The Telegraph this morning ran with the headline, "Pregnant women warned of 80pc increase in miscarriage from food heated in plastic". In fact, the work is "not nearly enough to prove a link", according to Dr Linda Giudice, president of the American Society for Reproductive Medicine.
The sample was tiny – 115 women – and preselected for a history of miscarriage and difficulty conceiving. Within that, there was a demonstrable effect that high bisphenol-A levels (BPA is the substance found in food packaging, in greater quantities once it's been heated) corresponded to probability of miscarriage. Sixty-eight of these women had miscarriages, 47 had live births. That is an incredibly high level of miscarriage, as you'd expect, from the fact that these women were being studied precisely because they had a history of miscarriage, and the largest single factor in miscarriage is having already had one. In other words, even if this study were of a useful size – say 100 times larger – there is no way you could apply its findings to the general population. The very most that it does is to give, as Guidice terms it, "biological plausibility" to BPA as a miscarriage factor, and give warrant to further research.
If it were as dangerous as the headline makes it out to be, it would not be unreasonable to ban its use altogether, with immediate effect – it's biologically implausible to find a toxin dangerous enough to choke dead a foetus, and yet perfectly safe for the rest of the human and animal kingdom. It is noticeable that the newspapers preaching a righteous path of impractical caution for pregnant people (and the study author is very clear on this – BPA is impossible to avoid altogether) never make any suggestion that would even ruffle the feathers of an industry. If there are toxins out there, the onus is placed entirely upon the individual to avoid them, and not at all upon business to stop producing them. It is a pretty rum but nevertheless widespread interpretation of consumer choice, that the freedom to choose whether or not to ingest a toxin is way better than the oppression of a regulatory system that protected you from toxins and took that choice away.
There is, in the field of pregnancy, massive overinterpretation of studies that are either too small or too partial to tell us what the headlines want to tell us; there is a marked tendency to expect of pregnant women risk-avoiding behaviour that is either unfounded or impossible or often both; and there is a "play it safe" mantra that is quasi-religious in its lack of evidence or even the ability to understand evidence. If that plastic bottle could harm your foetus, wouldn't you rather play it safe? What would you prefer, a new sofa or peace of mind? Since no amount of alcohol has been proved to be safe, why do you want to take the risk in drinking it?
Emily Oster wrote convincingly about the anti-intellectualism of this in her recent book, Expecting Better ("too much of many foods can be bad. If you have too many bananas, the excess potassium can be a real problem. But no doctor is going around saying 'No amount of bananas has been proven safe!'.") However, practical arguments against this kind of caution never quell it, because at its heart it is ideological; the goal of this hysterical, toxins-under-the-bed risk assessment is to erase bad luck as a peril of reproduction. A miscarriage, a stillbirth, a birth defect – all of these things, by the modern terms of engagement, are avoidable by righteous and informed behaviour during pregnancy. Now this simply isn't true, but if we can persuade ourselves that it is, all that natural empathy you would feel towards a woman in a dire situation can be safely tucked away under "She brought it on herself. With all the bottled water she left on her dashboard, and then drank." Beware the amplification of the minuscule study; its intended implications are greater than it would ever admit.
