Prenatal alcohol consumption and child IQ and cognition: a Mendelian randomization study

Background Previous studies using Mendelian randomization have found that fetal alcohol exposure may be associated with lower IQ and test scores in childhood. Objectives We aim to replicate and extend these findings in Generation R Study, a birth cohort based in Rotterdam, the Netherlands. Methods We used data from Generation R which recruited pregnant women between 2002 and 2006. Alcohol use was assessed via questionnaire during each trimester. IQ was measured in the children between ages 5 and 8 using the Snijders-Oomen Non-Verbal Intelligence Test. Scores from a national standardized test administered around age 12 were used as a measure of cognition. We estimated the associations between ten genetic variants in the mothers previously found to be related to alcohol consumption and metabolism and each of the outcomes. In the children, we also estimated the association between the same genetic variants as well as two polygenic scores for alcohol consumption and the outcomes. Results Maternal genetic variants were not found to be related to either outcome but wide confidence intervals did not preclude important effects. A few genetic variants in the children were suggestive of a decrease in IQ. Likewise, one genetic variant and the genetic score had estimates and confidence intervals consistent with increases in standardized test scores. Conclusions Our results provide slight support for associations between genetic variants in children related to maternal prenatal alcohol consumption and IQ and cognition outcomes. These findings are in line with two previous studies on this topic.