b; Nakajima et al., 1995). The assessment was conducted in a semi-structured interview by two clinicians with extensive training in and experience with the scale. Up to three primary obsessions and compulsions were listed for each subject, and the principal obsession and compulsion symptom categories were determined. Each of the 13 major Y-BOCS Symptom Checklist categories except for miscellaneous obsessions and compulsions, was scored as 0 (absent), 1 (present), or 2 (predominant). The Y-BOCS was then administered to assess current OCD symptom severity. Degree of insight in each subject was evaluated using the insight question of the Y-BOCS (Goodman et al., 1989a,b). The assessment procedure and inter-rater reliability have also been reported (Matsunaga et al., 2002). After the pre-treatment assessments, each subject was treated with a standardized combination of SRIs, (i.e., clomipramine (CMI), fluvoxamine (FLV), or paroxetine (PXT)) and cognitive–behavioral therapy (CBT). Treatment was usually initiated with daily dosages of CMI or FLV 50 mg, or PXT 10 mg. If the medication was tolerated, dosage was gradually increased over 4 weeks to a maximum of CMI or FLV 250 mg/day, or PXT 50 mg/day. CBT using exposure and response prevention was subsequently initiated with psychoeducational interventions and behavioral analysis. Atypical antipsychotics such as risperidone or olanzapine were added if patients were nonresponsive to at least 2 SRI trials and CBT (Pallanti and Quercioli, 2006; Matsunaga et al., 2009). One-year treatment response was evaluated using change in Y-BOCS total score. We previously performed a principal components analysis of the Y-BOCS Symptom Checklist symptom categories (Matsunaga et al., 2008). For the current study, subjects were categorized into four symptom dimension subgroups; 1) contamination/washing, 2) hoarding, 3) symmetry/repeating and ordering, and 4) forbidden thoughts/checking on the basis of their principal symptoms (score = 2 (predominant)) with kappa coefficients ranging from 0.72 to 1.0. Clinical variables such as gender, age, age at onset, global functioning, degree of insight, and 1-year treatment outcome were compared across these subgroups. For the main between-group comparisons, one-way and twoway analysis of variance (ANOVA) were used. For categorical data, chi-square tests with Yate's correction for discontinuity or Fisher's exact test (if the minimum expected cell size ≤ 5) was used. For investigating treatment predictors, stepwise multiple regression analyses with 1-year improvement rates of the Y-BOCS as the dependent variable were also performed. The possible indicators such as age at onset, GAFS score or pretreatment Y-BOCS total score, along with predominant symptom dimension score (present = 1) were included as independent variables. For all stepwise variable selection procedures, the probability of F to enter the regression equations was set at 0.05, and to remove at 0.10. Colinearity diagnostics were also performed. Significance level was set at P < 0.05. All statistical analyses in the study were conducted by using the SPSS statistical package (ver.14.0).
3. Results 3.1. Demographics The subjects were 142 men (41%) and 201 women (59%), ranging in age from 15 to 68 years (mean = 30.4, S.D. = 9.4). All patients had primarily suffered from OCD for at least 2 years with a mean duration
of 7.8 years (S.D. = 5.6). Mean age at onset was 22.6 years (S.D. = 7.4), mean duration of education was 12.9 (S.D. = 2.5), and mean GAFS score was 49.6 (S.D. = 7.5). The proportions of married subjects, subjects with involvement of others, or those with lifetime comorbidity of major depression were 37%, 38%, or 44%, respectively. In addition, 27% of the subjects had any lifetime history of impulsive behaviors. At the pre-treatment assessments for each participant, the percentage of subjects with poor insight was 43%. Mean Y-BOCS total score was 26.9 (S.D. = 4.8), mean Y-BOCS obsessions subscale was 13.7 (S.D. = 2.5), and mean Y-BOCS compulsions subscale was 13.2 (S.D. = 2.8) (Table 1). 3.2. OCD symptoms and factor analysis Table 2 shows the contents and prevalence of each primary obsession and compulsion categorized using the Y-BOCS symptom checklist in the participants. All but 17 (5%) patients with pure obsessions presented with a mixture of obsessions and compulsions. Contamination obsessions (48%) and cleaning/washing and checking compulsions were the most common symptoms. Using the principal components analysis of the 13 Y-BOCS symptom checklist categories, four factors that explained 57.7% of variance were identified: (1) contamination/washing (21.2%), (2) hoarding (14.3%), (3) symmetry/repeating rituals and ordering compulsions (11.9%), and (4) forbidden (aggressive) thoughts/checking (10.3%) (Matsunaga et al., 2008). 3.3. Baseline demographics For the current study, subjects were categorized by the clinician into four groups (contamination, hoarding, symmetry, and forbidden thoughts) on the basis of the identified predominant symptoms in each subject. In this procedure, 89 subjects (26%) were excluded from the study, because 1) they had any predominant symptom constellation other than the symptom dimensions (e.g., checking compulsion related to symmetry obsession), 2) the determination of predominant symptoms was considered too difficult and ambiguous because of multiple major symptoms co-existing. The remaining 254 (74%) subjects were divided into the four groups; 117 subjects were included in the “contamination/washing (CW)” group, 10 subjects were included in the “hoarding” group, 58 subjects were included in the “symmetry/ordering and repeating rituals (SOR)” group, and 69 Table 1 Demographic profiles, clinical features and psychometric test results of total subjects. Total N = 343 Gender (male/female) Age (years) Age at onset (years) Duration of illness (years) Education (years) GAFSa score Married subjects (N/%) Poor insight (N/%) Involvement of othersb (N/%) Lifetime major depression (N/%) Any history of impulsive behaviors (N/%)
142/201 30.4 ± 9.4 22.6 ± 7.4 7.8 ± 5.6 12.9 ± 2.5 49.6 ± 7.5 128 (37%)⁎ 147 (43%) 130 (38%) 151 (44%) 91 (27%)
Y-BOCSc Total Obsession Compulsion
26.9 ± 4.8 13.7 ± 2.5 13.2 ± 2.8
⁎ Values were expressed as numbers (%). a GAFS; Global Assessment of Functioning Scale (Axis V). b The presence of the involvement of others was assessed, if the OCD subjects involved someone in their OCD symptoms by forcing someone to help with rituals, or to offer them reassurance. c Yale–Brown Obsessive–Compulsive Scale.
H. Matsunaga et al. / Psychiatry Research 180 (2010) 25–29 Table 2 Obsessive–compulsive disorder (OCD) symptoms in the subjects. All patients (N = 343) Obsession Aggressive Contamination Sexual Hoarding Religious Symmetry and exactness Somatic Miscellaneous
Compulsion 125 (36%) 164 (48%) 35 (10%) 40 (12%) 27 (8%) 144 (42%) 40 (12%) 129 (38%)
Cleaning/washing Checking Repeating rituals Counting Ordering Hoarding Miscellaneous
160 161 108 47 74 40 108
(47%) (47%) (31%) (14%) (22%) (12%) (31%)
Values were expressed as numbers (%).
subjects were included in the “forbidden thoughts/checking (FC)” group. There were no significant differences in mean duration of education, the proportions of subjects with involvement of others in symptoms, those with lifetime comorbidity of major depression, or those with any history of impulsive behaviors across the symptom dimension groups. The proportion of female subjects in the CW group was significantly greater than that in the other groups (CW versus hoarding; P < 0.01 (Fisher's exact test), CW versus SOR; χ2 = 30.5, df = 1, P < 0.01, CW versus FC; χ2 = 30.2, df = 1, P < 0.01). Subjects in the SOR groups were significantly more likely than those in the CW group to have a younger mean age (F = 1.69, df = 3, 250, P < 0.05), a younger mean age at onset (F = 5.67, df = 3,250, P < 0.01), a lower mean score on the GAFS (F = 4.08, df = 3,250, P < 0.01), a more elevated mean total score on the pre-treatment Y-BOCS (F = 2.58, df = 3,250, P < 0.01), and a lower proportion of marriage (χ2 = 4.88, df = 1, P < 0.05). This group also had a significantly younger age at onset (P < 0.01), lower GAFS score (P < 0.01), higher mean score on the pre-treatment Y-BOCS (P < 0.05), and higher proportion of subjects with poor insight (χ2 = 4.91, df = 1, P < 0.05) than those in the FC group. In the hoarding group, similarly, the mean age at onset was younger than the CW group (P < 0.05), and the mean duration of illness was longer than the FC group (F = 2.02, df = 3,250, P < 0.05).
