Validation of the Comprehensive Effects of Alcohol Questionnaire in Older Adults.

Background: Alcohol expectancies have been established as an important construct in the prediction of alcohol use among younger adults, but there is little understanding about the perceived expected effects of consuming alcohol in older adults. Beliefs about the expected effects of alcohol may be quite different in older adults compared to young adults, which may also play a different role in their drinking behavior. In older cohorts there may be stigma may be associated with drinking and the nature of drinking experiences may differ to those of younger adults. Existing measures of expectancies are based on the perspectives of younger adults and therefore warrant validation in an older sample. Objectives: The aim of this study is to validate the comprehensive effects of alcohol questionnaire (CEOA) in a sample of older adults, and to investigate the relationships between alcohol expectancies, their evaluations and alcohol use. Methods: A sample of community dwelling older adults (N = 473) completed self-report measures assessing drinking behavior and alcohol expectancies. Exploratory factor analysis, confirmatory factor analysis, and hierarchical multiple regressions were conducted. Results: The final model of the CEOA supported two global dimensions of positive and negative expectancies, and showed excellent goodness-of-fit and internal consistency. Male drinkers were more likely to endorse positive expectancies and favorable evaluations of positive and negative expectancies. Favorable evaluations of positive expectancies were the strongest predictors of drinking frequency and quantity. Conclusions/Importance: The results support for the empirical validity of a two-factor expectancy model consisting of a 10-item negative expectancy factor and a 10-item positive expectancy factor. This more concise version could offer a more acceptable and time efficient measure of expectancies and valuations for older adults. Furthermore, the study provides important findings for the role of evaluations in their prediction of alcohol use.