Decline in alcohol consumption during two years of daily reporting

We conducted a formal analysis of the potential effects on alcohol consumption of daily reports to an Interactive Voice Response (IVR) system over a 2-year period.

Thirty-three white men who had been recruited from bars were invited to call a toll-free number daily to answer a 2-minute questionnaire regarding alcohol consumption the previous day, reasons for drinking or not drinking, mood and stress levels, and a few other relevant items. A recorded voice asked the questions and callers responded using the telephone keypad. During the 2-year study, all subjects were personally interviewed each quarter, including regular Timeline Follow Back (TLFB) assessments. We compared mean consumption levels reported for the first year with those reported the second year.

Although there were consistent seasonal variations and holiday peaks, reported consumption declined over the 2 years of study. A comparison of average consumption across all subjects revealed significant declines in drinks per day, drinking days per week, and drinks per drinking day. Analysis of individuals showed at least some reduction in 82% of the subjects and statistically significant reductions in 45%. There was an estimated 19% reduction in consumption from Year I to Year 2 based on the IVR reports and a 21% reduction as ascertained by the TLFB. No significant changes were observed in nonalcohol-related measures.

After exploring possible reasons for our results, we conclude that the most likely explanation of the decline in alcohol consumption is a reactivity effect due to reporting consumption regularly. The results suggest that IVR may have potential utility as an intervention for excessive alcohol consumption.