Date of Graduation


Document Type


Degree Type



Eberly College of Arts and Sciences



Committee Chair

Barry Edelstein.


Anxiety disorders are common in late-life and have detrimental effects of health and well-being (Stanley, Diefenbach, & Hopko, 2003; Wetherell et al., 2004). Generalized Anxiety Disorder (GAD) is the most common anxiety disorder in late-life (Beekman et al., 1998). Beyond the diagnostic symptoms, older adults with anxiety disorders may present with a variety of physical symptoms (Palmer, Jeste, & Sheikh, 1997). As a result, GAD is a heterogeneous disorder. Empirically-supported treatments are available for late-life GAD, but little is known about how people with different constellations of presenting symptoms respond to treatment, and whether they differ according to the way in which improvement is measured. The present study aimed to identify which presenting symptoms differentiated older adults with GAD from those without and how those symptoms predicted outcome across several domains of response. Participants were enrolled in a trial testing the efficacy of escitalopram for treatment of late-life GAD. One hundred and seventy-seven older adults with GAD and 41 older adults with no diagnosis participated. One hundred percent of the cases were correctly classified on the basis of the "Anxious Mood" and "Tension" SIGH-A items. A subset of participants with GAD who had been randomly assigned to received escitalopram and had completed 12 weeks of blinded treatment were further analyzed. Neither "Anxious Mood" nor "Tension" significantly predicted outcome in any domain. However, baseline scores on several of the outcome measures accounted for a significant amount of variance in week 12 scores, with lower scores being associated with better outcomes. These results indicate that baseline scores are the best predictors of outcome, and could have implications for treatment of GAD in late-life.