High prevalence of alcohol use disorders in 454 young adult offspring from the San Diego prospective study

Background: Preliminary evaluations of 212 drinking offspring from the San Diego Prospective Study (SDPD) indicated that over 50% developed alcohol use disorder (AUD) by their mid-20s. The present analysis evaluated if those findings remained robust when the group increased to 454 individuals, a sample size that facilitated a search for potential contributors to the high AUD prevalence.

Methods: Semistructured interviews were used to evaluate lifetime AUD diagnoses in 224 daughters and 230 sons from the SDPS (N = 454) by mean age 26. Analyses compared participants with and without AUD regarding demography, alcohol use, personality, and psychiatric diagnoses. Characteristics associated with AUD were entered together in a backward elimination regression analysis, and the results were entered in a structural equation model (SEM) to evaluate potential mediation of risks for alcohol problems.

Results: Lifetime AUD was documented for 61% of the sons and 41% of the daughters. Offspring with AUD reported averages of 13 maximum and five usual drinks per occasion and endorsed an average of 4 DSM AUD criteria. Even after considering personality characteristics, family AUD histories, and personal psychiatric histories, significant contributions to the regression analysis were limited to lower levels of response to alcohol, higher positive alcohol expectancies, and drinking to cope. Key elements of the hypothesized SEM were supported, and mediation between the low alcohol response and the number of alcohol problems was documented for expectancies, drinking to cope, and peer heavier drinking.

Conclusion: The results support prior high AUD rates in SDPS offspring and demonstrate that the AUD diagnoses were associated with robust alcohol intake and problems. The data also indicated mediation of the impact of the low alcohol response on the development of AUD through several characteristics proposed by prior work in other populations.