A real Swedish meatball of statistical inference debunked by reality.
Swedish researchers claim to correlate PCBs with atheroscerlosis in a population of elderly Swedish women.
Rather than debunk this with our standard weak association epidemiology analysis (which we could but lack the interest), we will just point out that PCB workers, presumably with higher exposure to PCBs than the elderly Swedish women, actually have less heart disease than the general population.
Your link to the journal article you claim suggests that incidence of heart disease in “PCB workers” is lower than the incidence of heart disease in the “general population” redirects to an article about the link between PCBs and cancer which, unlike the link between PCBs and atherosclerosis, is disputed. The following journal articles address the PCB-atherosclerosis link:
Heyler et al. 2009 “PCB-induced endothelial cell dysfunction: Role of poly(ADP-ribose) polymerase”
Lim et al. 2008 “Coplanar polychlorinated biphenyl-induced CYP1A1 is regulated through caveolae signaling in vascular endothelial cells”
Savouret et al. 2003 “The aryl hydrocarbon receptor and its xenobiotic ligands: a fundamental trigger for cardiovascular disease”
This is the “healthy worker” effect exposing this junk science for what it is, junk. People are told what the “scientists” want you to believe, even if it is just nonsense. Humans must have undergone a mutation of Genetic Modification status that has reduced the capacity to think for themselves in any rational manner.
Thank you, thank you, thank you!!!! Finally another web site with rational scientists writing. You might think that journalists would know enough statistics to realize that correlation does not equal causation, sadly, you would be mistaken. It is remarkable what is allowed to pass as science these days.
I detect statistics abuse. I did not see a single reference to a chi-square value – the single most important measure of variance in rare-event statistics.
But Rats and mice yield data more suitable to environmentalist’s causes.