Survey Questions on Quantity and Frequency Are Differentially Effective by Age in Predicting Future Alcohol Consumption

Cross sectional research has demonstrated that screening tool questions on frequency of alcohol consumption are a better predictor of dependence and harmful drinking in younger adults; questions about quantity per occasion are a better predictor in older adults. The aim of this study is to see if this relationship also holds longitudinally. A total of 9076 respondents aged 15 and over completed at least two waves of the longitudinal annual Household Income and Labour Dynamics in Australia survey 10 years apart between 2001-2010 and 2012-2020. Standardised scores from responses to questions on drinking quantity and frequency in the first survey were used to predict consumption 10 years later in groups stratified by age. Frequency of consumption was a significantly better predictor of future consumption than quantity in younger drinkers (aged < 36; β = 9.3, 95% confidence interval [CI] 8.6-10.0), than older drinkers (aged > 49; β = 5.1, 95% CI 4.8-5.5) while quantity was a better predictor in older drinkers (β = 8.2, 95% CI 7.2-9.3) than younger drinkers (β = 3.4, 95% CI 3.1-3.7). Some commonly used screening items, such as drinking quantity and frequency, are differentially effective at identifying future heavy drinkers between age groups. Development of age-specific screening tools could potentially lead to more accurate identification of people who could benefit from intervention to reduce their alcohol consumption.