Drinking behavior among low-income older adults: a multimethod approach to estimating alcohol use

Substance abuse is the fastest growing health concern for older adults. Heavy drinking among older persons is associated with an increased risk of health consequences such as diabetes, cognitive impairment, sleep issues, and depression. It is important to note, however, the prevalence estimates of alcohol use among older adults are often based on inconsistent methodology. To address these potential methodological shortcomings, this study examines drinking patterns among low-income older adults using both self-report and unobtrusive methods.

The study was conducted in a low-income residential senior center in the United States. A total of 174 participants, aged 60 years or older, completed 2 self-administered cross-sectional surveys. A bogus recycling program was implemented to assess the amount of alcohol consumed by residents. Logistic regression analysis was utilized to model predictors of drinking status and to determine predictors of 3 category Alcohol Use and Disorders Identification Test scores. Bivariate associations that predicted associations with alcohol use were included in the final multivariate model. Alcohol containers collected from recycling were converted to standard drink estimates in order to calculate the capital consumption of residents.

About 40% of respondents reported consuming alcohol and 25% reported drinking at least once a week. On average, a total of 1,079 drinks were consumed per month. There were 3 significant predictors of drinking status: age, education, and diagnosis of diabetes. Additionally, there appears to be an increase in recycled alcohol containers coinciding with the time residents received their social security checks.

Overall, the combination of self-report and unobtrusive measures of alcohol consumption has potential to highlight different aspects of drinking behavior with a population living in a single dwelling such as a senior center apartment complex, residential hall, and the like.