Analysis of selected studies suffers from flaws that render its conclusions misleadingly negative
Earlier this week, the Journal of the American Medical Association (JAMA) published a Greek team’s review of 20 studies.
The studies involved 68,680 people, many diagnosed with advanced heart disease and taking heart drugs (Rizos EC et al. 2012). It’s surprising that JAMA chose to publish an evidence review that suffers from the same shortcomings seen in other dubious evidence reviews.
The journal’s decision seems ill-advised, when the public health stakes are so high and the challenges of drawing reliable conclusions from a diverse group of studies in sick people are virtually insurmountable.
Public health agencies and medical societies worldwide agree that omega-3s clearly aid heart health. See the “Official Omega-3 Intake Recommendations, Worldwide” section of our “Omega-3 Facts & Sources” page.
And virtually everything known about the effects of omega-3s in humans suggests that they benefit cardiovascular and overall health.
Strikingly, a Harvard-based team concluded in 2009 that low omega-3 intake was the sixth deadliest of a dozen risk factors for heart-related death … see “Omega-3 Deficiency May Cause 84,000 Premature Deaths”.
Evidence review suffers fatal flaws
The authors, based in Greece, concluded that omega-3 fish oil supplements did not reduce the risk of death overall … or of adverse events (heart attack or stroke) or death related to cardiovascular disease (CVD).
As they noted, the idea that omega-3 fish oil can help prevent cardiovascular disease and heart-related death or injury “has been supported by a number of randomized clinical trials (RCTs) and refuted by others.” (Rizos EC et al. 2012)
And they acknowledged that major public health agencies and medical societies recommend increased intake of omega-3s from supplements or seafood … for healthy people, heart attack survivors, and patients at risk for heart attack, stroke, and heart-related death.
The Greek team – bases at the University Hospital of Ioannina – reviewed 20 randomized, controlled clinical trials to look for associations between omega-3s and major cardiovascular outcomes.
Their conclusion was remarkably bold and sweeping: “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.”
But their conclusion is also highly dubious, as we will explain. The authors’ stated conclusions were based on an ill-advised attempt to reconcile a highly divergent set of studies.
In fact, it’s likely that their analysis understated the benefits of omega-3s to most people to an egregiously misleading extent.
The methods and the participants in the 20 studies analyzed by the Greek team were so varied as to make a meaningful conclusion virtually impossible … so the authors shouldn’t have presented any sweeping conclusion.
Instead, the scientifically responsible approach is to examine the effects of omega-3s in randomized controlled trials that meet certain criteria:
As many expert commenters have noted in the days since publication of the Greek review, no studies have met all of these criteria … but the ones that came closest to doing so show the most benefit.
And significantly, some omega-3 studies show “dose-dependent” reductions in sudden cardiac death (which accounts for half of all heart-related deaths) and in second heart attacks … a strong indication of reliability.
Conclusion contradicted by official advice, prior analyses, and review’s own details
The Greek team’s evidence review is as inherently flawed as several others published in the past three years … for the same reasons.
We’ve reported on three similarly weak, misleading analyses that were persuasively rebutted by leading experts in clinical research on omega-3s and heart health.
This failure to recognize the difference between drugs and omega-3s is common to many such reviews.
Drugs such as statins are generally used to reduce risk factors for heart disease, and can reduce the risk of heart attacks or sudden cardiac death in people with diagnosed heart disease.
If a heart patient is already taking statins and other heart drugs, omega-3s generally do not further reduce the risk of second heart attacks and sudden heart-related death.
However, for heart patients taking drugs, omega-3s will produce physiological effects associated with general cardiovascular health … and they offer benefits related to brain, immune, and metabolic health.
The authors of one review echo the critiques experts are directing at the new evidence review from Greece (von Schacky C, Harris WS 2007):
(Cochrane Reviews is a respected organization that performs reviews of medical studies, certain types of which are called meta-analyses. The new Greek review features a meta-analysis of 20 studies.)
