A recent study identifying an association between autism and induction of labor that was published in the August issue of JAMA Pediatrics has gained a lot of media and patient attention.  Almost every major news outlet worldwide has featured an article this week citing the findings of a study conducted locally at the Duke Center for Human Genetics in Durham—this data has hit home, literally and figuratively, for many of our patients and has already prompted many thoughtful patient questions regarding autism and labor induction.  Some of these patients have even turned to our OB nurse line as a resource-- I am impressed that our patients are so aware. In fact, we received one call before I had even read the news of this story in the several OBGYN related “alert” emails I get daily.

For those unfamiliar with the study,  the author and his colleagues reviewed data of over 600,000 births in the state of North Carolina and then linked the birth records with educational records, specifically  any diagnosis of autism.  Their data showed a higher autism diagnosis rate in children who were born via  labors that were induced or augmented.  Overall, review of this population data showed that 1.3% of boys and 0.4% of girls born in  North Carolina were ultimately diagnosed with autism, these percentages  are consistent with many other studies on autism rates.

Importantly, despite what headlines report, the authors do not conclude that induction of labor, or augmentation of labor, are a cause of autism.  They specifically say this.  They describe a “link” between the circumstances around these deliveries and autism, but in no way can show causation.

“It remains unclear whether (1) the act of labor induction/augmentation itself or the medications used underlie the association; (2) the medical and obstetric conditions that are associated with labor induction/augmentation drive the association; or (3) acute intrapartum events that are more common among women with labor induction/augmentation drive the association with autism.”

The authors further state that the results of their study should NOT alter the standard of care regarding induction or augmentation.  Excellent medical evidence shows that induction and/or augmentation of labor, if used appropriately, can reduce stillbirth, neonatal respiratory difficulty, neonatal infection, meconium aspiration syndrome, c-section rates, and maternal mortality.  However, it also highlights the idea that all medical interventions should be given thoughtful consideration, including induction of labor, especially elective induction of labor.  To draw on another recent, timely and news-worthy event, if royalty can wait until a week after her due date, so can you!  (ha, ha,…)

There are a couple of other interesting pieces of data from this study.  The authors did not only look at the association between labor induction/augmentation and subsequent diagnosis of autism, but several other variables.  I found two very interesting.  First of all, if the mother was college-educated, her child had a higher probability of an autism diagnosis than if she did not go to college.  In fact, the increase in odds of autism was higher with maternal college education than the increase associated with labor induction/augmentation.  Second, mothers who smoked during pregnancy had children with lower chances of an autism diagnosis.  Applying the same rationale, I think stating that a mother should avoid college and start smoking during pregnancy to decrease the risk of autism would be similar to saying that we should eliminate induction or augmentation of labor when indicated.

However, I do not think this study in any way should be ignored .  Like many medical studies, it should encourage further investigation.  It would be amazing if someone could find some sort of preventable cause of autism.  In the meantime, I think we should continue to care for our patients by following the current guidelines for induction and augmentation of labor.



UPDATE: 9/13/16
A new study looking to build upon the one mentioned above used the Swedish database and had a superior control group. The link found with the NC study was NOT seen with this more thorough analysis and controlled review.