Of fish and fetuses

(If you're not up to speed on the latest flap about fish and mercury, please see the articles here and here. Then return to the next paragraph. Happy reading! A special award will be given to the alert reader who figures out what the FDA is actually going to announce in the spring -- i.e. whether they will be making their advisory more or less stringent. There seem to be contradictory reports on the matter.)

I am an epidemiologist, but I'm no expert on the effects of methylmercury expsure. But in looking through the National Academy of Sciences sober and thoughtful 2000 report on the matter, it appears that we have here a classic problem of epidemiology and public health: how do we translate risk assessment into legislation?

There are several parts to this controversy. One involves the "action standard" of the FDA -- this is the methylmercury dose that would require the agency to take legal action -- and whether this is to be based on the EPA's advisory. In April, the FDA indicated that this would indeed be the case, making its own action standards some five times stricter. But two other relevant pieces of the problem should not be overlooked: the wording of the FDA's warning itself, and, perhaps most important, how we should implement public policy based on these advisories.

Of course, these different parts of the problem are interdependent. The "action standard" of the FDA, like the advisory level of the EPA, demands legal action on the part of the agency. So one cannot agitate for the highest possible level of protection without taking into account the results that such protection will have.

Let us assume, for the moment, that the EPA advisory level is scientifically justified, albeit the matter is actually quite complicated. The NAS report fairly characterizes as "strong" the evidence for neurological effects due to methylmercury. These effects are strongest, states the report, also reasonably enough, in fetuses and newborns, whose neurological development is still proceeding. However, this evidence is based almost entirely on either animal studies or studies of populations affected by acute mercury exposure. The several epidemiologic studies concerned with chronic exposure to low levels of mercury (more relevant to us tuna-fish folks, for example) give a mixed picture, and in fact the largest of these studies (the so-called Seychelles study) finds very few significant neurological effects. The NAS report further states (in its conclusion) that the average exposure in the United States is quite low in comparison to the recommended standards.

In short, the EPA advisory is based on scientific conclusions which find their strongest base not among common, everyday levels of exposure, but in above-average, acute exposure, and in animal studies. If the FDA adopts the EPA level, we need to understand that we are setting the bar of protection higher, erring on the side of biological effects that, while plausible, might not be commonly found in daily life. If, for example, the new data on mercury levels in tuna suggest that exposure in fetuses, newborns, and perinatal women might be higher than previously thought, one must still compare this increased exposure with the recommended standard -- which is in turn set as low as possible, in keeping with a standard of maximum protection.

In other words, we are dealing with a strong, plausible biological connection, which might or might not be plausible given chronic exposures of an average intensity. It's not implausible to suggest that the FDA mandates exposure levels so low as to be incapable of causing any measurable effect.

Here's where the hard part starts. A price is paid for every legislative intervention, and someone, somewhere, will be paying the price of that intervention. Is it worth it? Is it worth assuring that tens of thousands of children a year have their mercury exposure reduced to levels that could not possibly have any physiological effect? How much do we value, say, a variegated diet during pregnancy, or the health values of different sorts of fish, or a non-panicked approach to dietary recommendations?

A spokesman from the Environmental Working Group (the chief critic of the FDA's less stringent recommendations with regard to tuna) accused the FDA of abdicating its responsibility to "protect the public." But protection against what, and at what cost? Are we to be protected against a minimal risk, even if such protection incurrs costly and possibly unwarranted intervention?

I would encourage all my readers (my Daily Dozen) to forward my speculations to those more expert in the matter than I am, so I can help clear up my thoughts and the suggestions of this blog.