June 25, 2015

expert reaction to study on association between Caesarean section births and diagnosis of autism spectrum disorder

A study has examined a possible link between Caesarean section births and probability of diagnosis of autism spectrum disorder (ASD).

Prof. Andrew Whitelaw, Emeritus Professor of Neonatal Medicine, University of Bristol, said:

“The paper by Curran and colleagues is an important and robust study. It was important to investigate whether the previously observed association between Caesarean section and autistic spectrum disorder (ASD) was likely to be due directly to the Caesarean or whether an earlier foetal characteristic increased the risk of both Caesarean and later ASD without the Caesarean making any causal contribution.  Caesarean section, particularly elective Caesarean, may deprive the foetus of important physiological and metabolic adaptive processes that are part of normal birth with long lasting changes.

“The fact that this is the largest study by a long way (nearly 2.7 million children born in Sweden, a country with all its residents on a database with a personal number that follows them through all official interactions), gives robustness. Furthermore, databases in Sweden can be linked over many years, across generations, and between siblings while still protecting personal confidentiality. Furthermore the observed association between Caesarean section and ASD could be adjusted to according to whether a sibling had ASD or not.  For many years there has been strong evidence that there is a genetic contribution to ASD. When adjustment was made for sibling ASD, there was no association between Caesarean section and ASD.  This is good evidence that Caesarean section does not cause ASD.

“The authors have gone a long way in considering and eliminating possible bias in their study but they acknowledge limitations. Nevertheless, their main conclusion looks to me to be justified.

“This is a good example of the advantages of having every resident in a country logged by number but not name, on a database which follows every aspect of life including health, education, employment, welfare and tax.  Each Swedish mother was not asked to consent specifically to this research study. The whole population of Sweden knows how statistics are collected and accepts that there are huge advantages in a welfare state knowing about its citizens. Britons should lose their fear of such databases. ”

‘Association between obstetric mode of delivery and autism spectrum disorder: A population-based sibling design study’ by Curran et al. published in JAMA Psychiatry on Wednesday 24th June.

Declared interests

None declared