The finding that lifetime conduct problems, FH of AUDs, prior THC use, and male gender were each individually associated with drinking onset is also consistent with previous research (Kuperman et al., 2001a, 2001b; McGue et al., 2001a, 2001b; Prescott & Kendler, 1999). This further highlights the importance of controlling for these robust risk factors when examining the relation between AFD and drinking problems later in life. Therefore, it may be premature to conclude that a delay in the AFD by itself could decrease risk for development of AUDs (Grant & Dawson, 1998). While this is a logical step to consider, the approach should be tested in future longitudinal studies to provide greater support for such causal inferences. Continued follow-ups in the San Diego Prospective Study will allow for further testing of whether AFD is a unique predictor of alcohol-related outcomes in early adulthood. Such information will also be critical to test whether AFD mediates the effects of the risk factors identified here (conduct problems, FH, THC use, and male gender) on a range of substance use outcomes in early adulthood.
