In an Slate article critiquing Marian Burros’ story about mercury levels in fish, Jack Shafer takes issue with the omission of any discussion of a review published in 2000 in Environmental Research entitled “Twenty-seven Years Studying the Human Neurotoxicity of Methylmercury Exposure”. Mr Shafer goes on for a while about the findings of this paper, but I find that his argument, on the face of it, to be very odd. Unless the review is a seminal piece of work, or the most up to date analysis, why should it be discussed? Hundreds of papers come out on mercury every year, and several discussing the epi studies done to date. This study loses on both fronts: It doesn’t discuss all the epi studies, doesn’t have the most recent information, and isn’t a work that is widely considered bedrock information by the scientific community. That’s not to say it’s a bad review, it’s simply inappropriate for why Mr Shafer is trying to use it for. Because this review came to a different conclusion than that of the Times article, Mr Shafer thinks Ms Burros is ‘scaremongering’ (I didn’t see anything in his critique that supported a conclusion that the article was trying to scare anyone, or did scare anyone, but no matter, maybe he was scared by it).
What is going on, in fact, is that Mr Shafer is cherry picking and is the one guilty of the sins he is accusing the Times of. First, Mr Shafer never even gets around to addressing why the Times’ use of the EPA’s numbers is wrong (this isn’t part of the Environmental Research paper either). Second, let’s try and find a easily available, seminal work on mercury exposure and the risks and benefits that everyone can agree comes from a group with impeccable credentials. Hmmm…Found. How about the National Academy of Sciences’ Institute of Medicine report, Seafood Choices: Balancing Benefits and Risks. It’s free, easy to find, recent, and comprehensive. Why isn’t it the article mentioned in the Mr Shafer’s critique? I’m not sure but it points to the problem that he is trying to slam the Times for. He doesn’t know the field and, it seems, either didn’t talk to someone who did or talked to only one or two people with some sort of agenda (the EPA/FDA/ATSDR feud on methylmercury is only one of many in the field). No one really knows why the Faroes and Seychelles studies came to different conclusions. Many ideas have been put forward, but none are compelling (to me or the Institute of Medicine). That doesn’t make one of them automatically wrong (especially without explanation).
But it’s this little snippet from the Slate article that really got me:
But dose determines toxicity, something a newspaper staffed with as many accomplished science reporters as the Times surely knows.
Hello, it’s the overused canard of “dose makes the poison”. Well, of course it does. But that doesn’t mean that a small dose is safe; in mercury as for lead, a little goes a long way. Interestingly, an analysis of the data from the Faroes, Seychelles, and New Zealand was done using maximum likelihood and Bayesian hierarchical models in 2007. A dose-response relationship was determined that suggests that children’s IQ declines by 0.18 points for each ppm increase in maternal hair mercury level. It may be possible and seems likely to me that there is no threshold for methylmercury effects on the nervous system, only a threshold for where you can clinically detect a deficit.
Mr Shafer is an excellent media critic but on this one, he was really out to lunch (hopefully not tuna sushi).