Center for Research in Neuropsychology and Cognitive Behavioral Intervention

CHARACTERIZATION: A self or hetero-response instrument designed to evaluate anxious symptoms in older adults. STRUCTURE AND SCORING: 20 items answered through two response alternatives, “I agree” (1 point) or “disagree” (0 points) considering the last week. The total score results from the sum of all the answers, varying from 0 points (without anxiety) 20 points (high anxiety). PSYCHOMETRIC CHARACTERISTICS: The mean of the GAI scores with 805 institutionalized older adults was 10.63 (SD = 6.84). The internal consistency was excellent (Cronbach’s α= 0.94, KR21 = 0.94). The factorial analysis of the scale structure revealed a one-component solution that explained 53.4% of the variance. Confirmatory analysis confirmed a single factor solution (χ2/df = 2.81, TLI = 0.96, CFI = 0.96, RMSEA = 0.05). The convergent validity was adequate, with a correlation of 0.80 (p < 0.001) with the Geriatric Depression Scale and 0.64 (p < 0.001) with the Negative subscale of the Positive and Negative Affect Schedule. The divergent validity was also adequate, with negative correlations with the Positive Affect subscale (r = -0.27; p < 0.001) and with Satisfaction With Life Scale (r = -0.17; p < 0.001). GAI discriminated older adults without anxiety disorder (M = 5.38, SD = 5.72) from older adults with generalized anxiety disorder (M = 14.47, SD = 4.22). CUTOFF POINT: The analysis of the cutoff scores through the Receiver Operating Curve (AUC = 0.92) revealed the value of 13 points (sensitivity = 100%, specificity = 84.7%).