Well, remember that the study was looking at blood levels and not simply fish oil intake. If anything, this is just another cautionary reminder that too much of a good thing ain't so good. Given that the difference between the highest and lowest quartiles was equivalent, so they say, to ~2 servings of salmon per week, the gray area between good and bad might end a little more abruptly than we think.
I'm curious as to what the relationship was across ALL quartiles, though.
Even if we dismissed the plethora of confounding variables that were not accounted for (read: too many to bother listing), is anyone surprised that the relationship between nutrition and health might be a little more nuanced than previously thought?
If anyone has a link to the study, or even just the name, toss it up plox. I'd very much enjoy a skim.