The combined results of 61 studies conducted in 19 countries suggest that when a woman gives birth by cesarean section, autism in the child is 33 per cent more likely and the odds of attention-deficit/hyperactivity disorder (ADHD) increase by 17 per cent.

The statistical analysis, published in JAMA Network Open, does not prove that having a C-section directly causes such problems. Other factors that lead doctors to order a surgical delivery of the baby may underlie the link.

The researchers collected studies covering more than 20 million deliveries dating back to 1999, to create a large pool of data to assess risks associated with cesarean delivery that may not be evident in the smaller individual studies.

When the team, led by Tianyang Zhang of the Karolinska Institute in Stockholm, assessed autism spectrum disorder, they looked at 29 study populations, 17 of which showed an increased risk that was large enough to be statistically meaningful.

Among the babies of women who had vaginal deliveries, for example, the autism rate was 0.8 per cent, according to data provided in the paper. A 33 per cent increase would put the rate in the cesarean group at 1.0 per cent.

For ADHD, Zhang's group assessed 14 study populations, only 5 of which showed a statistically significant risk with C-section, but those studies tended to be the largest so they carried more weight.

Babies born through vaginal delivery had an ADHD rate of 2.4 per cent. A 17 per cent increase in the C-section group would raise the rate by 0.4 percentage points, to 2.8 per cent.

"The impact of this is not going to be great because the increase is still pretty low," said Dr. Pankhuree Vandana, a pediatric psychiatrist at Nationwide Children's Hospital in Columbus, Ohio, who was not involved in the research.

"It's not very dramatic. It's statistically significant but from the population standpoint, the risk is not huge," she said in a phone interview. "And it absolutely does not prove that cesarean section is causing autism or ADHD. The risk factors that lead to a C-section may also be risk factors that lead to autism or ADHD."

Zhang, a PhD student at Karolinska, also emphasised in an email that "our study does not provide irrefutable proof that C-sections CAUSE neuropsychiatric disorders."

For example, "C-section is often recommended by obstetricians if the mother has diseases such as diabetes, hypertension, etc. Some of the diseases might already have an effect on a child's brain development when he or she is still in the womb," the author said. A combination of factors may contribute to the risk, so "it would therefore be wrong to demonize C-sections. C-sections should continue to be used when medically necessary."

However, "the results appear to further add to the known adverse health outcomes associated with cesarean delivery and suggest judicious use of cesarean delivery," the researchers write in their report.

There is already evidence that children born through C-section have a higher risk of obesity, allergies, childhood diabetes and a form of blood and bone marrow cancer known as acute lymphoblastic leukemia.

On the other hand, surgical delivery can be lifesaving.

Still, it is widely believed that the surgery is overused in many countries. The optimal rate is considered to be 10 per cent to 15 per cent but worldwide cesarean delivery rates, which were about 6 per cent in 1990, have risen to 21 per cent, according to 2015 data.

The study also found that whether a cesarean delivery was scheduled or done on an emergency basis had no effect on the autism or ADHD rates.

The Zhang team found hints of slightly higher rates of intellectual disability, obsessive-compulsive disorder, tic disorders and eating disorders among children born through C-section, but those associations were not statistically significant, meaning they could be due to chance.

The team did not find any link between cesarean delivery and other conditions such as depression.