Environmental Health News reports:
Exposure to bisphenol A (BPA) at levels commonly found in the general population may cut a woman’s chance of getting pregnant if she is undergoing fertility treatment, a study from Harvard University finds. Those with higher BPA levels were less likely to get pregnant than women with lower levels. The link was stronger in women having more intense fertility treatments. The pregnancies failed because the embryos did not attach to the uterus. While animal studies show similar results, this is the first time researchers report a link in people. BPA is widely used in some plastics, most food can linings and certain receipt paper.
The study is one of those human epidemiology studies that are hardly worth the paper — this time of women undergoing in vitro fertilization treatment. The study claims there is a “positive linear dose-response association between BPA urinary concentrations and implantation failure.” However, the study results actually show the opposite.
The results are not significant (p-value = 0.06) and only show a non-significant trend as a p-value of 0.05 or less is needed to be considered statistically significant. In short, these results could be due to chance alone.
Furthermore, even if the results had been statistically significant, they would only have indicated an association, not cause and effect. The authors, in their discussion and conclusion are not only trying to suggest that this chance association is significant, but also are trying to suggest that it is a meaningful biological cause and effect relationship. Furthermore, they attempt to present biological evidence (in the discussion) that BPA can cause implantation failure when the only credible evidence of BPA effects on reproduction is from animal studies at doses many thousand times higher than human exposures in these studies.
In short, the results of the study do not show what they claim, as the results in fact are not statistically significant. Even if the results were statistically significant, they wouldn’t indicate cause and effect. And the results have already been contradicted by high quality animals studies that show that BPA does not cause implantation failures or any reproductive or health effects at the tiny levels to which humans are exposed.
I’m sure the “geniouses” from Harvard do know statistics. It’s all about money (keeping it coming-trolling) and ideology. Not much science here.
If I recall, similar work originally established the ‘endocrine disruptor in plastic’ meme. Something about plastic lab equipment changing results of experiments. Trolling for research money with stuff like this, and the unfortunate headlines that result, should be considered a species of misconduct. It’s certainly a gross public disservice.
Don’t these genuses from Harvard know the differences between noise and real correlation. It seems that they have some magic power to see results in data that does not exist.