Symptom dimensions in OCD and their association with clinical characteristics and comorbid disorders

Abstract

The complex clinical phenotype of obsessive–compulsive disorder (OCD) can be summarized in to a few temporally stable and consistent symptom dimensions that may have distinct clinical and neurobiological correlates. We examined the relationship between symptom dimensions and clinical characteristics in 161 consecutive patients with DSM-IV diagnosis of OCD with the Yale–Brown Obsessive–Compulsive Scale severity score of ≥20 recruited from a specialty OCD clinic in India. Clinician administered version of the Dimensional-Yale–Brown Obsessive–Compulsive Scale (D-YBOCS) was used to assess symptom dimensions. Earlier age of onset of OCD was associated with sexual/religious, aggression and symmetry dimensions. Fear of contamination was associated with female sex, higher family loading, greater severity of illness, poorer insight, and somewhat poorer functioning and lower physical quality of life. Aggression dimension was associated with presence of anxiety disorders and social phobia in particular. Our study demonstrates relatively specific associations between OCD symptom dimensions and clinical characteristics supporting the view that symptom dimensions could be employed to reduce the heterogeneity of OCD. The study encourages research on neurobiological and genetic underpinnings of symptom dimensions and supports inclusion of symptom dimensions in characterizing OCD in DSM-5 text.

Introduction

The exact etiology of obsessive–compulsive disorder (OCD) is unknown. Systematic search for the genetic basis of OCD has not yielded any meaningful and replicable findings (Grados et al., 2003, Samuels, 2009). This is possibly due to the heterogeneous nature of OCD (Miguel et al., 2005, Samuels, 2009, Stein, 2000). Given this heterogeneity, there have been various attempts to subtype OCD into homogenous entities using clinical phenomenological characteristics such as age at onset of OCD (Hemmings et al., 2004), comorbidity profile (e.g. tic disorders) (Miguel, do Rosario-Campos, Shavitt, Hounie, & Mercadante, 2001), familiality (Viswanath, Narayanaswamy, Cherian, Reddy, & Math, 2011) and recently symptom dimensions (Mataix-Cols, Rosario-Campos, & Leckman, 2005).

There is wide variation in the thematic content of obsessions and compulsions. Empirical evidence suggests that the diversity in OCD symptoms can be reduced to a fewer number of symptom dimensions; as a result OCD is increasingly seen as a dimensional disorder (Mataix-Cols et al., 2005). Most widely used approach to reduce the heterogeneity is to generate symptom dimensions based on the factor analysis of the symptoms such as the ones generated by the Yale–Brown Obsessive–Compulsive Scale (YBOCS) symptom checklist (Goodman et al., 1989). A recent meta-analysis of 21 factor analytic studies of the YBOCS symptom checklist involving 5124 participants identified four symptom dimensions: (a) symmetry obsessions; counting, ordering and arranging compulsions; (b) forbidden thoughts (aggressive, sexual, religious and somatic obsessions; and related checking compulsions); (c) contamination/cleaning, and (d) hoarding (Bloch, Landeros-Weisenberger, Rosario, Pittenger, & Leckman, 2008). These symptom dimensions have been reported to be temporally (Mataix-Cols et al., 2002) and cross-culturally stable (Matsunaga et al., 2008), with distinct neural correlates (Mataix-Cols et al., 2004) and comorbidity patterns (Hasler et al., 2005). Brain-imaging and genetic studies have also provided preliminary evidence for the biological validity of these dimensions (Gilbert et al., 2008, Hashimoto et al., 2011, van den Heuvel et al., 2009).

There is also some evidence that symptom dimensions may have specific relationships with certain clinical characteristics such as age of onset and comorbid patterns. An integration of symptom dimensions with clinical characteristics may help identify homogeneous subtypes that may in turn help in elucidating the neurobiological and genetic underpinnings of OCD. However, very few studies have systematically examined the relationship between specific symptom dimensions and clinical characteristics. A recent study using factor and cluster analytic analyses to establish symptom dimensions, found associations between factor I (aggressive, sexual, religious and somatic obsessions, checking compulsions) and co-morbid anxiety disorders/depression, factor II (obsessions of symmetry, repeating, counting and arranging/ordering compulsions) and co-morbid bipolar disorders/panic disorder/agoraphobia and factor III (obsessions of contamination and washing compulsions) and tics/eating disorders (Hasler et al., 2005). In addition, factor I and factor II were associated with early onset of OCD. In a Brazilian study that examined the trajectory of comorbid psychiatric disorders associated with OCD according to the first manifested disorder, those who presented with separation anxiety disorder as first diagnosis had higher scores on the sexual/religious dimension of the Dimensional-Yale–Brown Obsessive–Compulsive Scale (D-YBOCS) (Rosario-Campos et al., 2006) and on the measures of anxiety depression, and a higher frequency of post-traumatic disorder (de Mathis et al., 2012). In another large Brazilian study, tic-related OCD was associated with more severe scores on the aggressive, sexual/religious and hoarding symptom dimensions of the D-YBOCS in addition to its association with early onset, male preponderance, sensory phenomena, and comorbidity with anxiety and impulse control disorders and attention deficit hyperactivity disorder (de Alvarenga et al., 2012). Association between aggressive dimension and Tourette syndrome has also been reported in a study from Japan (Kano et al., 2012).

This study aimed to examine the relationships between symptom dimensions and clinical characteristics such as age of onset, comorbidity, familiality and insight using the D-YBOCS (Rosario-Campos et al., 2006). The scale summarizes complex OCD phenotype into a few consistent and temporally stable symptom dimensions. It is a reliable and valid instrument to assess multiple aspects of OCD symptom severity (Rosario-Campos et al., 2006). The D-YBOCS is a more appropriate instrument to assess various symptom dimensions and their severity than the conventional statistical method of factor analysis of symptom checklists. Conventional instruments such as the Y–BOCS assess obsessions and compulsions separately and generate total severity score instead of generating severity of individual dimensions. The D-YBOCS assesses OCD symptoms within six distinct dimensions that combine thematically related obsessions and compulsions. The instrument allows classifying same type of ritual under different symptom dimensions based on the functionality of the ritual rather than just the form of it. For example, checking can be secondary sexual/religious, aggressive, somatic or even contamination obsessions. Severity ratings individual symptom dimensions are reported to be largely independent of each other with only modest correlation with global severity. Accordingly, the D-YBOCS may be better suited than other symptom measures for investigating the relationship between OCD symptom dimensions and clinical characteristics. Based on the limited literature on the relation between symptom dimensions, we hypothesized distinctive association between certain symptom dimensions and clinical characteristics. Specifically, we expected aggression, symmetry and sexual/religious dimensions to be associated with early onset; and aggression and sexual/religious dimensions to be comorbid with anxiety and depressive disorders. Although the literature suggests specific relation between impulse control disorders, tic disorders, and eating disorders with certain symptom dimensions, we refrained from hypothesizing such a relationship in view of low rates of these disorders in our OCD samples (Jaisoorya et al., 2003, Viswanath et al., 2012).