27
The hoarding group also had a significantly lower proportion of married subjects than the CW group (P < 0.05 (Fisher's exact test)), and a significantly greater proportion of subjects who were assessed as having poor insight than both the CW (P < 0.01 (Fisher's exact test)) and the FC (P < 0.01 (Fisher's exact test)) groups. 3.4. One-year treatment response In all of the subjects, improvement rate on Y-BOCS total score was 41% at the time of 1-year after starting the treatment. In these subjects, the mean maximum daily dosage of each SRI (S.D. (% of patients)) was 176.2 mg/day (50.9 (18%)) of CMI, 183.0 mg/day (45.9 (42%)) of FLV, or 39.7 mg/day (9.8(41%)) of PXT. The rate of the subjects who were assessed as “non-responder” to the adequate trials of at least two SRIs and subsequently received augmentation trials of atypical antipsychotics such as risperidone or olanzapine was 43%. Although there were no significant differences of the distribution of selected SRIs and a mean daily dosage of each SRI among the groups, the subjects who received augmentation trials of antipsychotics were significantly more likely to be in both the “hoarding” (P < 0.01(Fisher's exact test), and the “SOR” (versus CW; χ2 = 10.28, df = 1, P < 0.05, versus FC, P < 0.05) groups than in the other groups (Table 3). The 1year improvement rate on the Y-BOCS total score in the hoarding group was significantly lower than that in the other groups (F = 6.93, df = 3,206, P < 0.01). Subjects in the FC group also showed a significantly greater rate of improvement than the SOR (P < 0.05) group. In the subjects, finally, a worse response was significantly predicted in the stepwise multiple regression analyses by GAFS score (R2 = 0.199, Beta = 1.18. P < 0.001), and duration of illness (R2 = 0.21, Beta = 0.21, P = 0.045). 4. Discussion Our first finding was that a simple measure of predominant symptoms allowed the clinicians to categorize 74% of the subjects into one of four symptom dimension groups. Conversely, the finding that 26% could not readily be classified into the particular group might demonstrate a limitation of such a categorical approach for describing
Table 3 Comparisons of demographic profiles, clinical features and treatment outcome between the groups categorized by symptom dimensions. Contamination/washing
Hoarding
Symmetry/ordering and repeating rituals
Forbidden thoughts/checking
N Gender (male; female) Age (years) Age at onset (years) Duration of illness (years)
117 21;96 30.9 ± 9.3 23.0 ± 6.8 7.8 ± 5.5
10 10;0⁎⁎ 29.2 ± 6.9 18.2 ± 3.1⁎ 11.0 ± 7.0#
58 35;23⁎⁎ 27.9 ± 8.9⁎ 19.2 ± 6.1⁎⁎## 8.8 ± 6.9
69 40;29⁎⁎ 29.8 ± 8.8 22.8 ± 7.6 6.9 ± 4.2
Education (years) GAFS score Married patients (%) Poor insight (%) Involvement of others (%)a Lifetime major depression (%) Any history of impulsive
49.9 ± 6.5 44 45 44% 48% 22%
46.9 ± 7.5 0⁎ 100⁎⁎## 20% 20% 20%
46.6 ± 8.6⁎⁎## 26⁎ 58# 38% 36% 21%
50.6 ± 7.8 31 36 39% 42% 24%
Behaviors (%) Augmentation of antipsychotics (%)
24%
75%⁎⁎##
50%⁎#
30%
26.5 ± 4.9
28.1 ± 3.5
28.4 ± 5.1⁎⁎#
26.7 ± 4.4
39.2 ± 18.8
11.5 ± 6.5⁎⁎##$
36.4 ± 20.3#
45.5 ± 18.1
a
Y-BOCS total score Pre-treatment After 1-year treatment Improvement rate (%)
*P < 0.05, **P < 0.01; compared to the contamination/washing group. # P < 0.05, ##P < 0.01; compared to the forbidden thoughts/checking group. $ P < 0.05; compared to the symmetry/ordering and repeating rituals group. Group means of parametric variables were compared by one-way analysis of variance (ANOVA). Comparisons of non-parametric variables were made by chi-square tests with Yate's. Correction for discontinuity, or Fisher's exact tests, if the minimum cell size < 5.
28
H. Matsunaga et al. / Psychiatry Research 180 (2010) 25–29
heterogeneity of OCD. It also emphasizes that OCD is a complex disorder in which patients may suffer from multiple different symptom dimensions that are all equally predominant or disabling, or may have unusual symptoms that do not neatly fit in with the most frequently described dimensions. Our second finding was a simple measure of predominant symptom yielded subgroups with differential clinical features. The groups were differentially characterized by gender ratio, age at onset, level of global functioning, severity of OCD symptoms, and treatment responses. For instance, despite the small sample size, the hoarding symptom group in this study was significantly more likely than the other groups to show longer duration of illness, lower rate of marriage, poor insight and GAFS, and poorer responses to both SRIs and the combination treatments. Thus hoarding seems to be the most specific, disabling and treatment-resistant symptom, which is consistent with our factor-analyzed study (Matsunaga et al., 2008) and with a previous literature (Mataix-Cols et al., 1999; Mataix-Cols et al., 2002b; Samuels et al., 2002; Steketee and Frost, 2003; Saxena, 2007), including research on the unique neural circuitry associated with hoarding (Mataix-Cols et al., 2004; Saxena et al., 2004). While the effectiveness of SSRIs on hoarding symptoms remains controversial (Saxena et al., 2007), the findings may suggest that additional alternative strategies such as modified and intensive CBT, or MAOIs, should be considered in the treatment for patients with compulsive hoarding (Mataix-Cols et al., 1999; Mataix-Cols et al., 2002b; Samuels et al., 2002; Steketee and Frost, 2003; Saxena et al., 2004; Saxena, 2007; Pertusa et al., 2008). Recently, the CBT with specialized components to address problems with motivation, organizing, acquiring, and removing clutter has been proposed as a promising intervention for compulsive hoarding (Tolin et al., 2007). Similar to the hoarding group, the patients assigned in the “symmetry/ordering and repeating rituals” group were more likely than those in both the “contamination/washing” and/or “forbidden thoughts/checking” groups to exhibit severe clinical and psychopathological features such as lower GAFS, poorer insight and higher proportion of subjects who were assessed as SSRI-refractory and required the augmentation of antipsychotics. In the previous studies, this symptom dimension has been associated with clinical features such as elevated comorbidity of tic disorders and poor response to SRI treatment (Baer, 1994; Leckman et al., 1997; Mataix-Cols et al., 1999; Eichstedt and Arnold, 2001; Grados et al., 2001; Mathis et al., 2006; Matsunaga et al., 2008; Landeros-Weisenberger et al., 2010). Thus, taking into account the possibility that dopamine dysfunction may be preferentially involved in the neurobiological pathology of the symptom dimension, a better response to treatments found in this group compared to the hoarding group might be rather due to the direct effects of augmenting atypical neuroleptics to SRIs (Matsunaga et al., 2009). On the other hand, subjects in the “forbidden thoughts/ checking” group showed the most improved response among the symptom groups, which is consistent with previous studies along with a priori hypothesis (Mataix-Cols et al., 1999; Mataix-Cols et al., 2002b; Matsunaga et al., 2008; Landeros-Weisenberger et al., 2010). Nevertheless, a number of limitations to using a simple measure of symptom dimensions should be noted. In this study, for instance, only 25% of the subjects with hoarding symptoms (N = 40) were identified as predominantly having hoarding. It is possible that poorer insight and/or the ego-syntonic nature of the compulsive hoarding may lead to a minimization of the symptom group (Mataix-Cols et al., 2002b; Samuels et al., 2002), Moreover, compulsive hoarding should be heterogeneous; hoarding observed in some patients can be considered as a behavior that is secondary to other OCD symptom dimensions (Pertusa et al., 2008). In addition, even though the compulsive hoarding can be consistently seen as a significant indicator of treatment-resistance for the combination of pharmacotherapy and CBT (Matsunaga et al., 2008), a worse response was significantly predicted in the stepwise multiple regression analyses by GAFS score and duration of illness, not by the predominance of the
