According to the article, "In this type of study, cases of people who have a particular outcome - in this case, prostate cancer - are matched against a random group of people who do not develop the condition."
That sounds to me that in fact they were looking at cancer-free people as well.
Right. My point is that less health conscious people would have un-diagnosed cancer. This would make it appear that more health conscious people (higher omega-3) have higher rates of cancer than baseline.
That sounds to me that in fact they were looking at cancer-free people as well.