Obsessive– Compulsive Disorder 451
reported that washing/contamination, checking/doubt, order/symmetry, and hoarding
were the primary OCD subtypes (Rachman & Hodgson, 1980; Rasmussen & Eisen,
1998). However, in the last few years a series of more rigorous, empirical studies have
addressed OCD subtypes through factor analytic and cluster analysis of OCD symptom
measures like the Yale–Brown Obsessive– Compulsive Scale (Y-BOCS; Goodman et al.,
1989a, 1989b). Although there have been considerable inconsistency across studies, it
would appear that the symptom presentation of individuals with OCD can be roughly
classified into contamination/washing, harm obsessions/checking, pure obsessions with-
out overt compulsions, and hoarding (e.g., Calamari et al., 2004; see McKay et al.,
2004). These classifications may have some limited clinical utility in predicting treat-
ment response, with some evidence that those with predominantly hoarding and pure
obsessional symptoms may have a poorer response to standard CBT and pharmaco-
therapy for OCD (e.g., Abramowitz, Franklin, Schwartz, & Furr, 2003; see D. A. Clark
& Guyitt, 2008; Steketee & Frost, 2007). Specialized cognitive- behavioral treatment
protocols have been proposed for pure obsessions (Rachman, 2003), fear of contamina-
tion (Rachman, 2006), and hoarding (Steketee & Frost, 2007).
Caution must be raised before concluding that different types of OC symptom pre-
sentation will require their own unique cognitive treatment protocol. Radomsky and
Taylor (2005) raise a number of conceptual and methodological problems with OCD
subtype research, not the least being the probability that a dimensional approach to
symptoms may be more valid than a categorically based perspective (e.g., Haslam,
Williams, Kyrios, McKay, & Taylor, 2005). Furthermore, most individuals with OCD
have multiple obsessions and compulsions that cut across categories and the majority
of OCD patients will show change in their OC symptoms over the course of the illness
(e.g., Skoog & Skoog, 1999). Other researchers have investigated whether OCD samples
could be categorized according to cognitive variables such as type of OC-related dys-
functional beliefs. However, the initial studies proved somewhat discouraging, with the
most robust finding that individuals simply fall into a high and low belief group (Cala-
mari et al., 2006; Taylor et al., 2006). Given that there is substantial overlap in the treat-
ment strategies utilized in these specialized CBT packages, we believe that a thorough,
individualized cognitive case conceptualization is the most efficient clinical strategy for
dealing with the idiosyncratic and heterogeneous symptom presentation in OCD.
epiDemiology anD CliniCal featuresPrevalence
OCD has a lifetime prevalence rate of approximately 1–2% in the general population,
with 1-year estimates ranging from 0.7 to 2.1% (Andrews et al., 2001; Kessler, Ber-
glund, et al., 2005; Kessler, Chiu, et al., 2005; Regier et al., 1993; Weissman et al.,
1994). Moreover, an even larger number of nonclinical individuals experience milder
and less frequent obsessional phenomena that would not meet diagnostic criteria (e.g.,
Bebbington, 1998; Burns, Formea, Keortge, & Sternberger, 1995; see Rasmussan &
Eisen, 1998). Slightly more women than men develop OCD, with age of onset typically
between midadolescence to late 20s (Rachman & Hodgson, 1980; Rasmussen & Eisen,
1992; Kessler, Berglund, et al., 2005; Weissman et al., 1994). Men typically have an
earlier age of onset than women and so tend to begin treatment at a younger age (e.g.,