fifteenforty buries one of the biggest potential issues in his 4th para, so I'm going to highlight it: its not clear if this correlation was found post-hoc or n

fifteenforty buries one of the biggest potential issues in his 4th para, so I'm going to highlight it: its not clear if this correlation was found post-hoc or not. If you measure 10 factors and 10 outcomes then you have a potential 100 different correlations. If you assume that any correlation where p<0.05 is significant then you are going to get 5 "significant" correlations just by chance. The summary talks about how they didn't find correlations with acetaminophen before and after pregnancy, so that's 3 distinct factors just there.