Oddly, researchers keep expressing surprise when evidence reviews of studies in which omega-3s are given to people with advanced heart disease yield mixed or negative results.
Like those studies, the new one from Greece failed to take into account several critical factors:
The last point in our list of “confounding” factors – the unhealthful effects of excess omega-6 fat intake – is often overlooked in analyses of the effects of omega-3s on heart and overall health.
We addressed this topic in our article, “Heart Group's Omega-6 Advice Takes a Huge Hit”.
That article covered a detailed scientific critique of the American Heart Association’s seriously flawed analysis of and guidance on this subject … see also “Heart Association Appears Blind to Risks of America’s ‘Omega-Imbalance’”.
To learn more about this overlooked risk to heart and general health, read “Report Finds Americans Need Far More Omega-3s” and “America's Sickening Omega Fats Imbalance”.
The Greek team’s review also showed that older studies showed more benefit from omega-3 supplementation than the newer studies did.
This fits with the fact that the participants, like other people in developed countries, had probably been taking more cardiac drugs, eating more fish, and taking more fish oil supplements over the past 10 to 20 years.
These factors would mask or reduce any benefits of the added omega-3s being consumed by the study participants.
Vital Choice advisor raises key questions
We asked a true expert on omega-3 science for his take on the new review.
Doug Bibus, Ph.D., is more qualified than most scientists to comment on the Greek team’s analysis.
He studied under the late Ralph Holman, Ph.D., and later managed his lab.
Dr. Holman famously coined the term “omega-3”, first described these fatty acids’ basic structure, and discovered many of their properties. (See “Passing of an Omega-3 Pioneer”.)
Dr. Bibus raised key questions and critiques in an e-mail sent in response to our request for his comments:
Tremendous amounts of basic science and evidence from clinical and epidemiological studies have demonstrated the significance of omega-3 intake for preventing cardiovascular disease (CVD).
Many of the studies included in this review involved patients who were very sick and considered “high risk”, and therefore not likely to benefit from omega-3s.
Why does the current analysis ignore the findings of many cardinal studies that support the understanding of omega-3s in the prevention and development of CVD?
For example, the large, well-designed GISSI and JELIS trials reported significant protective effects of omega-3s … and the large MRFIT (Multiple Risk Factor Intervention Trial) study reported major dose-dependent reductions in risk for developing and dying of CVD.
But it may be too much to expect that any nutrient can reverse cardiovascular disease resulting from decades of poor diet, lack of exercise, and/or adverse genetic profiles.
To say that omega-3 fatty acids do not prevent cardiovascular disease is erroneous based on this review, which included many studies involving people with established heart disease, receiving late-stage cardiac drug therapy.
The authors state that all studies had good compliance … over 80%. How was this compliance measured? The 20% non-compliance rate reported by participants most likely represents a conservative estimate. How many patients will truthfully tell a researcher that they skipped their daily intake of nutrients?
The only true way to follow compliance is with a simple blood test to examine omega-3 levels. Drug studies routinely assay blood levels of medications to follow compliance. Why then has this been ignored in many of these trials and in this meta-analysis?
The outcome of this meta-analysis should be expected. How can you combine so many models, all with inherent variability, and determine a statistically significant effect after controlling for all other variables?
The authors detected trends toward reductions in risk for all adverse cardiac outcomes except for stroke, and for death from any cause … but unsurprisingly, the welter of “confounding” factors in these often deeply flawed, highly diverse studies rendered these trends statistically non-significant.
While the authors of this study find no overall statistically significant effect from additional dietary omega-3s, which side of the risk equation would you want to be on?
Lastly, there is a genuine bias against nutritional supplements in the old guard medical community. This status quo focuses on drug therapies to deal with avoidable cardiovascular disease after it has taken hold.
How often have we just heard, “eat a balanced diet” from medical professionals? What kind of balanced diet? The standard American diet is anything but balanced in its intake ratio of omega-6 and omega-3 fatty acids.
The bottom line? Don't believe the headlines on this deeply flawed, counterproductive evidence review!
Faulty Omega-3 Heart-Health Review