Citation

Sobell LC, Agrawal S, Sobell MB, Leo GI, Young LJ, Cunningham JA, Simco ER. J. Stud. Alcohol 2003; 64(6): 858-861.

Affiliation

Center for Psychological Studies, Nova Southeastern University, 3301 College Avenue, Fort Lauderdale, Florida 33314, USA. sobelll@nova.edu

Copyright

(Copyright © 2003, Rutgers Center of Alcohol Studies)

DOI

unavailable

PMID

14743950

Abstract

OBJECTIVE: Two major strategies have typically been used to assess recent drinking: (1) Daily Estimation (DE) measures such as the Timeline Followback (TLFB) and (2) Quantity-Frequency (QF) summary measures. Although QF measures provide a quick and easy measure of consumption, they have been criticized as not being able to capture sporadic and unpatterned drinking (e.g., days that reflect important social and/or health risks). The TLFB, a psychometrically sound drinking assessment method, is able to capture all drinking, including sporadic heavy days and unpatterned drinking. In some situations, however, recall of daily drinking may not be possible or practical (e.g., limited time; no resources). This article compares results obtained by using a QF measure and a DE measure to assess problem drinkers' pretreatment drinking. METHOD: The current study, part of a large community mail intervention with 825 alcohol abusers, compared results from two drinking measures covering the same time interval that were administered on two different occasions approximately 2.5 weeks apart. Both measures, the Quick Drinking Screen (QDS; a QF summary measure that collected data by telephone) and the TLFB (a self-administered daily estimation measure), collected drinking data for the year prior to the interview. RESULTS: Although the QDS and the TLFB are very different drinking measures, remarkably similar aggregate drinking data were obtained for five drinking variables. CONCLUSIONS: When it is not necessary or possible to gather detailed drinking data, the QDS produces reliable brief summary measures of drinking, at least for not severely alcohol dependent individuals. Also, respondents do not appear to use a repetitive response pattern when completing the TLFB.

Language: en