Ethics and Epidemiology
However much the authors of this paper would wish that it be considered as nothing more than a 'research pointer' and thus neutral in its effect on the general public I wish to suggest that both they and the editor were being disingenuous in this belief. Before the post arrived on 10 December bringing my copy of the BMJ I went to get my newspaper. The first headline I saw was on the front page of a popular tabloid - "Folic Acid Cancer Risk". The broadsheets also ran the story and, although there was more balance to their coverage, the residual message would inevitably cause concern to women taking folate to reduce the risk of fetal neural tube defect (NTD)in a desired pregnancy. In light of previous 'scares' of this kind, it is highly likely that a number of susceptible women will not take folate and some of these may conceive fetuses with NTDs. The greater risk in terms of numbers of women who may be affected will arise from an almost inevitable further delay in the fortification of wheat and corn flour around the world. What company or government will take responsibility for such fortification now that this question has been raised despite the much stronger contrary evidence quoted in the paper and the commentary that followed?
The authors will, of course, claim that they cannot be held responsible for such an outcome but this is, in my opinion, naive and ethically unsound. The main argument hinges on the interpretation of the data. The authors themselves point out that the numbers are small and the confidence intervals large. It was also associated with a much larger dose of folate than is routinely used to reduce the risk of NTD. They further confess that "this may be a chance finding". This is, in my opinion, not nearly a sufficient 'statistical health warning' since, as the commentary by Oakley and Mandel states, the overall evidence is, "the most likely explanation for the reported association is chance".
Thus, on one hand, we have an intervention - small dose folate supplementation- that confers a proven and considerable benefit. On the other we have a finding that not only goes against prior research findings but is also most likely to be due to chance. There can surely be no argument about the balance of benefit in this case!
Those who write such papers and those who publish them cannot disclaim ethical responsibility for how the data are interpreted and must consider more carefully their ethical responsibilities in such situations.
Competing interests: None declared
Competing interests: No competing interests