hoarding dimension. The current study also revealed that about a quarter of the participants could not easily be categorized, reflecting overlap in symptoms. Indeed, a considerable number of subjects equally had both washing and checking compulsions, and compared to both pure lifetime washers and pure lifetime checkers, they are significantly more likely to have pervasive and severe psychopathology and global dysfunction (Matsunaga et al., 2001). In this study, moreover, whereas a higher reliability for determining the predominant symptom in each subject was warranted, this determination and the classification of the subjects had been conducted by the psychiatrists specialized in OCD not by general psychiatric or primary care practice. Such possible problems related to reliability in clinical practice should also be taken into account. Thus, our findings could not sufficiently support the usefulness of such a simple measure of symptom dimensions in clinical practice of OCD. Nevertheless, symptom dimensions are not meant to be considered as mutually exclusive, discrete entities but rather as multiple complimentary domains (Mataix-Cols et al., 2005). Indeed, it has been suggested that the hoarding and symmetry factors are closely related to one another (Baer, 1994; Bloch et al., 2008). On the other hand, symptom dimensions have well been certified valid and useful especially for research (Mataix-Cols et al., 2004; Hasler et al., 2005; Mataix-Cols et al., 2005; Mathis et al., 2006; Rosario-Campos et al., 2006; Saxena, 2007; Leckman and Bloch, 2008). While their utility has also been demonstrated in studies of treatment outcome (Mataix-Cols et al., 1999; Mataix-Cols et al., 2002b; Mataix-Cols et al., 2005; Matsunaga et al., 2008), such a possible multidimensional “map” of vulnerability as symptom dimensions may be too complex for clinical application to be a useable nosology (Leckman et al., 2007). Moreover, even though the DY-BOCS has been developed as a useful tool in assessing OC symptom dimensions, time burden along with expert clinicians or their highly trained staffs should be needed for its administration in a reliable manner (Rosario-Campos et al., 2006) Thus, only a simple assessment of the predominant symptom dimension may be practical in clinical settings (Leckman et al., 2007). In this case, however, the limitations found in our study should be borne in mind, and further work on their validity and utility should be needed. Nevertheless, there are some methodological limitations in the current study. First, there may be a sample bias in this study; a larger part of the subjects contacted our clinic with an introduction from other psychiatrists seeking for more intensive and specialized treatments of OCD. Therefore, our subjects might be more likely to have severe clinical features related to OCD with greater deterioration in global functioning. Indeed, the higher rate of poor insight (43%) was observed in our subjects compared to those reported in the studies including our previous study (15–36%) (Eisen et al., 1998; Matsunaga et al., 2001; Alonso et al., 2008). Second, certain types of OCD symptoms such as miscellaneous obsessions and compulsions were not included in the analyses, and it remains to be elucidated whether these symptoms may form a unique or multiple separate dimensions (Mataix-Cols et al., 2005). Third, patients in the hoarding group had a lower sample size, and statistical power might be inadequate to analyze group differences. In addition to multiple testing, the threshold for statistical significance was set at P < 0.05, which may have caused Type 1-errors. Fourth, even though hoarding could also be a symptom of obsessive–compulsive personality disorder, we did not control for this comorbidity. Fifth, while our series of studies rather supported the trans-cultural stability of the symptom dimensions, socio-cultural factors may be important in shaping the content of OCD symptoms along with psychopathological features (Matsunaga and Seedat, 2007). Finally, a symptom-based approach seems to fail to take into account the tendency of patients to present with multiple types of symptoms, and the changing constellation of symptoms that can be observed over time (Calamari et al., 1999; Leckman et al., 1997). Even though symptoms of adult OCD patients have been suggested stable across time (Mataix-Cols et al., 2002a), the retrospective judging of
H. Matsunaga et al. / Psychiatry Research 180 (2010) 25–29
most prominent symptom dimension in each participant may cause some bias on the results in this study. Thus, further longitudinal prospective studies are necessary to better understand the stability, validity and usefulness of the simple measure of predominant symptoms categorized according to symptom dimensions in clinical practice. Acknowledgements This study was supported in part by a grant-in-aid for scientific research from the Japanese Ministry of Education, Culture, Sports, Science, and Technology (No. 16591154).
References Alonso, P., Menchón, J.M., Segalàs, C., Jaurrieta, N., Jiménez-Murcia, S., Cardoner, N., Labad, J., Real, E., Pertusa, A., Vallejo, J., 2008. Clinical implications of insight assessment in obsessive–compulsive disorder. Comprehensive Psychiatry 49, 305–312. American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Author, Washington, DC. Baer, L., 1994. Factor analysis of symptom subtypes of obsessive–compulsive disorder and their relation to personality and tic disorders. Journal of Clinical Psychiatry 55 (suppl 3), 18–23. Bloch, M.H., Landeros-Weisenberger, A., Rasario, M.C., Pittenger, C., Leckman, J.F., 2008. Meta-analysis of the symptom structure of obsessive–compulsive disorder. American Journal of Psychiatry 165, 1532–1542. Calamari, J.E., Wiegartz, P.S., Janeck, A.S., 1999. Obsessive–compulsive disorder subgroups; a symptom-based clustering approach. Behaviour Research and Therapy 37, 113–125. Eichstedt, J.A., Arnold, S.L., 2001. Childhood-onset obsessive–compulsive disorder: a tic-related subtype of OCD? Clinical Psychology Review 21, 137–158. Eisen, J.L., Phillips, K.A., Baer, L., Beer, D.A., Atala, K.P., Rasmussen, S.A., 1998. The Brown Assessment of Beliefs Scale—reliability and validity. American Journal of Psychiatry 155, 102–108. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B., 1997. The Structured Clinical Interview for DSM-IV Axis I Disorders: Biometric Research. New York State Psychiatric Institute, New York. Grados, M.A., Riddle, M.A., Samuels, J.F., Liang, K.Y., Hoehn-Saric, R., Bienvenu, O.J., Walkup, J.T., Song, D.H., Nestadt, G., 2001. The familial phenotype of obsessive– compulsive disorder in relation to tic disorders; the Hopkins OCD family study. Biological Psychiatry 50, 559–565. Goodman, W., Price, L., Rasmussen, S.A., Mazure, C., Fleischmann, R.L., Hill, C.L., Heninger, G.R., Charney, D.S., 1989a. The Yale–Brown Obsessive–Compulsive Scale, I: development, use, and reliability. Archives of General Psychiatry 46, 1006–1011. Goodman, W., Price, L., Rasmussen, S.A., Mazure, C., Delgado, P., Heninger, G.R., Charney, D.S., 1989b. The Yale–Brown Obsessive–Compulsive Scale, II: validity. Archives of General Psychiatry 46, 1012–1016. Hasler, G., Lasalle-Ricci, V.H., Ronquillo, J.G., Crawley, S.A., Cohran, L.W., Kazuba, D., Greenberg, B.D., Murphy, D.L., 2005. Obsessive–compulsive symptom dimensions show specific relationships to psychiatric comorbidity. Psychiatry Research 135, 121–132. Landeros-Weisenberger, A., Bloch, M.H., Kelmendi, B., Wegner, R., Nudel, J., Dombrowski, P., Pittenger, C., Krystal, J.H., Goodman, W.K., Leckman, J.F., Coric, V., 2010. Dimensional predictors of response to SRI pharmacotherapy in obsessive–compulsive disorder. Journal of Affective Disorders 121, 175–179. Leckman, J.F., Bloch, M.H., 2008. A developmental and evolutionally perspective on obsessive–compulsive disorder: whence and whither compulsive hoarding. American Journal of Psychiatry 165, 1229–1233. Leckman, J.F., Grice, D.E., Boardman, J., Zhang, H., Vitale, A., Bondi, C., Alsobrook, J., Peterson, B.S., Cohen, D.J., Rasmussen, S.A., Goodman, W.K., McDougle, C.J., Pauls, D.L., 1997. Symptoms of obsessive–compulsive disorder. American Journal of Psychiatry 154, 911–917. Leckman, J.F., Rauch, S.L., Mataix-Cols, D., 2007. Symptom dimension in obsessive– compulsive disorder: implication for the DSM-V. CNS Spectrums 12, 376–387. Lochner, C., Stein, D.J., 2003. Heterogeneity of obsessive–compulsive disorder; a literature review. Harvard Review of Psychiatry 11, 113–132. Mataix-Cols, D., Rauch, S.L., Manzo, P.A., Jenike, M.A., Baer, L., 1999. Use of factor-analyzed symptom dimensions to predict outcome with serotonin reuptake inhibitors and placebo in the treatment of obsessive–compulsive disorder. American Journal of Psychiatry 156, 1409–1416. Mataix-Cols, D., Rauch, S.L., Baer, L., Eisen, J.L., Shera, D.M., Goodman, W.K., Rasumussen, S.A., Jenike, M.A., 2002a. Symptom stability in adult obsessive–compulsive disorder:
29
