Information processing bias and pharmacotherapy outcome in older adults with generalized anxiety disorder

Information processing bias was evaluated in a sample of 25 older adults with generalized anxiety disorder (GAD) over the course of 12 weeks of escitalopram pharmacotherapy. Using the CANTAB Affective Go/No Go test, treatment response (as measured by the Hamilton Anxiety Rating Scale, Penn State Worry Questionnaire, and Generalized Anxiety Disorder Severity Scale) was predicted from a bias score (i.e., difference score between response latencies for negative and positive words) using mixed-models regression. A more positive bias score across time predicted better response to treatment. Faster responses to positive words relative to negative words were associated with greater symptomatic improvement over time as reflected by scores on the GADSS. There was a trend toward significance for PSWQ scores and no significant effects related to HAMA outcomes. These preliminary findings offer further insights into the role of biased cognitive processing of emotional material in the manifestation of late-life anxiety symptoms.

► Cognitive processing of emotional words was evaluated in a sample of 25 older adults with generalized anxiety receiving pharmacotherapy.
► We measured biased processing by subtracting the speed of response to negative words from the speed of response to positive words.
► Faster responses to positive words relative to negative words were associated with better response to medication.
► Findings represent a step toward understanding the role of emotional information processing bias in anxiety in later life.