There was plenty to like about this story. There are quotes from three experts who emphasize different perspectives on the results, and the take-home message about supplements — women should be taking them to prevent birth defects, regardless of any link to autism — is spot on. But the story gets tripped up when it uses active verbs (“may reduce” and “may prevent”) to describe the importance of the findings, and its use of relative risk figures was inadequate to convey the benefits. Presenting only the statistic “39% less likely” is particularly problematic. Even the abstract of the paper on which the story was based provided the absolute differences: rates of autistic disorder of 0.10% for the group taking folate, and 0.21% among nonusers. This is a very small absolute risk difference – and from an observational study at that.
As a result, readers will get a somewhat-too-rosy picture of what this study found and what it means.
It would be great if a simple intervention like taking a folic acid supplement — something women of childbearing age should be doing anyway — could help prevent autism. And these results do support the idea that folic acid might be protective. But there are also good reasons to question whether it’s the folic acid, or something else, that could be responsible for the findings. And it’s not clear that the benefits reported apply to women in the United States, who are already getting some extra folic acid from food fortification.
The cost of folic acid supplements is not in question.
The story says women who took folic acid “were 39% less likely to have children with autism.” This is inadequate, because it doesn’t reflect just how rare a diagnosis of autism was in this study. (Only 114 children out of about 85,000 received that diagnosis.) The findings in absolute terms would have been more informative, as they show that autism was diagnosed in 0.10% of children whose mothers took folic acid compared with 0.21% of children whose mothers didn’t.
There are hypothetical, but plausible concerns about harmful effects of getting too much folic acid, especially an increase in cancer risk. This is why many countries have not embraced mandatory folic acid fortification of the food supply. The story should have mentioned this.
This was a tough call. On the plus side, the story quotes a researcher who says “the data really do not establish anything close to a causal connection.” That’s terrific and absolutely true, since this was an observational study. But why then have a headline and lead sentence that say folic acid “may help reduce” and “may help prevent” autism, since these phrases clearly imply a causal relationship between folic acid and autism prevention? It’s also incorrect to suggest that supplements are a “tangible [thing] a woman can do to reduce her risk of giving birth to a child with the disorder.” That hasn’t been established.
We really like the way the story closed out with a great quote: “Caring for individuals with autism and their families would be a whole lot easier if we had simple answers about cause and risk. The reality is, autism is a complex disorder and our best answers about causes and treatment are going to be complex as well.”
But there are a couple of other important points that really should have been made, but weren’t:
Ultimately, we’ll rule this unsatisfactory, but with credit due as noted.
There was no overt disease mongering, but as discussed above, we wish that the story had been clearer about just rare how a diagnosis of autism was in this study.
The story includes a nice mix of voices, including experts from an autism advocacy group, the March of Dimes, and in independent autism research program. This was a strong point of the story.
There aren’t any proven strategies to reduce autism risk, although research suggests a link to maternal smoking, the father’s age at conception (older than 40 = higher risk), and exposure to air pollution and pesticides. The story could have mentioned these, but we won’t ding it for not doing so. The best we can rule it is Not Applicable.
The availability of folic acid supplements is not in question.
The story explains that folic acid is already recommended for the prevention of neural tube defects. But it could have noted that at least one other study has documented an association between folic acid use and reduced autism risk. The JAMA paper on which the story is based cites several prior research efforts showing association with folate in early pregnancy and severe language delay or autism.
There is enough original reporting that we can be sure this wasn’t based on a press release.