Obsessive-Compulsive Disorder: Course and Natural History

Course and Natural History

Age at Onset

Age at onset usually refers to the age when OCD symptoms (ob-sessions and compulsions) reach a severity level wherein they lead to impaired functioning or significant distress or are time-consuming (i.e., meet DSM-IV criteria for the disorder). Reported age at onset is usually during late adolescence. In one study drawn from an OCD clinic sample (N = 560), the onset for males occurred significantly earlier than for females (19.5 ± 9.2 years versus 22.0 ± 9.8 years). In this study, 83% of patients experi-enced the onset of significant symptoms between ages 10 and 24 years, whereas less than 15% experienced onset after age 35 years (Rasmussen and Eisen, 1998). People with OCD, however, usu-ally describe the onset of minor symptoms in childhood, well be-fore the onset of symptoms meeting full criteria for the disorder.

In several studies, earlier age at onset has been associated with an increased rate of OCD in first-degree relatives and sug-gest that there is a familial type of OCD characterized by early onset. Age at onset of OCD may also be a predictor of course. The vast majority of patients report a gradual worsening of obsessions and compulsions prior to the onset of full-criteria OCD, which is followed by a chronic course (see later). However, Swedo and colleagues (1998) have described a subtype of OCD that beginsbefore puberty and is characterized by an episodic course with intense exacerbations. Exacerbations of OCD symptoms in this subtype have been linked with Group A beta-hemolytic strep-tococcal infections, which has led to the subtype designation of pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections (PANDAS). In their study of 50 children with PANDAS, the average age of onset was 7.4 years. Whether the course of illness in patients with PANDAS continues to be episodic into adulthood, or, as is the case with postpubertal on-set, tends to be chronic, is not known.

In keeping with the older literature, a recent 2-year natu-ralistic prospective study of 65 adults with OCD, in which the effect of treatment was assessed, supported earlier findings that OCD is usually chronic with fluctuations in symptom intensity but no lasting remission; it is notable that this course was most common even during an era when effective treatments were avail-able. Although 50% of the subjects achieved partial remission in the first year of the study, the probability of subsequent relapse was 48%. Only 12% achieved full and sustained remission (Eisen et al., 1999). In contrast, a better outcome was found in a follow-up study of 144 people with OCD assessed in the 1950s and again in the 1990s (mean length of follow-up from illness onset was 47 years). Most subjects reported a significant decrease in OCDsymptom severity, which varied from complete recovery (20%), to recovery with continued subclinical symptoms (28%), to con-tinued OCD but with clear improvement (35%). Better outcome was associated with later age of onset and poorer social function-ing at baseline (Skoog and Skoog, 1999).

In a prospective study of children with OCD, the major-ity (52%) of the 25 patients had moderate to severe OCD in the 2- to 7-year follow-up period (Flament et al., 1990), which is con-sistent with the data on adults. A more recent prospective study of 54 children with OCD who were treated with clomipramine yielded a more hopeful picture of OCD’s course. At 2 to 7 years after initial referral, 43 of these patients still had symptoms that met criteria for OCD, but 73% were considered much or very much improved, and 11% were completely asymptomatic (Leonard et al., 1993). This study suggested that appropriate somatic treat-ment may improve outcome only while the patient continues to receive this treatment (see later).

Phenomenology

OCD’s clinical presentation is characterized by phenomenological subtypes based on the content of the obsessions and correspond-ing compulsions. The list of subtypes in the Yale-Brown Obses-sive–Compulsive Scale (Y-BOCS) (Table 51.1) was generated on

the basis of clinical interviews with OCD patients in the 1980s. These subtypes are remarkably consistent with phenomenologi-cal descriptions in the psychiatric literature beginning with scru-pulosity in the 15th century.

The basic types of obsessions and compulsions seem to be consistent across cultures. The most common obsession is fear of contamination, followed by pathological doubt, a need for symmetry and aggressive obsessions (Figure 51.1). The most common compulsion is checking, which is followed by washing, symmetry, the need to ask or confess and counting (Figure 51.2). Children with OCD present most commonly with washing com-pulsions, which are followed by repeating rituals.

Most patients have multiple obsessions and compulsions over time, with a particular fear or concern dominating the clini-cal picture at any one time. The presence of obsessions without compulsions, or compulsions without obsessions, is unusual. In the DSM-IV OCD field trial of 431 patients, only 2% had pre-dominantly obsessions and 2% had predominantly compulsions; the remaining 96% endorsed both obsessions and compulsions (Foa and Kozak, 1995). Patients who appear to have obsessions without compulsions frequently have unrecognized reassurance rituals or mental compulsions, such as repetitive, ritualized pray-ing, in addition to their obsessions. Pure compulsions are also unusual in adults, although they do occur in children, especially in the young (e.g., 6 to 8 years of age). Most people have both mental and behavioral compulsions; in the DSM-IV field trial 79.5% reported having both mental and behavioral compulsions, 20.3% had behavioral compulsions only and 0.2% had only men-tal compulsions.

The search for whether specific obsessions and compul-sions have predictive value in terms of treatment response biologic markers, or genetic transmission has not been particu-larly fruitful. There has been considerable interest in exploring whether certain clusters of obsessions and compulsions repre-sent specific OCD phenotypes. A number of studies have ad-dressed this question systematically using the Y-BOCS Symp-tom Checklist to identify groups of obsessions and compulsions that cluster together on factor analysis. Several studies have found between three and five such symptom dimensions: sym-metry/ordering, hoarding, contamination/cleaning, aggressive obsessions/checking and sexual/religious obsessions. The sym-metry dimension has been associated with comorbid tic disor-der; in one study, patients who scored high on this dimension had a relative risk for chronic tic disorder that was 8.5 times higher than those scoring low on this factor (Leckman et al., 1997). It appears that these symptom dimensions are stable over time, that is, although a patient’s specific obsessions and com-pulsions may change over time, new obsessions and compul-sions that develop are often within the same symptom dimen-sion as the previous symptoms. A study using positron emission tomography to evaluate neural correlates of these symptom di-mensions suggests that dysfunction in separate regions of the brain (e.g., striatum and prefrontal cortex) may mediate these factors (Rauch et al., 1998).

Data were analyzed from a number of placebo-controlled serotonin reuptake inhibitor (SRI) treatment studies to assess whether symptom factors or dimensions were associated with treatment response. No clear pattern emerged except that patients with hoarding obsession had a significantly poorer response to SRIs. Whether these identified dimensions are associated with response to behavioral treatment, biological markers, or genetic transmission has yet to be investigated.

The following descriptions of some common obsessions and compulsions illustrate the clinical presentation of these symptoms. In some cases a particular symptom may belong to more than one “type” of obsessive or compulsive grouping. Thus it is often up to the clinician to decide which category to place a symptom so that it best describes the patient’s symptoms overall; it may even be best to classify it in more than one category. For instance, a patient who has concerns about cancer may have handwashing as a compulsion related to her somatic obsession. If this is the only reason that she washes her hands, to avoid get-ting cancer, you might simply classify this as a somatic obsession with the accompanying compulsion. However if the patient also washes repeatedly to avoid contamination in general, not just for cancer, she would have both contamination and somatic obses-sions and compulsions.