Search

Objectives Assessing late-life anxiety using an instrument with sound psychometric properties including cross-cultural invariance is essential for cross-national aging research and clinical assessment. To date, no cross-national research studies have examined the psychometric properties of the frequently used Geriatric Anxiety Inventory (GAI) in depth. Method Using data from 3,731 older adults from 10 national samples (Australia, Brazil, Canada, The Netherlands, Norway, Portugal, Spain, Singapore, Thailand, and United States), this study used bifactor modeling to analyze the dimensionality of the GAI. We evaluated the “fitness” of individual items based on the explained common variance for each item across all nations. In addition, a multigroup confirmatory factor analysis was applied, testing for measurement invariance across the samples. Results Across samples, the presence of a strong G factor provides support that a general factor is of primary importance, rather than subfactors. That is, the data support a primarily unidimensional representation of the GAI, still acknowledging the presence of multidimensional factors. A GAI score in one of the countries would be directly comparable to a GAI score in any of the other countries tested, perhaps with the exception of Singapore. Discussion Although several items demonstrated relatively weak common variance with the general factor, the unidimensional structure remained strong even with these items retained. Thus, it is recommended that the GAI be administered using all items.

Background.The Geriatric Anxiety Inventory (GAI) is a widely used self-report measure of anxiety symptoms in older adults. Much research has been conducted on the psychometric properties of the GAI in various populations and using different language versions. Previous reviews of this literature have examined only a small proportion of studies in light of the body of research currently available and have not evaluated the methodological quality of this research. We conducted a systematic review of the psychometric properties of the GAI.Method.Relevant studies (N = 30) were retrieved through a search of electronic databases (Pubmed, PsycINFO, CINAHL, EMBASE and Google Scholar) and a hand search. The methodological quality of the included studies was assessed by two independent reviewers using the ‘‘COnsensusbased Standards for the selection of health status Measurement INstruments’’ (COSMIN) checklist.Results.Based on the COSMIN checklist, internal consistency and test reliability were mostly rated as poorly assessed (62.1% and 70% of studies, respectively) and quality of studies examining structural validity was mostly fair (60% of studies). The GAI showed adequate internal consistency and test-retest reliability. Convergent validity indices were highest with measures of generalized anxiety and lowest with instruments that include somatic symptoms. A substantial overlap with measures of depression was reported. While there was no consensus on the factorial structure of the GAI, several studies found it to be unidimensional.Conclusions.The GAI presents satisfactory psychometric properties. However, future efforts should aim to achieve a higher degree of methodological quality.

This study examined and compared the acceptability of three treatments for generalized anxiety disorder (GAD) in older adults: cognitive-behavioral therapy (CBT), guided self-help based on cognitive-behavioral principles (CBT-GSH), and selective serotonin reuptake inhibitors (SSRIs). Participants (N = 88; M age = 74.2 years, SD = 6.9) evaluated the acceptability of each treatment for the fictional case of a 70-year-old woman presenting mild to moderate or severe GAD. Results showed that CBT was more acceptable than SSRIs. Participants who indicated that they do not know what causes an anxiety disorder rated treatments as less acceptable. Also, the more participants considered the duration of treatment as an element influencing their choice of treatment for anxiety, the more they considered such treatment acceptable.

The Geriatric Anxiety Inventory (GAI) and a short form of this instrument (GAI-SF) were developed to assess the severity of anxiety symptoms in older adults in order to compensate for the lack of validated screening tools adapted to the elderly population. This study examined the psychometric properties of the French Canadian version of the GAI, in its complete (GAI-FC) and short form (GAI-FC-SF).A total of 331 community-dwelling seniors between 65 and 92 years old participated in this study.Both the GAI-FC and the GAI-FC-SF have sound psychometric properties with, respectively, a high internal consistency (α = .94 and .83), an adequate convergent validity (r = .50 to .86 with instruments known to evaluate constructs similar to the GAI or related to anxiety), a good test-retest reliability (r = .89 and .85), in addition to a single-factor structure.The results support the use of both the GAI-FC and the GAI-FC-SF. The GAI-FC-SF seems to be an interesting alternative to the GAI-FC as a screening tool when time available for assessment is limited.