Line Plot from Scientific Research

The predicted mean probabilities of each outcome across the distribution of z -standardized maternal polygenic scores, separately for prenatal heavy drinking. Not all measures were available in each cohort (for example, measures of heavy prenatal drinking were only available in the ALSPAC and MCS cohorts). Note, the E-Risk sample is a twin sample, hence rates of breastfeeding are lower than those in the ALSPAC and MCS cohorts. All effect sizes are reported for a 1-s.d. change in polygenic score. The error bars indicate 95% confidence intervals. The number of participants (mothers) included in the analysis were as follows: for prenatal heavy drinking ALSPAC n =7,144; MCS n =6,695.

DOI