Read on Twitter
New commentary out in @Ped_Research: Prenatal antidepressant exposure and neurodevelopmental problems in children: to get the right answer, we must ask the right question. A few thoughts below 
https://rdcu.be/cdzsT
https://rdcu.be/cdzsT
There is a lot of concern about antidepressant use in pregnancy and its possible effects on fetal brain development. Multiple studies showing small but persistent associations with behavioral, cognitive etc outcomes in kids (2/n)
Also plenty of null findings! But ongoing discussion is mainly focused on confounding by indication, i.e., the reason for taking antidepressant. (3/n)
But, not much concern over interventions implied by study design: all antidepressants grouped as a class, continuation versus initiation versus late discontinuation. I don't know of any study, for ex, looking at fx of switching to an SSRI from a TCA versus discontinuation (4/n)
Or for that matter, antidepressant taken for depression vs anxiety. (5/n)
Because there's so much conflicting literature, there are (at least) 2 ongoing RCTs of antidepressants in pregnancy for which neurodevelopment in kids is a primary or secondary endpoint (6/n)
One studies initiation of a specific SSRI during pregnancy for women with moderate depression; the other randomizes to tapering vs continuation. This is a gift for observational studies! (7/n)
These trials should be informing our study design. At this point, I don't believe we should be comparing any antidepressant for any reason to a general-population control. (8/n)
So, that's that!
