WOMEN who reported not taking prenatal vitamins immediately before and during pregnancy were twice as likely to have a child with autism, a study has found.

If the women also had a mutation in a high-risk gene, they were seven times more likely to have a child with the developmental disorder, researchers from the University of California, Davis, reported in the online edition of the journal Epidemiology. The researchers studied about 700 families with two- to five-year-old children with autism. In telephone interviews, the mothers were asked if and when they took prenatal vitamins and the type.

The team found that mothers who took vitamins prenatally or during the first month of pregnancy were only half as likely to have a child with autism as those who did not (when measured against a control group).

Advertisement: Story continues below For mothers who began taking vitamins in the second month of pregnancy there was no effect.

This suggests that by the time most women learn they are pregnant, beginning to take the vitamins will provide little or no benefit in terms of autism.

The researchers also found strong associations with two gene mutations that had previously been linked to autism risk.

''The good news is that if this finding is replicated it will provide an inexpensive, relatively simple evidence-based action that women can take to reduce risks for their child, which is to take prenatal vitamins as early as possible … and even when planning for a pregnancy,'' said the epidemiologist Irva Hertz-Picciotto, the study's senior author. It was important to replicate it, she added, because the study relied on women's recall of vitamin use, which was not always accurate.

Los Angeles Times

The researchers studied about 700 families with two- to five-year-old children with autism....

The team found that mothers who took vitamins prenatally or during the first month of pregnancy were only half as likely to have a child with autism as those who did not (when measured against a control group).

Anyone else notice that there are absolutely no hard numbers in that report? Seems suspect to me.