Autism spectrum traits in children and adolescents with obsessive-compulsive disorder (OCD)

Abstract

Objective

Assess the prevalence of autistic traits (AST) in pediatric obsessive-compulsive disorder (OCD) and relate them to OCD co-morbidity and compare them with published normative data.

Methods

Pediatric patients with obsessive-compulsive disorder (n = 109) according to the DSM-IV were studied using parent ratings of the Autistic Symptom/Syndrome Questionnaire to assess AST symptoms as a continuous rather than categorical trait. The KSADS, a semi-structured psychiatric interview, was used for the psychiatric diagnostic evaluation. Also, the Children's Yale-Brown Obsessive-Compulsive Scale was used to assess OCD severity and other clinical features.

Results

AST was common among our patients. Symptom scores were highest in cases with co-morbid Autistic Spectrum Disorders, but cases with other co-morbidities as tics/Tourette and attention/behavioral disorders also scored higher. All sub-groups, including OCD without these co-morbidities scored higher than the Swedish normative group. Using ANOVA, co-morbid ASD and tics/Tourette (plus a term for gender by tic interaction indicating that girls with tics scored high, otherwise low) and pathological doubt contributed (R² = .41) to the AST-traits, while OCD severity and co-morbid anxiety- and depressive disorders did not.

Conclusion

AST traits are prevalent in OCD and seem to be intricately associated with the co-morbidities as well as the OCD syndrome itself. The findings might have implication for our nosological understanding of OCD which currently is discussed.

Introduction

Obsessive-compulsive disorder (OCD) is a not uncommon disorder among children and adolescents (Heyman et al., 2001), has often a chronic course (Stewart et al., 2004) and is classified in the DSM as an anxiety disorder in view of the anxiety associated with obsessions and the function of rituals to ease this anxiety. However, there has been considerable debate about the nosological placement of OCD recently (Bartz & Hollander, 2006). This is the consequence of many aspects of OCD phenomenology. One reason is the remarkably heterogeneous features found in OCD patients (McKay et al., 2004). For one, the OCD phenomenology itself shows many diverse features (Calamari, Wiegartz, & Janeck, 1999; Ivarsson & Valderhaug, 2006; Leckman, Grice, Boardman, & Zhang, 1997). Following this lead, a term, OC spectrum disorders has been coined as an “umbrella concept” to express similarities that are found between OCD and various disorder, for example hypochondriasis, body dysmorphic disorder, trichotillomania, but also tic disorders and autism (see Bartz & Hollander, 2006) for a recent review of this issue).

Other grounds for this debate are findings from the way other disorders are associated with OCD in family studies (Grados et al., 2001; Hanna, Himle, Curtis, & Gillespie, 2005; Pauls, Alsobrook, Goodman, Rasmussen, & Leckman, 1995) and findings from studies of OCD co-morbidity patterns (Geller et al., 2000, Hanna et al., 2002, Ivarsson et al., in press). Furthermore, findings from the study of endophenotypes, i.e. the association between different anxiety disorders and the neurocircuitry involved in the symptom formation have fuelled the debate as well as (Bartz & Hollander, 2006).

One reason for the question if OCD is properly placed among the anxiety disorders is the overlap phenomenologically between OCD and the autism spectrum disorders (ASD). Many OCD patients are characterized by repetitive behaviors just like ASD-patients are. In OCD, many patients have ordering and symmetry compulsions, as well as repetition compulsions harboring a wish for a “just right” feeling (Rapoport, 1989; Rasmussen & Eisen, 1990), even to the extent that this can be thought of as a particular OCD-factor, i.e., “Symmetry and Ordering” (Baer, 1994, Leckman et al., 1997) or a group of patients with such symptoms (Calamari et al., 1999; Ivarsson & Valderhaug, 2006). Likewise some patients with strong traits of hoarding (Baer, 1994, Leckman et al., 1997) seem to be attached to objects in a way that is akin to that seen in ASD.

ASD is frequently a co-morbid problem in OCD, both in pediatric (Ivarsson et al., in press) and in adult populations (LaSalle et al., 2004), at levels that exceeds those that are found in the general population (Ehlers & Gillberg, 1993) (for a review see Williams, Higgins, and Brayne (2006)). Also, family members of ASD probands show compulsive personality traits, something that further emphasizes the link between the disorders.

Furthermore, Bejeroth found a subgroup of adult OCD-patients with autistic traits shown among other things through low sociability (Bejerot, Nylander, & Lindstrom, 2001). These findings are similar to findings by (Ivarsson & Winge-Westholm, 2004), who studying temperament in OCD, found that roughly half of the patients were characterized by low levels of activity, high levels of shyness and low levels of sociability. Do those individuals that are most extreme with regard to low sociability border on ASD pathology?

If OCD might be thought of as less related to the anxiety disorders and more fundamentally related to other OC-spectrum disorders including ASD (Bartz & Hollander, 2006), several implication ensue. One would be that pediatric OCD-patients ought to show more clear-cut ASD traits, i.e. the ASD relationship should not only be expressed through the OCD-symptoms, but as core ASD symptoms present in the patients to a significant degree, both above and below the level for an ASD diagnosis. Thus, OCD-patients should have ASD symptoms that could be perceived by people who know them well, i.e. parents or partners. While autistic traits could be noted in adults with OCD as rated by the clinician (Bejerot et al., 2001), ratings by the parents in pediatric OCD ought to be more reliable, especially using valid and reliable methods developed for this purpose. However, a serious problem arises with regard to the interpretation of possible ASD-traits in OCD. When above the diagnostic threshold, should these ASD-traits be seen as a case of co-morbidity, i.e. the co-occurrence of two separate disorders or is there “really” just one disorder with expression of divergent symptoms, i.e. OCD and ASD in these cases? What about ASD traits below the threshold even for an ASD UNS diagnosis? Are those related to the OCD-disorder per se, or are they related to other co-morbidities than ASD, e.g. ADHD and tic/Tourette, that can show such traits too (Ehlers, Gillberg, & Wing, 1999)? As we factor out other contributing factors, what, if anything, remains?

The aim of the present study was to study ASD symptoms in pediatric OCD-patients using a dimensional approach, as well as the categorical/diagnostic approach and to compare them with previously published data on ASD symptoms in different clinical groups to study the pattern of relationships with ASD, co-morbidities and OCD.