data from a naturalistic two-year follow-up study. American Journal of Psychiatry 159, 263–268. Mataix-Cols, D., Marks, I.M., Greist, J.H., Kobak, K.A., Baer, L., 2002b. Obsessive–compulsive symptom dimensions as predictors of compliance with and response to behavior therapy; results from a controlled trial. Psychotherapy and Psychosomatics 71, 255–262. Mataix-Cols, D., Wooderson, S., Lawrence, N., Brammer, M.J., Speckens, A., Phillips, M.L., 2004. Distinct neural correlates of washing, checking, and hoarding symptom dimensions in obsessive–compulsive disorder. Archives of General Psychiatry 61, 564–576. Mataix-Cols, D., Rosario-Campos, M.C., Leckman, J.F., 2005. A multidimensional model of obsessive–compulsive disorder. American Journal of Psychiatry 162, 228–238. Mathis, M.A., Diniz, J.B., Rosario, M.C., Torres, A.R., Hoexter, M., Hasler, G., Miguel, E.C., 2006. What is the optimal way to subdivide obsessive–compulsive disorder? CNS Spectrums 11, 762–779. Matsunaga, H., Seedat, S., 2007. Obsessive–compulsive spectrum disorders: cross-national and ethnic issues. CNS Spectrums 12, 392–400. Matsunaga, H., Kiriike, N., Matsui, T., Iwasaki, Y., Miyata, A., Fujimoto, K., 2000. Gender differences of social and interpersonal features and personality disorders among Japanese patients with obsessive–compulsive disorder. Comprehensive Psychiatry 41, 266–272. Matsunaga, H., Kiriike, N., Matsui, T., Iwasaki, Y., Koshimune, K., Ohya, K., Stein, D.J., 2001. A comparative study of clinical features between pure lifetime checkers and pure washers categorized using a lifetime symptom rating method. Psychiatry Research 105, 221–229. Matsunaga, H., Kiriike, N., Matsui, T., Oya, K., Iwasaki, Y., Koshimune, K., Miyata, A., Stein, D.J., 2002. Obsessive–compulsive disorder patients with poor Insight. Comprehensive Psychiatry 43, 150–157. Matsunaga, H., Kiriike, N., Oya, K., Okino, K., Miyata, A., Matsui, T., Kosimune, K., Stein, D.J., 2005. Impulsive-control disorders in Japanese adult patients with obsessive– compulsive disorder. Comprehensive Psychiatry 46, 43–49. Matsunaga, H., Maebayashi, K., Hayashida, K., Okino, K., Matsui, T., Iketani, T., Kiriike, N., Stein, D.J., 2008. Symptom structure in Japanese patients with obsessive– compulsive disorder. American Journal of Psychiatry 165, 251–253. Matsunaga, H., Nagata, T., Hayashida, K., Ohya, K., Kiriike, K., Stein, D.J., 2009. A long-term effectiveness and safety of atypical antipsychotic agents in augmenting SSRIrefractory obsessive–compulsive disorder. Journal of Clinical Psychiatry 70, 863–868. McDougle, C.J., Goodman, W.K., Leckman, J.F., Lee, N.C., Heninger, G.R., Price, L.H., 1994. Haloperidol addition in fluvoxamine-refractory obsessive–compulsive disorder; a double-blind, placebo-controlled study in patients with and without tics. Archives of General Psychiatry 51, 302–308. Nakajima, T., Nakamura, M., Taga, C., Yamagami, S., Kiriike, N., Nagata, T., Saitoh, M., Kinoshita, T., Okajima, Y., Hanada, M., Tazoe, H., Yamaguchi, K., 1995. Reliability and validity of the Japanese version of the the Yale–Brown Obsessive Compulsive Scale. Psychiatry Clinical Neurosciences 49, 121–126. Pallanti, S., Quercioli, L., 2006. Treatment-refractory obsessive–compulsive disorder; methodological issues, operational definitions and therapeutic lines. Progress in Neuropsychopharmacology & Biological Psychiatry 30, 400–412. Pertusa, A., Fullana, M.A., Singh, S., Alonso, P., Menchon, J.M., Mataix-Cols, D., 2008. Compulsive hoarding; OCD symptom, distinct clinical syndrome, or both? American Journal of Psychiatry 165, 1289–1298. Rosario-Campos, M.C., Miguel, E.C., Quatrano, S., Chacon, P., Ferrao, Y., Findley, D., Katsovich, L., Scahill, L., King, R.A., Woody, S.R., Tolin, D., Hollander, E., Kano, Y., Leckman, J.F., 2006. The dimensional Yale–Brown Obsessive–Compulsive Scale (DYBOCS); an instrument for assessing obsessive–compulsive symptom dimensions. Molecular Psychiatry 11, 495–504. Samuels, J.F., Bienvenu, O.J., Riddle, R.A., Cullen, B., Grados, M.A., Liang, K.Y., HoehnSaric, R., Nestadt, G., 2002. Hoarding in obsessive–compulsive disorder: results from a case-control study. Behaviour Research and Therapy 40, 517–528. Saxena, S., 2007. Is compulsive hoarding a genetically and neurobiologicaly discrete syndrome? Implications for diagnostic classification. American Journal of Psychiatry 164, 380–384. Saxena, S., Brody, A.L., Maidment, K.M., Smith, E.C., Zohrabi, N., Katz, E., Baker, S.K., Baxter, L.R., 2004. Cerebral glucose metabolism in obsessive–compulsive hoarding. American Journal of Psychiatry 161, 1038–1048. Saxena, S., Brody, A.L., Maidment, K.M., Baxter, L.R., 2007. Paroxetine treatment of compulsive hoarding. Journal of Psychiatry Research 41, 481–487. Steketee, G., Frost, R., 2003. Compulsive hoarding; current status of the research. Clinical Psychology Review 23, 905–927. Summerfeldt, L.J., Richter, M.A., Antony, M.M., Swinson, R.P., 1999. Symptom structure in obsessive–compulsive disorder; a confirmatory factor analytic study. Behaviour Research and Therapy 37, 297–311. Tolin, D.F., Frost, R.O., Steketee, G., 2007. An open trial of cognitive–behavioral therapy for compulsive hoarding. Behaviour Research and Therapy 45, 1461–1470.
Contents lists available at ScienceDirect
Psychiatry Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s
The clinical utility of symptom dimensions in obsessive–compulsive disorder Hisato Matsunaga a,⁎, Kazuhisa Hayashida a, Nobuo Kiriike a, Kensei Maebayashi a, Dan J Stein b,c a b c
Department of Neuropsychiatry, Osaka City University Medical School, Osaka, Japan MRC Unit on Anxiety Disorders, University of Cape Town, Cape Town, South Africa Department of Psychiatry, Mt. Sinai Medical School, New York, USA
a r t i c l e
i n f o
Article history: Received 27 April 2009 Received in revised form 3 September 2009 Accepted 11 September 2009 Keywords: Obsessive–compulsive disorder Symptom dimension Yale–Brown Obsessive–Compulsive Scale (Y-BOCS) Clinical utility
a b s t r a c t Factor analyses in obsessive–compulsive disorder (OCD) have consistently identified several different symptom dimensions. Nevertheless the clinical utility of identifying such symptom dimensions remains somewhat unclear. On the basis of their principal symptoms, 343 OCD patients were divided into four symptom dimension subgroups; 1) contamination/washing, 2) hoarding, 3) symmetry/repeating and ordering, and 4) forbidden thoughts/checking. Clinical variables including 1-year treatment outcome were compared across these patient subgroups. Most patients (74%) could distinctively be categorized as falling into a particular symptom subgroup. The groups were differentially characterized by some demographic and clinical features. For instance, both the symmetry and hoarding groups were significantly associated with decreased global functioning and greater OCD severity. Moreover the hoarding group was significantly more likely than the others to show longer duration of illness, lower rate of marriage, poor insight, and poorer outcome. However, about a quarter of the participants could not be classified definitively into a particular group. Our findings provide partial support for the clinical utility of a simple measure of symptom dimensions in OCD. In clinical settings, however, the limitations of such a simple measure of predominant symptom dimensions should be borne in mind and further work on their validity and utility is needed. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction There is increasing evidence that obsessive–compulsive disorder (OCD) is a heterogeneous condition (Lochner and Stein, 2003; MataixCols et al., 2005; Mathis et al., 2006; Bloch et al., 2008). If OCD subtypes were characterized by a distinct psychobiology, then this would account for a variance in studies on biological markers, and would potentially impact on treatment strategies. For example, it has been suggested that some forms of OCD are etiologically related to tic disorders (Eichstedt and Arnold, 2001; Mathis et al., 2006), and OCD with comorbid tics has been characterized by specific phenomenological, genetic, and neuro-imaging features and a differential treatment response (McDougle et al., 1994; Eichstedt and Arnold, 2001; Grados et al., 2001; Mathis et al., 2006). Recent factor and cluster analyses have provided consistent evidence that distinct obsessive–compulsive symptom dimensions exist, including forbidden thoughts, contamination/washing, hoarding and symmetry/repeating rituals and ordering (Baer, 1994; Leckman et al., 1997; Calamari et al., 1999; Mataix-Cols et al., 1999; Summerfeldt et al., 1999; Bloch et al., 2008). These symptom dimensions have been differentially related to clinical features such ⁎ Corresponding author. Department of Neuropsychiatry, Osaka City University Medical School, 1-4-3 Asahi-machi, Abeno-ku, Osaka, 545-8585, Japan. Tel.: +81 6 6645 3821; fax: +81 6 6636 0439. E-mail address: [email protected] (H. Matsunaga). 0165-1781/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2009.09.005
as comorbidity (Baer, 1994; Leckman et al., 1997; Hasler et al., 2005; Mataix-Cols et al., 2005). Early-onset or tic-related OCD is a unique subtype characterized by more symmetry/ordering compulsions (Mathis et al., 2006), and OCD patients with hoarding tend to exhibit specific and disabling clinical features such as a lower level of global functioning and greater overall illness severity as well as poorer response to serotonin reuptake inhibitors (SRIs) (Mataix-Cols et al., 1999; Steketee and Frost, 2003; Mataix-Cols et al., 2005; Saxena, 2007; Matsunaga et al., 2008). It has also been hypothesized that each symptom dimension may be underpinned by a distinctive set of biobehavioral mechanisms (Mataix-Cols et al., 2004; Saxena et al., 2004). Data that OCD subtypes and symptom dimensions are stable across different cultures and ethnic groups (Bloch et al., 2008; Matsunaga et al., 2008), support the hypothesis that OCD is mediated by universal psychobiological mechanisms. Nevertheless, it remains unclear whether classifying patients as suffering predominantly from one or other symptom dimension has clinical utility. In research settings, it is possible to use instruments such as the Dimensional Yale–Brown Obsessive–Compulsive Scale (DYBOCS) (Rosario-Campos et al., 2006). In clinical settings, only a simple assessment of the predominant symptom dimension such as “OCD with prominent contamination obsessions and cleaning compulsions” may be practical (Leckman et al., 2007). It is unclear as to whether such a measure is always possible, and whether it would usefully inform clinical practice. Based on the previous findings (Mataix-Cols et al., 1999; Mataix-Cols et al., 2002b; Steketee and Frost, 2003; Mataix-Cols et
