No, your premature baby isn’t going to be Autistic

An article has been floating around, announcing that babies born premature are at a greater risk of developing autism later in life. This stems from a new study performed by Karolinska University in Stockholm, Sweden.

Medical professionals aboard the Military Sealift Command hospital ship USNS Comfort (T-AH 20) care for a baby born 7 weeks pre-mature.

As happens with many studies that are reported by the media, the results of the study are being exaggerated. The problem is that this isn’t what the study says. And media outlets are using sensational headlines such as “Pre-mature babies at higher risk of developing autism” that dismiss the study’s actual findings. Suffice it to say, there’s simply not enough evidence to assume that your premature baby is likely to become autistic.

What the Media Reports

I’m taking quotes directly from the press release issued by Karolinska Institute:

Extremely premature babies run a much higher risk of developing autism in later childhood, and even during the neonate period differences are seen in the brains of those who do.

…However, babies born more than 13 weeks prematurely run a serious risk of brain damage, autism, ADHD and learning difficulties…

…In this present study, the researchers examined over 100 babies who had been born extremely prematurely (i.e. before week 27, the beginning of the third trimester)…

…almost 30 per cent – of the extremely preterm-born children had developed ASD symptoms…

…Amongst children born after full term pregnancy, the corresponding figure is 1 per cent…

…The researchers found that it was more common in the group of children who had developed ASD for there to have been complications during the neonate period, such as surgery, than it was amongst their prematurely born peers who had not developed ASD…

…’Our study shows that environmental factors can also cause autism,’ says Dr Ådén. ‘The brain grows best in the womb, and if the developmental environment changes too early to a life in the atmosphere, it can disrupt the organisation of cerebral networks. With new therapeutic regimes to stimulate the development of such babies and avoid stress, maybe we can reduce the risk of their developing ASD.’

The problem here is that this press release misrepresents both the findings of the study as well as their statistical strength. In order to see where the press release is misleading, we first need to look at what the actual study found.

The Study

Researchers studied 84 pre-term babies over the course of six years. The babies were born earlier than 27 weeks. Twenty-three of these children later developed Autism Spectrum Disorder (ASD) at age 6. To check for brain lesions, “Structural brain morphometric studies were performed in 33 infants with high-quality MRI and no evidence of focal brain lesions.”¹⁾Padilla, N., Eklöf, E., Mårtensson, G. E., Bölte, S., Lagercrantz, H., & Ådén, U. (2015). Poor Brain Growth in Extremely Preterm Neonates Long Before the Onset of Autism Spectrum Disorder Symptoms. Cerebral Cortex, bhv300. http://doi.org/10.1093/cercor/bhv300

First, lets de-compact a few things. Yes, the risk in this sample size of developing ASD was about 30%, compared with the global population which is consistently at about 1-2% ²⁾Facts About ASDs. (n.d.). Retrieved December 29, 2015, from http://www.cdc.gov/ncbddd/autism/facts.html. That’s a compelling jump in risk, and the study does show that, in this sample, premature babies do likely have an increased risk of developing ASD.

What does it all mean?

However, there are a few issues with this conclusion which the media and press release ignores. Primarily, the sample size is only 83. With such low of a sample size, the study simply doesn’t have enough statistical power make any real inferences toward all pre-term babies born each year.

In addition, the press release glosses over the fact that this study focused on babies born before 27 weeks, whereas a baby born before 37 weeks can be qualified as pre-term. As an aside roughly 15 million babies in the world are born pre-term each year, most of them are in Africa, India and China. I couldn’t find exact numbers for how many babies are born before 27 weeks, but we can make some rough estimations. About 1% (rounding up) of babies born in the US are born before 28 weeks. Sweden and the US are fairly similar in regards to health, so we can use this number without too much error. Rounding up again about 12,000 babies are born in Sweden every year. One per cent of that number comes out to 120 babies each year born before 27 weeks. That’s fairly representative actually.

But only 1% of pre-term babies are born before 27 weeks, which is what the study looked at. And babies born before 27 weeks are already known to have a list of health deficits ranging from severe physical disabilities such as problems with hearing, vision, breathing, heart disease and intestinal problems. And that glosses over the mental deficiencies that are already known to occur in babies born that early.

The real issue with this study, and it’s an important study and I couldn’t find anything wrong with the actual findings from a statistical sense, is that the sample they pulled from were already at a higher risk of developing difficulties.

There are so few babies born before 27 weeks that, should your baby be born in week 36 or 37 as most pre-mature babies are, you don’t need to worry about your child’s risk of developing autism. The outcomes for most pre-mature babies, especially in the US are quite good. Yes, pre-mature babies do have a risk of health complications, but the survival rate is 90-95% and health outcomes in general are fairly good for most pre-term babies.³⁾Behrman, R. E., Butler, A. S., & Outcomes, I. of M. (US) C. on U. P. B. and A. H. (2007). Mortality and Acute Complications in Preterm Infants.

So should you be worried if your baby is born pre-term? Yes, there are some significant health complications. Should Autism be one of those worries? Not really, not unless your baby is born significantly early, which less than 1% of infants are.

What would be interesting to note is the rate of developing ASD compared to gestational age at birth. We know (from this study) that babies born before 27 weeks run a higher risk, but what about all pre-mature babies? I couldn’t find any numbers for this, which would seem to indicate that the risk factor for all pre-mature babies is rather low compared to the general population, but again there isn’t enough data to make a conclusion.

The Take Home

Suffice it to say, among the risk factors for developing ASD, the incidence rate of pre-mature babies is so low as to not be a serious thing to worry about unless your baby happens to fall under that very small class.

But that’s not what the news articles are saying. Many news articles are insinuating that, either by misleading headlines or misleading text, that each of the 4 million babies born pre-mature each year has a 30% higher risk of developing autism later in life. This ignores two things: first is the fact that the study only looked at a small portion of the population which is in no way representative of the overall population of newborns, and that Autism Spectrum Disorder is not a one sized fits all diagnosis. Some Autistics are non-verbal and highly aggressive, but this is a very small proportion. Most Autistics function reasonably well in society and many show few if any outward signs of autism.

Quite frankly speaking there are worse health outcomes for pre-term babies than developing Autism.