Alcohol challenge-based and SRE-based LR values overlap 60% in predicting alcohol outcomes (Schuckit et al., 2009a, b), and operate similarly when used to identify low and high LR subjects for further study (Trim et al., 2010). Furthermore, SRE- and alcohol challenge-generated LRs operate similarly when placed in the context of additional risk factors predicting outcomes in structural equation models (Schuckit et al., 2010, 2011). When the same individuals were given alcohol challenges and retrospective SREs at the same time, with correlations between the two LR measures of > −.20, a finding generally consistent with correlations and beta weights reported in another evaluation in the literature (Pederson and McCarthy, 2009; Schuckit et al., 2009b).
