Validity of self-reported alcohol consumption in nondependent drinkers with unintentional injuries

Background: Self-report has become an anchor for alcohol assessment in the acute and primary care populations. The purpose of the study was to determi ne the validity of self-reported alcohol consumption after unintentional in juries iri hospitalized, nondependent drinkers. Methods: Non-alcohol-dependent subjects 18 years of age and older with unin tentional injuries (n = 209) were enrolled in the study and were interviewe d if they had either an admitting blood alcohol concentration (BAC) greater than or equal to 10 mg/dl (0.01 g/dl) or a positive screen for a history o f problem drinking. The self-reported number of standard drinks, time that drinking commenced, sex, and weight were used to calculate estimated blood alcohol concentration (EBAC), which was then compared to the admission BAG. Results: We had data to calculate the EBAC on 141 of the 209 subjects. Seve n men and no women with positive (greater than or equal to 10 mg/dl) BAC de nied drinking. Of the 134 subjects for whom we had data to calculate EBAC a nd who acknowledged drinking, mean BAC was 147.06 mg/dl and mean EBAC was 6 8.66 mg/dl. For women (n = 30), mean BAC was 149.53 mg/dl and mean EBAC was 114.67 mg/dl; for men (n = 104), mean BAC was 146.35 mg/dl and mean EBAC w as 55.38 mg/dl. The Spearman's rho between laboratory BAC and EBAC was 0.46 1 (p < 0.001) for all subjects (n = 134), 0.275 (NS) for women (n = 30), an d 0.532 (p < 0.001) for men (n = 104). For women and men separately, multip le regression analyses were performed to predict BAC by using weight and re ported number of drinks. For women, weight and number of drinks accounted f or 3% of the variance in laboratory BAC [r = 0.181, F(2,47) = 0.797, p = NS ]. In contrast, for men these same predictors accounted for 34% of the vari ance [r = 0.585, F(2,135) = 35.203,p < 0.001). Conclusions: Most nondependent patients with unintentional injury acknowled ged drinking before injury. After injury, women and men have different patt erns of reporting their drinking, with men more frequently underreporting b ut reporting more accurately and women more random in their self-reports.