The current data require several additional comments. First, in many clinical and research settings information might not be readily available to allow for the use of covariates. Thus, it is reassuring to note the similar patterns for how alcohol challenge and SRE scores correlated with current and future maximum drinks and alcohol problems in the zero-order correlations in Tables 1 and 2 compared with analyses that considered covariates in Tables 5 and 6. In addition, because the same underlying LR trait is likely to affect drinking behaviors at different time points in a person’s life (including Time 1 and Time 2 in the current analyses) covarying analyses for the same drinking-related characteristic from Time 1 when considering Time 2 could underestimate the true impact of LR. That said, many Time 2 analyses of the impact of Time 1 LR that covaried for Time 1 drinking characteristics were still significant.
