Fish oil's heart benefit shrank as the studies got bigger

Fish oil is one of the most widely taken supplements, and omega-3 fatty acids have long been promoted as protection for the heart. Yet the randomised trials testing that promise have returned conflicting results, some strongly positive and others flat. That confusion matters most for people at high cardiovascular risk, the group with the most to gain if the supplements genuinely work. This analysis set out to pool every relevant trial, estimate the true effect on survival, and hunt for the reasons the earlier studies disagreed with each other.

The researchers systematically searched the Cochrane Library, MEDLINE and EMBASE for interventional trials comparing omega-3 fatty acids against placebo or usual diet in high-risk cardiovascular patients. The primary outcome was death from any cause; secondary outcomes were coronary restenosis, the re-narrowing of arteries after a stent procedure, and safety. Twenty-nine randomised trials covering 35,144 patients met the inclusion criteria, with 25 reporting mortality and 14 reporting restenosis. The team pooled results using Bayesian random-effects models and used meta-regression to explore why the trials pointed in different directions.

Omega-3 supplementation was not associated with a statistically significant reduction in deaths, with a relative risk of 0.88, nor with prevention of restenosis, at a relative risk of 0.89. The probability of at least some benefit did remain high for both outcomes, at 0.93 and 0.90 respectively. The most telling result came from the meta-regressions: there was a greater than 90% probability that larger studies, and those with longer follow-up, showed smaller benefits. No serious safety problems were identified across the trials.

This analysis helps explain the conflicting picture from individual omega-3 trials: much of the apparent benefit traces back to smaller and shorter studies, exactly the pattern expected when weaker methods or selective publication inflate an effect. It does not close the question, since a modest real benefit remains plausible, and the authors call for additional high quality trials to clarify omega-3's role in secondary prevention of cardiovascular disease. It stands as a caution against taking the early positive trials at face value rather than a definitive verdict that the supplements do nothing.

For anyone taking fish oil in the hope of protecting their heart, this is a reason to hold that expectation loosely. In the highest-risk patients studied, the people most likely to show a benefit if one exists, the pooled effect on survival did not reach statistical significance, and the more rigorous the trial, the smaller the benefit appeared. There was no safety signal, so the analysis gives no reason for alarm, but it suggests any protective effect is modest at best and that some of the enthusiasm rests on weaker evidence.