We examined time-to-alcohol initiation in mixed effects Cox regression models using the coxme package (Therneau, 2020) for R (R Development Core Team, 2019). We examined the likelihood of alcohol use disorder diagnosis with generalized linear mixed effects models with a logit link using the Ime4 package (Bates et al., 2015). Because COGA included related individuals (i.e., individuals nested in families), these mixed effect models adjusted for familial clustering with a random intercept for family grouping.
