Top hits identified in the first phase of analyses were regressed against the likelihood of drinking cessation at different time-points during pregnancy, in crude analyses as well as adjusting for pre-pregnancy alcohol consumption levels (likely on the causal pathway).
