Both ALSPAC and S4S participants were administered the SRE items across multiple assessments. For ALSPAC, we examined data from questionnaires/clinic visits at average ages 15.5, 16.5, and 17.5. S4S participants were also administered SRE items at average ages 18.5, 19.0, 19.9, and 21.0. For both samples, only participants who reported having initiated alcohol use were administered the SRE items; others were coded as “NA”. Where scores were available for a participant across multiple waves, the first score was used for GWAS, reasoning that this wave represented the assessment most temporally proximal to the initiation of alcohol use and therefore least susceptible to recall bias. The size of a standard drink, to which respondents are asked to refer when completing the SRE items, differs in the US and the UK (14g vs. 8g of ethanol, respectively). Therefore, raw SRE scores were standardized for GWAS analyses, ensuring that effect sizes observed across the ALSPAC and S4S samples were on the same scale.
