Accurately Estimating Alcohol Consumption: A Comparison of Self-Administrated and Interview Methods

Abstract

Background: There have been two methods of gaining retrospective self-report estimates of alcohol consumption, quantity frequency (QF) and retrospective diary (RD), offering distinct advantages and disadvantages. The typical and atypical drinking diary (TADD) was developed to benefit from each method’s advantages. Objectives: To compare estimates obtained from the TADD with corresponding estimates obtained from a QF measure (Khavari alcohol test; KAT) and an RD measure, the timeline follow-back (TLFB). It was hypothesized that the TADD’s alcohol consumption estimates would correlate strongly with those obtained from the TLFB; and the strength of the TADD–TLFB correlation would be significantly stronger than the KAT–TLFB correlation. Method: One hundred and sixty-eight (168) undergraduates (73% female) estimated their alcohol consumption for the preceding 12 weeks using all three measures. Results: The TADD (r = 0.83, p < .001) and the KAT (r = 0.066, p < .001) produced estimates of total alcohol consumption that correlated very strongly and strongly, respectively, with the TLFB. Notably, the TADD–TLFB correlation was significantly stronger than the KAT–TLFB correlation (Z = 5.10, p < .001). Moreover, the TADD’s estimates of number of drinking days (r = 0.70, p < .001) and heavy drinking episodes per week (0.77, p < .001) correlated strongly with corresponding estimates from the TLFB. The TADD and KAT provided higher estimates of alcohol consumption than the TLFB. Conclusion: Clinicians and researchers should consider using the TADD to estimate alcohol consumption in time-limited situations.