26
H. Matsunaga et al. / Psychiatry Research 180 (2010) 25–29
al., 2005; Saxena, 2007; Landeros-Weisenberger et al., 2010), it can be hypothesized that the predominance of hoarding symptoms would be associated with poorer treatment response and social disability, and forbidden thoughts might be associated with positive response to SRIs. In this study, we sought to examine 1) whether a simple clinical measure of predominant symptoms was consistent with the data from a more complex principle component factor analysis, and 2) whether such a measure is useful in informing practice, insofar as it is associated differentially with particular clinical characteristics of OCD. 2. Methods Subjects were 343 outpatients who met Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV; APA, 1994) criteria for OCD, consecutively admitted to our hospital between April, 2005 and December, 2007 and gave us informed consent to participate in this study. Each subject was diagnosed with the Structured Clinical Interview for DSM-IV Patient version (SCID-P) (First et al., 1997), received the standardized combination treatments for at least 1 year, and had participated in our recent study (Matsunaga et al., 2008). Information was obtained regarding demographic profile, family and medical history, and clinical features and course. Global functioning was assessed using the DSM-IV Axis V Global Assessment of Functioning Scale (GAFS) (APA, 1994). The involvement of others – i.e. whether OCD subjects involved others in their symptoms by forcing them to help with their rituals, or asking them for reassurance, or to perform ritualistic behaviors similar to their own – (Matsunaga et al., 2000) was also assessed in each subject. The procedure to assess the lifetime incidence of impulsive behaviors such as substance abuse, stealing, selfmutilation, suicide attempts, and destructive behaviors has been described in our previous study (Matsunaga et al., 2005). Ratings of OCD symptoms were obtained with the Japanese version of the Y-BOCS Symptom Checklist (Goodman et al., 1989a,b; Nakajima et al., 1995). The assessment was conducted in a semi-structured interview by two clinicians with extensive training in and experience with the scale. Up to three primary obsessions and compulsions were listed for each subject, and the principal obsession and compulsion symptom categories were determined. Each of the 13 major Y-BOCS Symptom Checklist categories except for miscellaneous obsessions and compulsions, was scored as 0 (absent), 1 (present), or 2 (predominant). The Y-BOCS was then administered to assess current OCD symptom severity. Degree of insight in each subject was evaluated using the insight question of the Y-BOCS (Goodman et al., 1989a,b). The assessment procedure and inter-rater reliability have also been reported (Matsunaga et al., 2002). After the pre-treatment assessments, each subject was treated with a standardized combination of SRIs, (i.e., clomipramine (CMI), fluvoxamine (FLV), or paroxetine (PXT)) and cognitive–behavioral therapy (CBT). Treatment was usually initiated with daily dosages of CMI or FLV 50 mg, or PXT 10 mg. If the medication was tolerated, dosage was gradually increased over 4 weeks to a maximum of CMI or FLV 250 mg/day, or PXT 50 mg/day. CBT using exposure and response prevention was subsequently initiated with psychoeducational interventions and behavioral analysis. Atypical antipsychotics such as risperidone or olanzapine were added if patients were nonresponsive to at least 2 SRI trials and CBT (Pallanti and Quercioli, 2006; Matsunaga et al., 2009). One-year treatment response was evaluated using change in Y-BOCS total score. We previously performed a principal components analysis of the Y-BOCS Symptom Checklist symptom categories (Matsunaga et al., 2008). For the current study, subjects were categorized into four symptom dimension subgroups; 1) contamination/washing, 2) hoarding, 3) symmetry/repeating and ordering, and 4) forbidden thoughts/checking on the basis of their principal symptoms (score = 2 (predominant)) with kappa coefficients ranging from 0.72 to 1.0. Clinical variables such as gender, age, age at onset, global functioning, degree of insight, and 1-year treatment outcome were compared across these subgroups. For the main between-group comparisons, one-way and twoway analysis of variance (ANOVA) were used. For categorical data, chi-square tests with Yate's correction for discontinuity or Fisher's exact test (if the minimum expected cell size ≤ 5) was used. For investigating treatment predictors, stepwise multiple regression analyses with 1-year improvement rates of the Y-BOCS as the dependent variable were also performed. The possible indicators such as age at onset, GAFS score or pretreatment Y-BOCS total score, along with predominant symptom dimension score (present = 1) were included as independent variables. For all stepwise variable selection procedures, the probability of F to enter the regression equations was set at 0.05, and to remove at 0.10. Colinearity diagnostics were also performed. Significance level was set at P < 0.05. All statistical analyses in the study were conducted by using the SPSS statistical package (ver.14.0).
3. Results 3.1. Demographics The subjects were 142 men (41%) and 201 women (59%), ranging in age from 15 to 68 years (mean = 30.4, S.D. = 9.4). All patients had primarily suffered from OCD for at least 2 years with a mean duration
of 7.8 years (S.D. = 5.6). Mean age at onset was 22.6 years (S.D. = 7.4), mean duration of education was 12.9 (S.D. = 2.5), and mean GAFS score was 49.6 (S.D. = 7.5). The proportions of married subjects, subjects with involvement of others, or those with lifetime comorbidity of major depression were 37%, 38%, or 44%, respectively. In addition, 27% of the subjects had any lifetime history of impulsive behaviors. At the pre-treatment assessments for each participant, the percentage of subjects with poor insight was 43%. Mean Y-BOCS total score was 26.9 (S.D. = 4.8), mean Y-BOCS obsessions subscale was 13.7 (S.D. = 2.5), and mean Y-BOCS compulsions subscale was 13.2 (S.D. = 2.8) (Table 1). 3.2. OCD symptoms and factor analysis Table 2 shows the contents and prevalence of each primary obsession and compulsion categorized using the Y-BOCS symptom checklist in the participants. All but 17 (5%) patients with pure obsessions presented with a mixture of obsessions and compulsions. Contamination obsessions (48%) and cleaning/washing and checking compulsions were the most common symptoms. Using the principal components analysis of the 13 Y-BOCS symptom checklist categories, four factors that explained 57.7% of variance were identified: (1) contamination/washing (21.2%), (2) hoarding (14.3%), (3) symmetry/repeating rituals and ordering compulsions (11.9%), and (4) forbidden (aggressive) thoughts/checking (10.3%) (Matsunaga et al., 2008). 3.3. Baseline demographics For the current study, subjects were categorized by the clinician into four groups (contamination, hoarding, symmetry, and forbidden thoughts) on the basis of the identified predominant symptoms in each subject. In this procedure, 89 subjects (26%) were excluded from the study, because 1) they had any predominant symptom constellation other than the symptom dimensions (e.g., checking compulsion related to symmetry obsession), 2) the determination of predominant symptoms was considered too difficult and ambiguous because of multiple major symptoms co-existing. The remaining 254 (74%) subjects were divided into the four groups; 117 subjects were included in the “contamination/washing (CW)” group, 10 subjects were included in the “hoarding” group, 58 subjects were included in the “symmetry/ordering and repeating rituals (SOR)” group, and 69 Table 1 Demographic profiles, clinical features and psychometric test results of total subjects. Total N = 343 Gender (male/female) Age (years) Age at onset (years) Duration of illness (years) Education (years) GAFSa score Married subjects (N/%) Poor insight (N/%) Involvement of othersb (N/%) Lifetime major depression (N/%) Any history of impulsive behaviors (N/%)
142/201 30.4 ± 9.4 22.6 ± 7.4 7.8 ± 5.6 12.9 ± 2.5 49.6 ± 7.5 128 (37%)⁎ 147 (43%) 130 (38%) 151 (44%) 91 (27%)
Y-BOCSc Total Obsession Compulsion
26.9 ± 4.8 13.7 ± 2.5 13.2 ± 2.8
⁎ Values were expressed as numbers (%). a GAFS; Global Assessment of Functioning Scale (Axis V). b The presence of the involvement of others was assessed, if the OCD subjects involved someone in their OCD symptoms by forcing someone to help with rituals, or to offer them reassurance. c Yale–Brown Obsessive–Compulsive Scale.
