Reports that “fish oil” might “cause” cancer

A couple of clients have asked about the media reports which have suggested a causal link between fish oil supplementation and prostate cancer. I do not want to spend a lot of time on this so here are the major problems with the reporting:

  • No participant in this study was tracked on their use of fish oil or consumption of fish. For all anyone knows, none of the participants consumed any fish or fish oil at all.
  • The participants in this study were men over 55, more than 70% of whom were overweight or obese and more than 65% were current or former smokers.
  • No causal link can be established from this methodology.

Yet despite these issues people were subjected to headlines such as:

  • “Those Fish Oil Supplements Might Cause Cancer” (Forbes)
  • “Omega-3 supplements linked to prostate cancer” (Fox News)
  • “Hold the salmon: omega-3 fatty acids linked to higher risk of cancer” (Time)

However the pushback on this study has been pretty swift. Here are several for anyone who really wants more details:

And, of course, if you want to read the study yourself:


There are 4 other important points I think are worth making:

  • By critiquing the reporting on this study, especially by emphasizing the limits of this research, I do not wish to imply that I think fish oil is always a good supplement for people to take. I concur with some of the sentiments expressed in some of the responses linked above that too much fish oil could be detrimental, or at least not beneficial, to people’s health. I have observed that fish oil can be useful in remediating a deficiency in omega-3’s but it is not something people should continue to take after that point as it can contribute to an imbalance of excess omega-3’s.
  • The underlying assumptions in the research and the media reporting is very reminiscent of the research on cholesterol. The implication from the conclusion of the paper is that elevated serum DHA “is” the result of excess dietary consumption of omega-3’s. However, as with cholesterol, there is no reason to assume a simple relationship between dietary consumption and serum levels of a nutrient. I would suggest that it is more likely that elevated serum levels of a nutrient represents an increase need for that nutrient that has upregulated its production/release or downregulated its breakdown/elimination or a combination of both.
  • A final strange finding that I didn’t think was important to list as a major problem but worth noting is that serum levels of 2 types of trans-fatty acids were very slightly but significantly lower in the high-grade cancer subjects. This would seem contradictory to the large body of research supporting a link between trans-fatty acids and the development of cancer. However, I think there is a hypothesis that explains this apparent contradiction: the lower serum levels of trans-fatty acids may indicate those fats were already incorporated into cells in those with the high grade cancer and thus appeared at lower levels in the blood.
  • As has been stated in some of the other responses to this research, a statement by the lead researcher suggests a bias against nutritional supplementation. His statement to reporters: “We’ve shown once again that use of nutritional supplements may be harmful.”

Leave a Reply

Your email address will not be published. Required fields are marked *