Psychometric Validation of the Brief Alcohol Social Density Assessment in an Adult Community Sample.

Abstract

Individuals' social networks exert a strong influence on alcohol use, but valid assessment of network drinking behavior is typically lengthy and high in participant burden. The aim of this study was to validate the Brief Alcohol Social Density Assessment (BASDA), an efficient measure of perceived alcohol use within a person's social network, in a sample of adult drinkers from the general community. Specifically, the convergent, criterion-related, incremental validity and internal validity were investigated by examining the BASDA in relation to other established measures of drinking motives, weekly drinking level, and severity of involvement. Participants were 903 (56% female) adults who reported drinking in the last year and who completed the BASDA, the Drinking Motives Questionnaire, the Daily Drinking Questionnaire, and the Alcohol Use Disorders Identification Test (AUDIT). Significant positive correlations were found between the BASDA and drinking motives, drinking quantity, and the AUDIT (rs=0.21 to 0.51, ps<0.001), providing support for convergent validity. There was a significantly higher BASDA score for those scoring at or above an AUDIT cutoff for hazardous drinking (p<0.001), providing support for criterion-related validity. Finally, beyond motives and covariates, the BASDA was significantly associated with total AUDIT score (ΔR2 =0.09, p<0.001), indicating its additive contribution and providing support for incremental validity. Confirmatory factor analysis revealed excellent fit, and all items significantly loaded onto a single factor (p<0.0001), providing evidence of internal validity. The resulting alcohol social density latent variable was significantly and robustly associated with drinks per week and AUDIT total score. These findings provide further support for the BASDA as a valid and efficient measure of social network alcohol density for understanding social influences on alcohol misuse.