Previous: Dec. 16, 2015 - HCT LIVE
Next: Questions and Answers With Aaron Carroll and John Green



View count:23,016
Last sync:2024-05-03 22:15
I'm getting lots of tweets, emails, tests, etc. "Antidepressant Use During Pregnancy and the Risk of Autism Spectrum Disorder in Children". The is Healthcare Triage News.

Those of you who want to read more can go here:

John Green -- Executive Producer
Stan Muller -- Director, Producer
Aaron Carroll -- Writer
Mark Olsen -- Graphics

And the housekeeping:

1) You can support Healthcare Triage on Patreon: Every little bit helps make the show better!
2) Check out our Facebook page:
3) We still have merchandise available at
I'm getting lots of Tweets, e-mails, texts, etc talking about the panic de jour, antidepressant use during pregnancy and the risk of autism spectrum disorder in children.  This is Healthcare Triage News.


Let's go straight to the research!  In this study, the authors found that after adjusting for confounders, including maternal psychiatric conditions, that the use of antidepressants during the second and/or third trimester was associated with an increased risk of a child later being diagnosed with autism.  The hazard ratio was 1.87.  Cue the scary headlines.

Now, let's break this down.  In the entire cohort of 145,456 kids, 0.7% of them had an autism spectrum disorder.  That's actually on the low side.  Of course, some of the kids in the cohort were still too young to be reliably diagnosed.  Moreover, they excluded any prematurely born infants from the analysis, and preemies are at an increased risk of autism.

Let's talk numbers first.  Of the 145,000+ kids born into this cohort, only 4,724 were exposed to antidepressants.  The other 140,732 were not.  Of the 4,724 exposed to anti-depressants, 46 developed an ASD and 4,678 did not.  Of course, 1,008 kids who were not exposed to antidepressant were diagnosed with an autism spectrum disorder, too.  Antidepressants in the first trimester were not found to be significantly related to the risk of an ASD, though.  Of those 46 kids, only 31 were exposed to an antidepressant in the second or third trimester.  

In other words, the headline finding is a comparison between the kids who were never exposed to antidepressants and got an ASD, 1,008 of the 145,456 versus the kids who were exposed to antidepressants in the second or third trimester and got an ASD, 31 out of 2,532.  Those calculate to 0.7% and 1.2%.  The adjusted relative risk increase was 87% when taking into account person years.  The absolute increase, which is what you should care about, was 0.5%.  

If you accept this at face value, the number needed to harm was there for about 200.  The vast majority of kids who were exposed to antidepressants in the second and/or third trimester did not develop an ASD.  Further, only SSRIs were shown to be related when looking at classes of drugs individually.  They do have different mechanisms, after all.  There were 22 kids exposed to SSRIs in the second and/or third trimester who developed an ASD, out of the 145,456 total kids in the cohort.

The point of laying out all these numbers was to show you that this isn't a problem of epidemic proportions.  At the end of the day, the study hinges on a small number of children.  That doesn't mean that the findings aren't robust or that they aren't real, but you have to place things in context.  The mothers are theoretically getting a benefit from the drug or they wouldn't be on them.  Those benefits need to be weighed against risks and anyone who says that any risk is too big when it comes to a child should never be putting their kid in a car.

We also need to consider this finding in context.  There have been any number of studies in this area.  One study showed an association between maternal use of SSRIs and a lower risk for preterm birth and many birth outcomes.  Depression in moms in general has been shown to be associated with low birth weight, preterm birth, and being small for gestational age.  Another case control study from earlier this year found antidepressant use wasn't linked to ASD but that it might be linked to ADHD.  

I didn't tout any of those studies as truth when they were published any more than I would this one.  It's a data point and we should adjust our priors accordingly.  Finally, there are limitations to this work as there are with all research.  They did adjust for the presence of depression in mothers, but not necessarily the severity.  It's possible that the most severe cases were treated with antidepressants, and the severity of the depression is what's related to ASD in kids, not the drugs.  It's possible that mothers prescribed antidepressants are more plugged into the healthcare system and that their kids were more likely to get a diagnosis of ASD than those whose mothers aren't as plugged in.  It's possible that just filling a prescription is not a great marker for actual antidepressant use in pregnancy.  It's possible that the preemies might have made a difference if they'd been included.  It's even possible that something else is at play that we're all missing.

Here's the tl;dr or dw, I guess.  This was a statistically significant finding, but from a relatively small group of children, regardless of the size of the entire cohort.  It's only significant for SSRIs and in the second or third trimester, which is 22 kids total, and the absolute risk increase was only 0.5%.  There are limitations to this study and other studies have found different results.  My take home would be that this deserves more work and attention and that any potential harms from the antidepressants should be weighed against the known benefits for these pregnant moms.  They should discuss that with their physicians.  

Healthcare Triage is supported in part by viewers like you through, a service that allows you to support the show through a monthly donation.  Your support helps us make this bigger and better.  We'd especially like to thank our research associate, Cameron Alexander, and our first ever surgeon admiral, Sam.  Thanks Cameron.  Thanks, Sam.  More information can be found at