Assessing older adults with generalized anxiety: a replication and extension.

Anxiety is a major health problem for older adults. The cornerstone for further work in this area is research that establishes the psychometric utility of standardized measurement strategies to characterize anxiety in older adults. The goals of the current study were to replicate and extend prior research addressing the psychometric properties of five self-report measures of anxiety in a sample of 57 older adults with generalized anxiety disorder (GAD). Data addressed the descriptive characteristics, internal consistency, test-retest reliability, interrelation of subscales, and convergent and divergent validity of the Penn state worry questionnaire (PSWQ), worry scale (WS), Spielberger state-trait anxiety inventory - Form Y (STAI), and two versions of the fear questionnaire (FQ). Descriptive data generally replicated prior findings, with some increased state and trait anxiety in the current sample. Measures were internally consistent, with evidence that a revised version of the FQ that focuses on severity of fear irrespective of associated avoidance demonstrated greater internal consistency than the original version. Measures of content-specific fears and worries (WS, FQ) were stable over time and demonstrated convergent validity. Evidence of convergent validity also was evident for measures of worry and general anxiety (WS, PSWQ, STAI-Trait), but two of these measures (PSWQ, STAI) were not reliable over time. Only the PSWQ showed evidence of divergent validity with respect to self-report measures of depression.