Another potential problem is that most of the research on fish oil and heart health—including all of the trials included in the recent JAMA analysis—have involved subjects who already have heart disease or established risk factors. Whereas this isn't necessarily a problem in itself, it means that very little research has addressed whether fish oil supplements benefit healthy people. "The jury is still out on whether omega-3 supplements can prevent a first cardiovascular event in people at usual risk," says JoAnn Manson, an epidemiologist at Brigham and Women's Hospital in Boston who is conducting a trial to answer this question, which she estimates will be finished in 2016.
Moreover, because so many trials have involved subjects with heart problems, subjects in recent years have been "taking multiple medications, such as aspirin and statins, which can obscure the effects of supplements," Manson says. (Half of the studies included in the JAMA analysis were conducted after statins became commonplace.) This fact could also help explain the outcome discrepancies between recent trials and older ones carried out during the pre-statin era. Indeed, a February 2012 analysis of a large European clinical trial reported that fish oil supplements do not prevent second heart attacks among people taking statins, but that it cuts risk by half among people who don't take the medications. (Because there were so few non-statin users enrolled in the trial, this finding did not quite reach statistical significance.)
People who enroll in and complete omega-3 trials may differ from typical Americans in important ways, too. "Members of the public who volunteer to join randomized controlled studies are frequently healthier and more active than the average for the population," Dangour says, which could affect outcomes in unknown ways. In addition, he adds, people who drop out of trials are often the sickest, and they might be the ones who would most benefit from supplementation.
There's also more than one type of fish oil. Typically, in omega-3 intervention studies, subjects take pills containing a near-equal mixture of two fats, docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA)—but some research suggests that in certain situations and for certain outcomes, one may be better than the other. For instance, a 2011 meta-analysis concluded that fish oil capsules don't help treat depression, but a group of British and Norwegian researchers challenged these findings, citing evidence that pills containing at least 60 percent EPA do seem to provide mood benefits. Another controversial question is whether the omega-3 dose is important in and of itself or whether the ratio of omega-3 to omega-6 fats one consumes is more important. (Although omega-6 fats are important for survival, Americans tend to consume far more of these fats than they need.)
Finally, whereas the recent JAMA analysis concluded that fish oil has no effect on cardiovascular outcomes, the researchers did find that omega-3s reduced the risk of cardiac death by 10 percent, an effect that was statistically significant (having a "p value" of 0.01). The researchers did not report the finding in their conclusions because they subsequently modified their statistical calculations to account for the fact that they had used the same data set to ask a number of different "exploratory" questions: In this case, does fish oil prevent heart attacks? Strokes? What about sudden cardiac death? The team wanted to tighten their definition of statistical significance to account for the fact that the more questions one asks, the more likely one is to get a positive result by chance. Still, Mozaffarian says, "if you combine all the data and look only at cardiac death, there is a statistically significant benefit. A 10 percent reduction in the number-one cause of death in both men and women in the U.S. is a big deal."