Objective: To develop a shorter version of the Anxiety Disorder Scale (ADS) for use as a rapid screening instrument in primary care.
Design: Two-stage screening design. Primary care attenders aged 65 and over were screened for generalized anxiety in the surgery with the 11-item generalized anxiety subscale of the ADS (ADS GA), a selected subsample then proceeding to a clinical validation interview.
Interventions: None.
Main outcome measures: Scores on the ADS GA, non-hierarchical ICD-10 caseness for generalized anxiety established by brief clinical interview by an old age psychiatrist.
Results: The prevalence rate of generalized anxiety was 16% using the established cutpoint and showed an age-related decline. A cutpoint of 2-3/11 appeared to give optimal performance in this small sample (sensitivity 85%, specificity 77%, positive predictive value 52%), suggesting that 36% of elderly general practice attenders might be diagnosed as having generalized anxiety. A reduced four-item version gave a predicted sensitivity of 77%, a specificity of 83% and a positive predictive value of 63% (cutpoint 1-2/4).
Conclusions: A four-item version of the ADS GA, the FEAR (frequency of anxiety; enduring nature of anxiety; alcohol or sedative use; restlessness or fidgeting), has potential as a rapid screening instrument for use in primary care.