Discriminant validity of the symptoms constituting the DSM-III-R and DSM-IV associated symptom criterion of generalized anxiety disorder

Abstract

The associated (somatic) symptom criterion for DSM-IV generalized anxiety disorder (GAD) requires at least 3 of a list of 6 symptoms that previously resided in the Motor Tension (MT) and Vigilance and Scanning (VS) clusters of DSM-III-R. The decision to retain these six symptoms while eliminating all symptoms from the Autonomic Hyperactivity (AH) cluster, was made with limited data on discriminant validity and classification accuracy of the proposed DSM-IV associated symptom criterion. The present study examined this issue in a large sample ( N = 390) of patients with a DSM-III-R anxiety or mood disorder, all of whom were administered the GAD associated symptom ratings from the Anxiety Disorders Interview Schedule-Revised. Scores from the MT, VS, and DSM-IV symptom clusters correlated more strongly ( ps < .05) with measure of GAD (e.g., Penn State Worry Questionnaire) than did AH scores. Analyses that differentiated patients with GAD from other anxiety disorder groups produced the largest effect sizes using MT, VS, and DSM-IV cluster scores; GAD differed from mood disorders only on scores from the AH cluster. Whereas 98.6% of patients with GAD met the “3 of 6 symptoms” criterion proposed in DSM-IV, a large percentage of patients with other diagnoses met this criterion also (range = 35.3% to 92.9%). Raising the criterion to require four or more symptoms increased specificity without adversely affecting sensitivity. Implications for conceptual models of GAD are considered.