Although the findings of this preliminary study are important and warrant additional exploration, we can’t yet say with any confidence whether prenatal vitamins are protective against autism, as this story prematurely suggests.
Autism, as the story notes, affects about 1 in 110 children, and its prevalence seems to be rising. It would be wonderful if simply taking a vitamin pill could help prevent autism and the huge burden it places on children and families. However, there have been too many cases in the past where vitamin supplements seemed to be beneficial in early research but proved to have no benefit or worse when definitive studies were conducted. We need to temper our enthusiasm for prenatal vitamins pending replication of these initial results in more rigorous studies. Meanwhile, women of childbearing age should continue taking prenatal vitamins for their other well-documented benefits, including the prevention of spina bifida and other birth defects.
There was no discussion of costs, which is unfortunate considering that there can be big differences in the price of prescription prenatal vitamins vs. OTC brands. Some women may also do fine with a basic store-brand multivitamin. Costs can also vary depending on insurance coverage, since some plans may cover the cost of prescription vitamins but not OTC products. At a minimum, the story should have mentioned the range of costs for a month’s supply in order to satisfy the criterion. Some discussion of the other tradeoffs between prescription and OTC brands (varying formulations, tolerability, etc) would also have been helpful.
The story communicates the results using relative terms, noting that women who didn’t take vitamins had “about a 60% higher risk of having a child with autism.” In women with a high-risk genetic profile who didn’t take vitamins, the risks were 4.5 to 7 times higher compared with women who didn’t have the high risk genes and took prenatal supplements, according to the story. While these relative numbers probably give readers an overinflated sense of the benefits of taking supplements (or the harm of not taking them), using absolute numbers, which we normally prefer, would be inappropriate and probably even more misleading in coverage of a case-control study such as this one — for the reasons discussed here. The use of relative numbers is acceptable in a case-control study if there is some indication of the underlying population-wide risk, which this story provides when it says that autism affects “up to one in 110 children” — or about 1%. While admittedly complicated and imperfect, this mix of numbers should help readers draw appropriate conclusions about the significance of the findings.
There should have been some discussion of the possible downsides of getting too high a dose of certain vitamins, and that more is not necessarily better when it comes to supplements. Especially now that many foods are fortified with folic acid, there is some concern about women receiving folic acid doses that could overwhelm the body’s systems for metabolizing this vitamin, which could lead to unknown adverse effects on the fetus. Some women also report gastrointestinal problems, including nausea and constipation, when taking prenatal vitamins.
The story uses inappropriate language to discuss the results of this observational study. The headline and lead play up the possibility that prenatal vitamins “may reduce” autism risk, suggesting a cause-and-effect relationship that this study wasn’t designed to assess. In addition, there was no discussion of the many limitations that we need to consider when evaluating the importance of these results. For example, women who are health conscious-enough to take vitamins may also take other health-promoting steps that might reduce the risk of having a child with autism. This might confound the data. Even this press release issued for the study acknowledged problems that it says “could have biased the results.” A researcher quoted in the release explains: “Because the mothers were asked about their vitamin use years after their pregnancies and after their child’s developmental status was known, some error is expected in their reporting. Moreover, in comparison with mothers who have an affected child, mothers whose children are healthy and show typical developmental milestones may be less likely to remember accurately, simply because they have less reason to reflect on and be concerned about their behaviors years earlier.”
There was no disease-mongering in this story. Autism incidence was clearly stated and there were no dramatic anecdotes.
The story quotes an independent expert who was not involved with the research and who is identified as working for an advocacy group. There don’t seem to be any other potential conflicts to disclose.
The story should have mentioned a healthful diet as a means to ensure adequate intake of folate and other nutrients that are important for fetal development.
Readers will know that the vitamins discussed in the story are widely available either by prescription or as over the counter (OTC) supplements, so we’ll award a satisfactory here. However, the story could have been more precise about what exactly these women were taking. Were they using formulations specifically designed for women of childbearing age or simply taking a standard multivitamin supplement during the prenatal period? Did the benefits extend to women who only took folic acid, which is widely regarded as the most important vitamin to supplement during the prenatal period? In addition, the story might have noted that there are different formulations of prenatal vitamins with varying doses of key nutrients. It would have been helpful to include some discussion of the nutrients women should look for and why.
The association between prenatal vitamins and autism does appear to be a novel and newsworthy finding, as the story suggests.
This story quotes an independent researcher and doesn’t seem to rely excessively on any news release.