Stress and alcohol consumption in heavily drinking men: 2 years of daily data using interactive voice response.

Background: The relationship between stress and alcohol consumption has been shown in recent research to be more complex than originally thought. Cross-sectional and short-term longitudinal studies may not provide adequate data to address subtle but important relationships that impact consumption frequency and/or quantity.

Methods: Participants were 33 males recruited from local bars who reported their alcohol consumption, stress, and other related variables every day for 2 years. Reports were provided by automated telephone to a computer-based interactive voice response (IVR) system. By closely monitoring call completion, we were able to collect over 95% of the daily calls.

Results: Using hierarchical linear modeling, daily stress was found to be associated with daily alcohol consumption, but in the opposite direction than would be predicted by a "drinking to cope" hypothesis. That is, same-day stress was found to be inversely related to consumption levels. Prior day's stress was not significantly associated with subsequent day's consumption; however, prior day's consumption was predictive of subsequent stress for up to 2 days. Similar analyses using week, rather than day, as the unit of measure paralleled these findings.

Conclusions: This study highlights the utility of using long-term daily process data to address important research and clinical questions in the alcohol field.