Article Text

Download PDFPDF
O-26 Folic acid during pregnancy and the risk of autism: a nested case-control study
  1. Koren1,
  2. Fono1,
  3. Sharman Moser2,
  4. Rotem2,
  5. Davidovich2,
  6. Chodik1,
  7. Shalev1
  1. 1Tel Aviv University, TEL AVIV, Israel
  2. 2Maccabi Research Institute, TEL AVIV, Israel


Background During the past two decades there has been a dramatic increase in the prevalence of autistic spectrum disorders (ASD) among children worldwide, concurring with growing use of periconceptional folic acid supplements for the prevention of neural tube de-fects (NTD). This has raised the question of possible as-sociation between maternal folic acid exposure and ASD. We aimed to examine the association between the cu-mulative dose of folic acid purchased by the mother from 3 months before and throughout pregnancy, and the risk of autism.

Methods In a nested case-control study, we identi-fied 1650 children with ASD diagnosed from a cohort of 5 04 028 children born in a large health organisation in Israel from 2000 through 2013. ASD patients were in-dividually matched in a ratio of 1:5 to ASD-free children (n=7591) from the cohort on age and maternal age, sex, residential area and level of socio-economic status. Odds ratios and 95% confidence intervals by mean daily dose of supplemented folic acid during the 12 month period were calculated using unconditional multivariable logis-tic regression. The model was adjusted for potential con-founders including age of mother, place of the child in the family, having a fertility problem and being enrolled in our fertility register, suffering from epilepsy, maternal BMI, and serum concentrations of vitamin B12.

Results In univariate analysis, mean daily dose of folic acid purchases among ASD cases (177.84 µg, SD=250.7) during the 12 month study period was significantly high-er when compared to controls (145.87 µg, SD=214.2) (p<0.001). However, significantly more ASD children were first born, and mothers purchased significantly more folic acid during the first pregnancy than in the second preg-nancy, and even less in the third pregnancy. Similarly, the ages of ASD mothers were significantly older, they exhibited significantly more subfertility, visited significant-ly more often at their physicians’ offices. In multivariable analysis, accounting for these confounders, there were no apparent differences in the amount of folic acid pur-chased between the groups and no dose –response ef-fect of folic acid on occurrence of autism was discerned. In a sensitivity analysis we compared folic acid purchases between healthy and ASD first born children while ac-counting for all other variables; here too there was no association between higher folic acid purchases and ASD occurrence.

Conclusion No association was found between the amount of folic acid purchased and the occurrence of ASD. The univariate finding of higher folic acid exposure in autistic children is most probably the result of colinear-ity between the order of birth (first born) and the trend of mothers to consume significantly more folic acid in their first pregnancy.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.