> A hypothesis that Christel Nielsen's research group had before the study was that the size of the tattoo would affect the lymphoma risk. They thought that a full body tattoo might be associated with a greater risk of cancer compared to a small butterfly on the shoulder, for example. Unexpectedly, the area of tattooed body surface turned out not to matter.
As someone pointed out on the other thread about this[1], the lack of dose response makes it very difficult to see this as a direct correlation and not reflective of other confounding factors.
Why so? Zero-order effects are not unheard of in biology. Example: the elimination rate of ethyl alcohol is dose-independent because any practically noticeable concentration of ethyl alcohol will saturate available alcohol dehydrogenases.
As someone pointed out on the other thread about this[1], the lack of dose response makes it very difficult to see this as a direct correlation and not reflective of other confounding factors.
[1] https://news.ycombinator.com/item?id=40492364