SUN-005 Do Peer, Parental, and Personality Factors Explain the Effects of Early Life Stress and Testosterone on Female Alcohol Initiation?

Abstract Disclosure: A. Donovan: None. C. Grella: None. S. Assari: None. M. Shaheen: None. L. Richter: None. T.C. Friedman: None. Background: Early life stress (ELS) is associated with early adolescent substance use initiation. ELS alters stress and sex hormone levels during adolescence, a time of increased neuroplasticity. Previous findings indicated a significant effect of early life stress (ELS) and a sex-specific effect of testosterone on the rate of alcohol initiation (defined as consuming a full drink of alcohol) among females. This effect may be due to a more advanced pubertal status resulting in association with older, substance-initiated peers. Additionally, higher testosterone levels are associated with increased trait impulsiveness. Though neither of these peer or personality factors were correlated with testosterone, inclusion of these factors may explain the association between testosterone levels, ELS domain predictors, and female alcohol use. Methods: Baseline measures from a nationally representative longitudinal cohort study, the Adolescent Brain Cognitive Development study, were used to examine factors related to ELS by domain (ELS Environmental, Family, Trauma), pubertal development, salivary testosterone level (pg/mL), age in months, peer alcohol use, parental rules regarding alcohol, availability of alcohol, impulsiveness (UPPS), and alcohol initiation. Sex-stratified analysis was performed to generate descriptive statistics followed by a proportional hazards cox regression using a complex samples method. Youth at baseline (age 9.9 SD ±0.6, N= 7290, 47%F) were followed up through wave 3 (age 12-13) for alcohol initiation. Results: Any peer alcohol use increased the rate of alcohol initiation by 70% (p= 0.02) among females but not males. Easy availability of alcohol among females increased the rate of initiation by 47% (p=0.03), while rules about alcohol use were not significant for males or females. A 1 pt increase in total UPPS score increased the rate of initiation among females by 4% (p<0.01). These additional factors did not explain the effect of testosterone on female alcohol initiation, and each 10 pg/mL increase in testosterone was associated with a 17% increase in the rate of first use (p=0.02) among females. Furthermore, ELS Family score remained a significant predictor of alcohol initiation (11% increase in rate of first use for each 1pt increase in score) for both females (p=0.03) and males (p=0.05). Conclusions: Though none of the ELS domains or testosterone effects were explained by the additional factors, our findings describe a sex-specific mix of peer, parental, personality, and hormonal direct effects on alcohol use initiation as well as an overall effect of family-related early life stress for both sexes. These findings have implications for developing prevention interventions that address both sex-specific and general risks of alcohol use initiation. Presentation: Sunday, July 13, 2025