Abstract
Accruing evidence of an association between drinking and smoking relapse suggests that it is important to measure alcohol use in smoking cessation studies. However, most studies do not do so, often because of the extra time burden required for these assessments. Data from participants (N=634) in two smoking cessation clinical trials were used to examine the relationship between short and longer periods of monitoring for a number of Timeline Followback (TLFB) drinking metrics at baseline and during treatment. High intercorrelations were found between short (7 and 14 days) and longer (30 and 60 days) time windows for baseline drinking data. Intercorrelations between short (last 7 days of treatment) and longer (entire treatment period) time windows of drinking data during the smoking cessation treatment period were also mostly in the high range. Although total abstinence was significantly overestimated with shorter time windows, for those who were misclassified, percentage of days abstinent was high and percentage of heavy drinking days and number of drinks per drinking day were low during the longer period. Thus, a brief estimate of alcohol use over 7 days at baseline is likely to provide a representative assessment of percentage of days abstinent, percentage of heavy drinking days, and number of drinks per drinking day. To estimate abstinence at baseline and during treatment, however, a more comprehensive period of monitoring may be required.