Square Pegs and Round Slots: Evaluating Key DSM-5 Changes for Anxiety Disorders
Friday, May 23, 2014,
4:00 PM - 5:20 PM
These studies examine the validity and clinical implications of key DSM-5 definitional changes for the assessment of social anxiety disorder, specific phobias, and agoraphobia.
The recent release of DSM-5 brought numerous revisions to existing nosology. Some changes, such as a newly independent agoraphobia diagnosis and the replacement of previous social anxiety subtypes, remain controversial within the field. Other diagnoses such as specific phobia remained unchanged despite continued questions of validity. These studies evaluate the recently revised criteria and persistent questions across the anxiety disorders. Presenters will discuss empirical implications for future iterations of the “living DSM-5 document.”
The first study evaluates the new performance subtype of social anxiety disorder (SAD) in treatment-seeking anxious youth. Although research in adults supports a performance/public speaking subtype (Blote et al., 2009; Bogels et al., 2010) in favor of the previous “generalized subtype” of SAD, similar analyses in youth yield divergent results. The present analyses examined SAD symptoms in outpatient anxious youth (N=204). Using DSM-IV-TR criteria, participants were identified as “generalized” or “nongeneralized” subtype, and were evaluated for fit into a panel-generated performance subtype to match the new DSM-5 performance-only specifier. Linear regression analyses revealed significantly higher rates of comorbid depression among youth with generalized versus nongeneralized SAD (p < .05). Furthermore, though most participants (93.6%) endorsed performance items, all such cases also reported additional SAD items; no cases from the present sample met criteria for performance-only subtype, raising questions concerning the validity of the new DSM-5 nosology.
In the second study, we examine the classification of the newly independent agoraphobia diagnosis, and the new criteria requiring fear/avoidance across two or more distinct fear domains. The present study compared DSM-IV and approximate DSM-5 diagnoses of 1,000 adult outpatients with some amount of agoraphobic avoidance, as diagnosed by the ADIS-IV. Over ¼ of the patients were diagnosed differently between editions, with the majority being diagnosed as present in DSM-IV and absent in DSM-5 (sensitivity of .75, specificity of .7). This was despite similar levels of impairment between patients who were diagnosed as present by both editions and present in DSM-IV but absent in DSM-5 (e.g., mean ASI = 35.4 and 33.5, respectively). Given that one reason for this change was concern that agoraphobia was being under-identified and consequently under-treated (Wittchen et al., 2010), this lack of sensitivity raises serious concerns.
In the third study, latent class analyses were conducted to examine the presentation of Specific Phobias in an outpatient sample of anxious children (N=691, 53.8% female, MAge=12.04) assessed via the Anxiety Disorders Interview Schedule for DSM-IV. The presence of clinically significant specific fear intensities were entered as dichotomous indicators. A 3-class solution demonstrated the best fit (AIC= 7208.060, BIC=7445.664, Entropy=0.818, adj LMR p=.01, BLRT p<.001). The three classes were characterized as (1) Children with moderate probability of experiencing animal, natural environment (heights, thunder, and water), situational (transportation), and other (loud noises and choking) fears; (2) children with high probability of experiencing fear of shots and blood tests and moderate probability of experiencing fear of animals and darkness; (3) children with low probability of experiencing any specific fears. The clinical correlates and utility of these classes will be discussed.