H. Matsunaga et al. / Psychiatry Research 180 (2010) 25–29 Table 2 Obsessive–compulsive disorder (OCD) symptoms in the subjects. All patients (N = 343) Obsession Aggressive Contamination Sexual Hoarding Religious Symmetry and exactness Somatic Miscellaneous
Compulsion 125 (36%) 164 (48%) 35 (10%) 40 (12%) 27 (8%) 144 (42%) 40 (12%) 129 (38%)
Cleaning/washing Checking Repeating rituals Counting Ordering Hoarding Miscellaneous
160 161 108 47 74 40 108
(47%) (47%) (31%) (14%) (22%) (12%) (31%)
Values were expressed as numbers (%).
subjects were included in the “forbidden thoughts/checking (FC)” group. There were no significant differences in mean duration of education, the proportions of subjects with involvement of others in symptoms, those with lifetime comorbidity of major depression, or those with any history of impulsive behaviors across the symptom dimension groups. The proportion of female subjects in the CW group was significantly greater than that in the other groups (CW versus hoarding; P < 0.01 (Fisher's exact test), CW versus SOR; χ2 = 30.5, df = 1, P < 0.01, CW versus FC; χ2 = 30.2, df = 1, P < 0.01). Subjects in the SOR groups were significantly more likely than those in the CW group to have a younger mean age (F = 1.69, df = 3, 250, P < 0.05), a younger mean age at onset (F = 5.67, df = 3,250, P < 0.01), a lower mean score on the GAFS (F = 4.08, df = 3,250, P < 0.01), a more elevated mean total score on the pre-treatment Y-BOCS (F = 2.58, df = 3,250, P < 0.01), and a lower proportion of marriage (χ2 = 4.88, df = 1, P < 0.05). This group also had a significantly younger age at onset (P < 0.01), lower GAFS score (P < 0.01), higher mean score on the pre-treatment Y-BOCS (P < 0.05), and higher proportion of subjects with poor insight (χ2 = 4.91, df = 1, P < 0.05) than those in the FC group. In the hoarding group, similarly, the mean age at onset was younger than the CW group (P < 0.05), and the mean duration of illness was longer than the FC group (F = 2.02, df = 3,250, P < 0.05).
27
The hoarding group also had a significantly lower proportion of married subjects than the CW group (P < 0.05 (Fisher's exact test)), and a significantly greater proportion of subjects who were assessed as having poor insight than both the CW (P < 0.01 (Fisher's exact test)) and the FC (P < 0.01 (Fisher's exact test)) groups. 3.4. One-year treatment response In all of the subjects, improvement rate on Y-BOCS total score was 41% at the time of 1-year after starting the treatment. In these subjects, the mean maximum daily dosage of each SRI (S.D. (% of patients)) was 176.2 mg/day (50.9 (18%)) of CMI, 183.0 mg/day (45.9 (42%)) of FLV, or 39.7 mg/day (9.8(41%)) of PXT. The rate of the subjects who were assessed as “non-responder” to the adequate trials of at least two SRIs and subsequently received augmentation trials of atypical antipsychotics such as risperidone or olanzapine was 43%. Although there were no significant differences of the distribution of selected SRIs and a mean daily dosage of each SRI among the groups, the subjects who received augmentation trials of antipsychotics were significantly more likely to be in both the “hoarding” (P < 0.01(Fisher's exact test), and the “SOR” (versus CW; χ2 = 10.28, df = 1, P < 0.05, versus FC, P < 0.05) groups than in the other groups (Table 3). The 1year improvement rate on the Y-BOCS total score in the hoarding group was significantly lower than that in the other groups (F = 6.93, df = 3,206, P < 0.01). Subjects in the FC group also showed a significantly greater rate of improvement than the SOR (P < 0.05) group. In the subjects, finally, a worse response was significantly predicted in the stepwise multiple regression analyses by GAFS score (R2 = 0.199, Beta = 1.18. P < 0.001), and duration of illness (R2 = 0.21, Beta = 0.21, P = 0.045). 4. Discussion Our first finding was that a simple measure of predominant symptoms allowed the clinicians to categorize 74% of the subjects into one of four symptom dimension groups. Conversely, the finding that 26% could not readily be classified into the particular group might demonstrate a limitation of such a categorical approach for describing
Table 3 Comparisons of demographic profiles, clinical features and treatment outcome between the groups categorized by symptom dimensions. Contamination/washing
Hoarding
Symmetry/ordering and repeating rituals
Forbidden thoughts/checking
N Gender (male; female) Age (years) Age at onset (years) Duration of illness (years)
117 21;96 30.9 ± 9.3 23.0 ± 6.8 7.8 ± 5.5
10 10;0⁎⁎ 29.2 ± 6.9 18.2 ± 3.1⁎ 11.0 ± 7.0#
58 35;23⁎⁎ 27.9 ± 8.9⁎ 19.2 ± 6.1⁎⁎## 8.8 ± 6.9
69 40;29⁎⁎ 29.8 ± 8.8 22.8 ± 7.6 6.9 ± 4.2
Education (years) GAFS score Married patients (%) Poor insight (%) Involvement of others (%)a Lifetime major depression (%) Any history of impulsive
49.9 ± 6.5 44 45 44% 48% 22%
46.9 ± 7.5 0⁎ 100⁎⁎## 20% 20% 20%
46.6 ± 8.6⁎⁎## 26⁎ 58# 38% 36% 21%
50.6 ± 7.8 31 36 39% 42% 24%
Behaviors (%) Augmentation of antipsychotics (%)
24%
75%⁎⁎##
50%⁎#
30%
26.5 ± 4.9
28.1 ± 3.5
28.4 ± 5.1⁎⁎#
26.7 ± 4.4
39.2 ± 18.8
11.5 ± 6.5⁎⁎##$
36.4 ± 20.3#
45.5 ± 18.1
a
Y-BOCS total score Pre-treatment After 1-year treatment Improvement rate (%)
*P < 0.05, **P < 0.01; compared to the contamination/washing group. # P < 0.05, ##P < 0.01; compared to the forbidden thoughts/checking group. $ P < 0.05; compared to the symmetry/ordering and repeating rituals group. Group means of parametric variables were compared by one-way analysis of variance (ANOVA). Comparisons of non-parametric variables were made by chi-square tests with Yate's. Correction for discontinuity, or Fisher's exact tests, if the minimum cell size < 5.
