A paper (BPA.pdf) published in the March issue of The Journal of Allergy and Clinical Immunology about the association between bisphenol A (BPA) and childhood asthma is nothing short of mind-boggling. Possibly enough so to create a new acronym of data interpretation-- GIMBIO--garbage in, mind-blowing idiocy out.
The authors set out to demonstrate that exposure to bisphenol A (BPA) -- a perennial favorite of anti-chemical, anti-everything groups, which has been portrayed as being responsible for everything from enlarged male breasts (undesirable, at least in my book) to the kidnapping of the Lindberg baby -- has something to do with childhood asthma. Despite the authors' attempt to establish a link between the two, they have actually succeeded in doing just the opposite.
There are so many things wrong here that I need to restrain myself, lest I crash the Internet singlehandedly. But the following should suffice.
Of the entire 13-page article, one paragraph on the first page (results) is more than sufficient to relegate the entire paper to the compost bin. I still can't figure out how they even wrote this with a straight face, let alone got it published.
The paper may be filled with meaningless data, but at least they put in a lot of it. The authors talk about animal models of asthma, analytical methods to detect BPA in urine, the percentage of Dominicans in the study, what antigens contribute to human allergy, nitric oxide, the presence of nicotine metabolites in blood, cytokines and cytomegalovirus. I didn't see the Yankee's spring training lineup in there, but then again, I did not request the supplemental information.
And all of it is meaningless. Which was patently obvious by simply examining one finding at the very beginning: the presence of BPA can either exacerbate, or prevent asthma, depending on age. Children whose mothers had higher levels of BPA in their urine while pregnant had less asthma at age five, but five-year olds with higher BPA in their urine had more asthma. Please.
To make matters worse, although we are all chronically exposed to BPA in minute quantities because of its ubiquitous use (mostly in canned food as a sealant) the levels of children aged three and five didn't even correlate with levels of children at age seven. So, which do you believe? How about neither?
And if this smells fishy, it gets worse. The incidence of asthma was determined by a questionnaire--a notoriously unreliable method of data gathering, along with one examination by a doctor per child at ages that ranged from ages five through twelve.
When a paper is filled with statistics that are supposed to support a hypothesis and the authors have to resort to ridiculous speculation to explain away obvious inconsistencies, the answer is almost always the same--garbage numbers. Which is what I strongly suspect is the sum total of this paper.
Yet this did not stop them from concluding: Asthma prevalence has increased dramatically over the past 30 years, which suggests that some as-yet-undiscovered environmental exposures may be implicated... Our study indicates that one such exposure may be BPA.
Uh, no it doesn't.
For this to be true, we need to accept that BPA, which has the ability to either increase or decrease asthma (which itself was measured in an unreliable manner) and has been in routine use for 60 years, is responsible for rising asthma rates over the past 30 years--depending on how old you are.
Given an impossible concoction of gyrations necessary to justify the paper's conclusions, or common sense telling me that the data are all nonsense, I will cheerfully pick the latter.
Publications like this do cumulative harm in the pursuit of legitimate science-based policy, as they make flashy headlines that contribute to the anti-chemical feeding frenzy, while at the same time obscuring the real message of the study.
Perhaps The Journal of Allergy and Clinical Immunology ought to reexamine its publication standards.