CRN debunks omega-3 meta-analysis as 'inherently inconclusive'
CHICAGO — As part of a meta-analysis spanning 70,000 patients, a study published in the Sept. 12 issue of JAMA determined supplementation with omega-3 fish oils was not associated with a lower risk in heart disease. Lead researcher Evangelos Rizos of the University Hospital of Ioannina in Greece, concluded that omega-3 supplement recommendations may be overblown. "Our findings do not justify the use of omega-3 as a structured intervention in everyday clinical practice or guidelines supporting dietary omega-3 … administration," he wrote.
However, given the many shortcomings around meta-analyses of a nutritional supplement's impact on disease states, "consumers should not discount the many proven benefits of omega-3 fatty acids in all stages of life," countered Duffy MacKay, VP scientific and regulatory affairs for the Council for Responsible Nutrition.
"This study does not change the current recommendations by authoritative bodies, such as the World Health Organization, American Heart Association and the U.S. National Academies of Science, who recommend adequate consumption of omega-3 fats," MacKay said.
The value in meta-analyses is found in combining comparable smaller clinical studies to assess whether similarities in the combined results exist. The problem is in finding studies that are actually comparable. "This meta-analysis combined studies that were not comparable in their design … which makes the results more skewed," MacKay noted. "Second, omega-3 fatty acids are vital nutrients and not drugs. Many of the studies included in the meta-analysis were conducted on diseased individuals already undergoing treatment with one or more drugs, such as statins, which may mask the less potent and more long-term effects of omega-3 fats."
Along these lines, the researchers apparently did not examine within each individual study included in the meta-analysis whether individuals in the placebo group were sufficient or insufficient in their dietary intake omega-3 fats, MacKay suggested. "In this regard, studies on drugs are far simpler than those of nutrients [where] the treatment group gets the drug and the placebo group does not," he said. "With nutrients, if participants in the control group already have a diet sufficient in that substance, then it will be that much harder to demonstrate any benefit among the treatment participants. It is impossible for five researchers to control the diet of almost 70,000 patients over several years (particularly as a retrospective meta-analysis), as omega-3 fats are widespread throughout a variety of foods."