28
H. Matsunaga et al. / Psychiatry Research 180 (2010) 25–29
heterogeneity of OCD. It also emphasizes that OCD is a complex disorder in which patients may suffer from multiple different symptom dimensions that are all equally predominant or disabling, or may have unusual symptoms that do not neatly fit in with the most frequently described dimensions. Our second finding was a simple measure of predominant symptom yielded subgroups with differential clinical features. The groups were differentially characterized by gender ratio, age at onset, level of global functioning, severity of OCD symptoms, and treatment responses. For instance, despite the small sample size, the hoarding symptom group in this study was significantly more likely than the other groups to show longer duration of illness, lower rate of marriage, poor insight and GAFS, and poorer responses to both SRIs and the combination treatments. Thus hoarding seems to be the most specific, disabling and treatment-resistant symptom, which is consistent with our factor-analyzed study (Matsunaga et al., 2008) and with a previous literature (Mataix-Cols et al., 1999; Mataix-Cols et al., 2002b; Samuels et al., 2002; Steketee and Frost, 2003; Saxena, 2007), including research on the unique neural circuitry associated with hoarding (Mataix-Cols et al., 2004; Saxena et al., 2004). While the effectiveness of SSRIs on hoarding symptoms remains controversial (Saxena et al., 2007), the findings may suggest that additional alternative strategies such as modified and intensive CBT, or MAOIs, should be considered in the treatment for patients with compulsive hoarding (Mataix-Cols et al., 1999; Mataix-Cols et al., 2002b; Samuels et al., 2002; Steketee and Frost, 2003; Saxena et al., 2004; Saxena, 2007; Pertusa et al., 2008). Recently, the CBT with specialized components to address problems with motivation, organizing, acquiring, and removing clutter has been proposed as a promising intervention for compulsive hoarding (Tolin et al., 2007). Similar to the hoarding group, the patients assigned in the “symmetry/ordering and repeating rituals” group were more likely than those in both the “contamination/washing” and/or “forbidden thoughts/checking” groups to exhibit severe clinical and psychopathological features such as lower GAFS, poorer insight and higher proportion of subjects who were assessed as SSRI-refractory and required the augmentation of antipsychotics. In the previous studies, this symptom dimension has been associated with clinical features such as elevated comorbidity of tic disorders and poor response to SRI treatment (Baer, 1994; Leckman et al., 1997; Mataix-Cols et al., 1999; Eichstedt and Arnold, 2001; Grados et al., 2001; Mathis et al., 2006; Matsunaga et al., 2008; Landeros-Weisenberger et al., 2010). Thus, taking into account the possibility that dopamine dysfunction may be preferentially involved in the neurobiological pathology of the symptom dimension, a better response to treatments found in this group compared to the hoarding group might be rather due to the direct effects of augmenting atypical neuroleptics to SRIs (Matsunaga et al., 2009). On the other hand, subjects in the “forbidden thoughts/ checking” group showed the most improved response among the symptom groups, which is consistent with previous studies along with a priori hypothesis (Mataix-Cols et al., 1999; Mataix-Cols et al., 2002b; Matsunaga et al., 2008; Landeros-Weisenberger et al., 2010). Nevertheless, a number of limitations to using a simple measure of symptom dimensions should be noted. In this study, for instance, only 25% of the subjects with hoarding symptoms (N = 40) were identified as predominantly having hoarding. It is possible that poorer insight and/or the ego-syntonic nature of the compulsive hoarding may lead to a minimization of the symptom group (Mataix-Cols et al., 2002b; Samuels et al., 2002), Moreover, compulsive hoarding should be heterogeneous; hoarding observed in some patients can be considered as a behavior that is secondary to other OCD symptom dimensions (Pertusa et al., 2008). In addition, even though the compulsive hoarding can be consistently seen as a significant indicator of treatment-resistance for the combination of pharmacotherapy and CBT (Matsunaga et al., 2008), a worse response was significantly predicted in the stepwise multiple regression analyses by GAFS score and duration of illness, not by the predominance of the
hoarding dimension. The current study also revealed that about a quarter of the participants could not easily be categorized, reflecting overlap in symptoms. Indeed, a considerable number of subjects equally had both washing and checking compulsions, and compared to both pure lifetime washers and pure lifetime checkers, they are significantly more likely to have pervasive and severe psychopathology and global dysfunction (Matsunaga et al., 2001). In this study, moreover, whereas a higher reliability for determining the predominant symptom in each subject was warranted, this determination and the classification of the subjects had been conducted by the psychiatrists specialized in OCD not by general psychiatric or primary care practice. Such possible problems related to reliability in clinical practice should also be taken into account. Thus, our findings could not sufficiently support the usefulness of such a simple measure of symptom dimensions in clinical practice of OCD. Nevertheless, symptom dimensions are not meant to be considered as mutually exclusive, discrete entities but rather as multiple complimentary domains (Mataix-Cols et al., 2005). Indeed, it has been suggested that the hoarding and symmetry factors are closely related to one another (Baer, 1994; Bloch et al., 2008). On the other hand, symptom dimensions have well been certified valid and useful especially for research (Mataix-Cols et al., 2004; Hasler et al., 2005; Mataix-Cols et al., 2005; Mathis et al., 2006; Rosario-Campos et al., 2006; Saxena, 2007; Leckman and Bloch, 2008). While their utility has also been demonstrated in studies of treatment outcome (Mataix-Cols et al., 1999; Mataix-Cols et al., 2002b; Mataix-Cols et al., 2005; Matsunaga et al., 2008), such a possible multidimensional “map” of vulnerability as symptom dimensions may be too complex for clinical application to be a useable nosology (Leckman et al., 2007). Moreover, even though the DY-BOCS has been developed as a useful tool in assessing OC symptom dimensions, time burden along with expert clinicians or their highly trained staffs should be needed for its administration in a reliable manner (Rosario-Campos et al., 2006) Thus, only a simple assessment of the predominant symptom dimension may be practical in clinical settings (Leckman et al., 2007). In this case, however, the limitations found in our study should be borne in mind, and further work on their validity and utility should be needed. Nevertheless, there are some methodological limitations in the current study. First, there may be a sample bias in this study; a larger part of the subjects contacted our clinic with an introduction from other psychiatrists seeking for more intensive and specialized treatments of OCD. Therefore, our subjects might be more likely to have severe clinical features related to OCD with greater deterioration in global functioning. Indeed, the higher rate of poor insight (43%) was observed in our subjects compared to those reported in the studies including our previous study (15–36%) (Eisen et al., 1998; Matsunaga et al., 2001; Alonso et al., 2008). Second, certain types of OCD symptoms such as miscellaneous obsessions and compulsions were not included in the analyses, and it remains to be elucidated whether these symptoms may form a unique or multiple separate dimensions (Mataix-Cols et al., 2005). Third, patients in the hoarding group had a lower sample size, and statistical power might be inadequate to analyze group differences. In addition to multiple testing, the threshold for statistical significance was set at P < 0.05, which may have caused Type 1-errors. Fourth, even though hoarding could also be a symptom of obsessive–compulsive personality disorder, we did not control for this comorbidity. Fifth, while our series of studies rather supported the trans-cultural stability of the symptom dimensions, socio-cultural factors may be important in shaping the content of OCD symptoms along with psychopathological features (Matsunaga and Seedat, 2007). Finally, a symptom-based approach seems to fail to take into account the tendency of patients to present with multiple types of symptoms, and the changing constellation of symptoms that can be observed over time (Calamari et al., 1999; Leckman et al., 1997). Even though symptoms of adult OCD patients have been suggested stable across time (Mataix-Cols et al., 2002a), the retrospective judging of
H. Matsunaga et al. / Psychiatry Research 180 (2010) 25–29
most prominent symptom dimension in each participant may cause some bias on the results in this study. Thus, further longitudinal prospective studies are necessary to better understand the stability, validity and usefulness of the simple measure of predominant symptoms categorized according to symptom dimensions in clinical practice. Acknowledgements This study was supported in part by a grant-in-aid for scientific research from the Japanese Ministry of Education, Culture, Sports, Science, and Technology (No. 16591154).
