An evaluation of the level of response to alcohol, externalizing symptoms, and depressive symptoms as predictors of alcoholism.

Objective: The development of alcohol-use disorders (AUDs) reflects a complex relationship between genetic influences and environmental/cultural forces. Some genes operate through intermediate phenotypes, including a low level of response (LR) to alcohol, externalizing symptoms (EXT), and internalizing characteristics such as depressive syndromes (DEP). This article evaluates the impact of these three intermediate phenotypes and additional domains in a structural equation model (SEM).

Method: Data were available from baseline at approximately age 20 for LR, as well as from additional domains at the 10- and 15-year follow-up periods for 393 men from the San Diego Prospective Study. Correlational analyses and an AMOS-based SEM were used to evaluate the development of alcohol problems, including AUDs, with the hypothetical model based on results from prior studies evaluating each key intermediate phenotype separately.

Results: The SEM explained 51% of the variance of the 15-year outcome, and had good fit characteristics. The family history ofAUDs (FHalc) was linked, directly or indirectly, to all three key domains. The combination of LR and EXT mediated the relationship between FHalc and 15-year alcohol outcomes, with a trend (p = .07) for LR to mediate between FHalc and the 10-year outcome. DEP, by itself, did not mediate FHalc to alcoholism. The LR predicted the 15-year outcome both through alcohol problems at 10 years and via drinking to cope (COPE), with each of these domains functioning as mediators. The relationship of EXT to outcome was mediated by alcohol expectations (EXPECT) and by COPE. DEP added to the model in the context of an FH of independent depressions, stress, and lower social supports, subsequently affecting COPE.

Conclusions: The results indicate that the development of AUDs reflects several genetically influenced endophenotypes in the context of multiple additional domains. Both EXPECT and COPE represented important pathways through which the phenotypes influenced the AUD risk.