References Alonso, P., Menchón, J.M., Segalàs, C., Jaurrieta, N., Jiménez-Murcia, S., Cardoner, N., Labad, J., Real, E., Pertusa, A., Vallejo, J., 2008. Clinical implications of insight assessment in obsessive–compulsive disorder. Comprehensive Psychiatry 49, 305–312. American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Author, Washington, DC. Baer, L., 1994. Factor analysis of symptom subtypes of obsessive–compulsive disorder and their relation to personality and tic disorders. Journal of Clinical Psychiatry 55 (suppl 3), 18–23. Bloch, M.H., Landeros-Weisenberger, A., Rasario, M.C., Pittenger, C., Leckman, J.F., 2008. Meta-analysis of the symptom structure of obsessive–compulsive disorder. American Journal of Psychiatry 165, 1532–1542. Calamari, J.E., Wiegartz, P.S., Janeck, A.S., 1999. Obsessive–compulsive disorder subgroups; a symptom-based clustering approach. Behaviour Research and Therapy 37, 113–125. Eichstedt, J.A., Arnold, S.L., 2001. Childhood-onset obsessive–compulsive disorder: a tic-related subtype of OCD? Clinical Psychology Review 21, 137–158. Eisen, J.L., Phillips, K.A., Baer, L., Beer, D.A., Atala, K.P., Rasmussen, S.A., 1998. The Brown Assessment of Beliefs Scale—reliability and validity. American Journal of Psychiatry 155, 102–108. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B., 1997. The Structured Clinical Interview for DSM-IV Axis I Disorders: Biometric Research. New York State Psychiatric Institute, New York. Grados, M.A., Riddle, M.A., Samuels, J.F., Liang, K.Y., Hoehn-Saric, R., Bienvenu, O.J., Walkup, J.T., Song, D.H., Nestadt, G., 2001. The familial phenotype of obsessive– compulsive disorder in relation to tic disorders; the Hopkins OCD family study. Biological Psychiatry 50, 559–565. Goodman, W., Price, L., Rasmussen, S.A., Mazure, C., Fleischmann, R.L., Hill, C.L., Heninger, G.R., Charney, D.S., 1989a. The Yale–Brown Obsessive–Compulsive Scale, I: development, use, and reliability. Archives of General Psychiatry 46, 1006–1011. Goodman, W., Price, L., Rasmussen, S.A., Mazure, C., Delgado, P., Heninger, G.R., Charney, D.S., 1989b. The Yale–Brown Obsessive–Compulsive Scale, II: validity. Archives of General Psychiatry 46, 1012–1016. Hasler, G., Lasalle-Ricci, V.H., Ronquillo, J.G., Crawley, S.A., Cohran, L.W., Kazuba, D., Greenberg, B.D., Murphy, D.L., 2005. Obsessive–compulsive symptom dimensions show specific relationships to psychiatric comorbidity. Psychiatry Research 135, 121–132. Landeros-Weisenberger, A., Bloch, M.H., Kelmendi, B., Wegner, R., Nudel, J., Dombrowski, P., Pittenger, C., Krystal, J.H., Goodman, W.K., Leckman, J.F., Coric, V., 2010. Dimensional predictors of response to SRI pharmacotherapy in obsessive–compulsive disorder. Journal of Affective Disorders 121, 175–179. Leckman, J.F., Bloch, M.H., 2008. A developmental and evolutionally perspective on obsessive–compulsive disorder: whence and whither compulsive hoarding. American Journal of Psychiatry 165, 1229–1233. Leckman, J.F., Grice, D.E., Boardman, J., Zhang, H., Vitale, A., Bondi, C., Alsobrook, J., Peterson, B.S., Cohen, D.J., Rasmussen, S.A., Goodman, W.K., McDougle, C.J., Pauls, D.L., 1997. Symptoms of obsessive–compulsive disorder. American Journal of Psychiatry 154, 911–917. Leckman, J.F., Rauch, S.L., Mataix-Cols, D., 2007. Symptom dimension in obsessive– compulsive disorder: implication for the DSM-V. CNS Spectrums 12, 376–387. Lochner, C., Stein, D.J., 2003. Heterogeneity of obsessive–compulsive disorder; a literature review. Harvard Review of Psychiatry 11, 113–132. Mataix-Cols, D., Rauch, S.L., Manzo, P.A., Jenike, M.A., Baer, L., 1999. Use of factor-analyzed symptom dimensions to predict outcome with serotonin reuptake inhibitors and placebo in the treatment of obsessive–compulsive disorder. American Journal of Psychiatry 156, 1409–1416. Mataix-Cols, D., Rauch, S.L., Baer, L., Eisen, J.L., Shera, D.M., Goodman, W.K., Rasumussen, S.A., Jenike, M.A., 2002a. Symptom stability in adult obsessive–compulsive disorder:
29
data from a naturalistic two-year follow-up study. American Journal of Psychiatry 159, 263–268. Mataix-Cols, D., Marks, I.M., Greist, J.H., Kobak, K.A., Baer, L., 2002b. Obsessive–compulsive symptom dimensions as predictors of compliance with and response to behavior therapy; results from a controlled trial. Psychotherapy and Psychosomatics 71, 255–262. Mataix-Cols, D., Wooderson, S., Lawrence, N., Brammer, M.J., Speckens, A., Phillips, M.L., 2004. Distinct neural correlates of washing, checking, and hoarding symptom dimensions in obsessive–compulsive disorder. Archives of General Psychiatry 61, 564–576. Mataix-Cols, D., Rosario-Campos, M.C., Leckman, J.F., 2005. A multidimensional model of obsessive–compulsive disorder. American Journal of Psychiatry 162, 228–238. Mathis, M.A., Diniz, J.B., Rosario, M.C., Torres, A.R., Hoexter, M., Hasler, G., Miguel, E.C., 2006. What is the optimal way to subdivide obsessive–compulsive disorder? CNS Spectrums 11, 762–779. Matsunaga, H., Seedat, S., 2007. Obsessive–compulsive spectrum disorders: cross-national and ethnic issues. CNS Spectrums 12, 392–400. Matsunaga, H., Kiriike, N., Matsui, T., Iwasaki, Y., Miyata, A., Fujimoto, K., 2000. Gender differences of social and interpersonal features and personality disorders among Japanese patients with obsessive–compulsive disorder. Comprehensive Psychiatry 41, 266–272. Matsunaga, H., Kiriike, N., Matsui, T., Iwasaki, Y., Koshimune, K., Ohya, K., Stein, D.J., 2001. A comparative study of clinical features between pure lifetime checkers and pure washers categorized using a lifetime symptom rating method. Psychiatry Research 105, 221–229. Matsunaga, H., Kiriike, N., Matsui, T., Oya, K., Iwasaki, Y., Koshimune, K., Miyata, A., Stein, D.J., 2002. Obsessive–compulsive disorder patients with poor Insight. Comprehensive Psychiatry 43, 150–157. Matsunaga, H., Kiriike, N., Oya, K., Okino, K., Miyata, A., Matsui, T., Kosimune, K., Stein, D.J., 2005. Impulsive-control disorders in Japanese adult patients with obsessive– compulsive disorder. Comprehensive Psychiatry 46, 43–49. Matsunaga, H., Maebayashi, K., Hayashida, K., Okino, K., Matsui, T., Iketani, T., Kiriike, N., Stein, D.J., 2008. Symptom structure in Japanese patients with obsessive– compulsive disorder. American Journal of Psychiatry 165, 251–253. Matsunaga, H., Nagata, T., Hayashida, K., Ohya, K., Kiriike, K., Stein, D.J., 2009. A long-term effectiveness and safety of atypical antipsychotic agents in augmenting SSRIrefractory obsessive–compulsive disorder. Journal of Clinical Psychiatry 70, 863–868. McDougle, C.J., Goodman, W.K., Leckman, J.F., Lee, N.C., Heninger, G.R., Price, L.H., 1994. Haloperidol addition in fluvoxamine-refractory obsessive–compulsive disorder; a double-blind, placebo-controlled study in patients with and without tics. Archives of General Psychiatry 51, 302–308. Nakajima, T., Nakamura, M., Taga, C., Yamagami, S., Kiriike, N., Nagata, T., Saitoh, M., Kinoshita, T., Okajima, Y., Hanada, M., Tazoe, H., Yamaguchi, K., 1995. Reliability and validity of the Japanese version of the the Yale–Brown Obsessive Compulsive Scale. Psychiatry Clinical Neurosciences 49, 121–126. Pallanti, S., Quercioli, L., 2006. Treatment-refractory obsessive–compulsive disorder; methodological issues, operational definitions and therapeutic lines. Progress in Neuropsychopharmacology & Biological Psychiatry 30, 400–412. Pertusa, A., Fullana, M.A., Singh, S., Alonso, P., Menchon, J.M., Mataix-Cols, D., 2008. Compulsive hoarding; OCD symptom, distinct clinical syndrome, or both? American Journal of Psychiatry 165, 1289–1298. Rosario-Campos, M.C., Miguel, E.C., Quatrano, S., Chacon, P., Ferrao, Y., Findley, D., Katsovich, L., Scahill, L., King, R.A., Woody, S.R., Tolin, D., Hollander, E., Kano, Y., Leckman, J.F., 2006. The dimensional Yale–Brown Obsessive–Compulsive Scale (DYBOCS); an instrument for assessing obsessive–compulsive symptom dimensions. Molecular Psychiatry 11, 495–504. Samuels, J.F., Bienvenu, O.J., Riddle, R.A., Cullen, B., Grados, M.A., Liang, K.Y., HoehnSaric, R., Nestadt, G., 2002. Hoarding in obsessive–compulsive disorder: results from a case-control study. Behaviour Research and Therapy 40, 517–528. Saxena, S., 2007. Is compulsive hoarding a genetically and neurobiologicaly discrete syndrome? Implications for diagnostic classification. American Journal of Psychiatry 164, 380–384. Saxena, S., Brody, A.L., Maidment, K.M., Smith, E.C., Zohrabi, N., Katz, E., Baker, S.K., Baxter, L.R., 2004. Cerebral glucose metabolism in obsessive–compulsive hoarding. American Journal of Psychiatry 161, 1038–1048. Saxena, S., Brody, A.L., Maidment, K.M., Baxter, L.R., 2007. Paroxetine treatment of compulsive hoarding. Journal of Psychiatry Research 41, 481–487. Steketee, G., Frost, R., 2003. Compulsive hoarding; current status of the research. Clinical Psychology Review 23, 905–927. Summerfeldt, L.J., Richter, M.A., Antony, M.M., Swinson, R.P., 1999. Symptom structure in obsessive–compulsive disorder; a confirmatory factor analytic study. Behaviour Research and Therapy 37, 297–311. Tolin, D.F., Frost, R.O., Steketee, G., 2007. An open trial of cognitive–behavioral therapy for compulsive hoarding. Behaviour Research and Therapy 45, 1461–1470.