Got science? You should. That’s because science underpins the success of the nutrition world.
It’s a trick, of course, to conduct pharma-style double-blind, placebo-controlled human clinicals on slower-acting nutrients, and there is a growing chorus of researchers and advocates pushing for a different research paradigm based on biomarkers that can suggest health outcomes.
Regardless, there are three points I’d like to make related to nutrition science before I get to the punch line you’re all waiting for: what we are to make of this week’s omega-3 study.
1. Meta-analyses are convenient methods of taking stock of the state of the science regarding a particular item, but they should always be taken with a grain of salt.
That’s because bias, either via inclusion or exclusion, invariably creeps in. These are not new and original studies where patients are given a bioactive and researchers see what the effects are; these are when librarians scan the online literature to try to make heads or tails of a bioactive. And it always depends on exactly what they’re looking for, and two sets of researchers can and do come up with different conclusions.
2. One study does not a conclusion make.
This is certainly true with a meta-analysis, but also for even a perfectly constructed and conducted human clinical trial. That’s why the FTC is deigning to make the “two clinical trial” rule the rule of the land for making any sort of implied health claim with nutrients. You need to validate your results with another study, by a different set of researchers if you please. Put another way: Trust but verify.
3. Any time a published paper contains researchers’ words to the effect that their study makes a bullet-proof conclusion, that no further studies need be conducted, that Truth is contained herein, that you can disregard the efficacy of said ingredient by virtue of these particular single study results—big fat grain of salt alert.
In fact, when I read studies, I start by reading the last paragraph to see if the authors make such a bold conclusion. Researchers tend to be pretty close-to-the-vest people, but when they make such end-game assertions, I stop reading right there. Hubris doesn’t impress.
Picking apart the JAMA study
Which brings me to the infamous meta-analysis of omega-3s published in the influential Journal of the American Medical Association (JAMA) today.
It was fine, as far as meta-analyses go. Selected studies had to be longer than one year in duration and the mean omega-3 dose was 1.51 g/day (770mg EPA and 600mg DHA).
Note, however, the wide variations in dosages and in the specific disease states being studied.
For example, one study included in the analysis was of 120 subjects with leg muscle pain caused by poor blood flow (claudication). Patients took 270mg EPA/day to discern if that would help all-cause mortality, cardiac death, heart attacks and stroke. Conclusion: “small reduction in non-fatal coronary events…that warrants further investigation.” I’m thinking that is pretty impressive given only 270mg of only EPA.
Another study included in the JAMA meta, meanwhile, used a much larger dose—2,900 mg EPA and 1,900mg DHA—in patients with hardened arteries (atherosclerosis) and found a statistically significant 30 percent reduction in triglyceride levels (hello, Lovaza!) but only small changes in its primary end point, the diameter of the hardened arteries.
Incredulously, the researchers made mention of the first big study on fish oil: the Italian GISSI trial of 2002, which gave 1,000mg of fish oil supplements for a year to more than 11,000 patients who had suffered a heart attack in the previous three months. The fish oil group had significantly fewer deaths. However, the researchers did not include the GISSI in their final analysis.
Which made it much easier to come to the conclusion that omega-3 PUFAs vis a vis major cardiovascular disease events had a utility “lingering around a small effect and borderline statistical significance.”
Is this, then, the best we can hope for with omega-3s?
I say no. That’s because these studies tended to study people who already were suffering from a disease state. They already had diagnosed atherosclerosis, had already suffered a heart attack or were already cruising for a bruising.
In addition, these patients were routinely already receiving the standard of care for mainstream medicine: lots of pharmaceuticals. So, if you have seriously ill patients taking a suite of drugs, and then add a supplement to the mix, and the results don’t change significantly, does that mean the supplement does not work, or that it’s not as powerful as a drug, or what?
“I fear that in the fog of science the first one to get shot is the dietary supplement,” is how industry veteran Loren Israselsen put it to me.
Now, although more than a few nutritionists and advocates declare that nutritional supplementation at appropriate doses can and do actually treat disease, according to the regulatory structure we have in the U.S. you cannot officially say that. Supplements, according to the regulators, are strictly for supporting healthy bodily structures and functions.
Because omega-3s look like they are only slightly effective for those with serious pre-existing cardio conditions, and they look like they at least keep healthy people healthy, I’m going to hazard that the time remains ripe for those who are somewhere between healthy and diseased.
Here’s an idea from Harry Rice, PhD, vice president of scientific and regulatory affairs for the trade group GOED Omega-3. He said, "We know from the studies that have shown a benefit that future research in this area should only analyze studies that do not include confounding medications, are longer than two years in duration, and use dosages greater than 1-2 grams of omega-3s per day. Very few of the studies included in this meta-analysis met these criteria."
Mainstream media's response
On ABC news last night, Diane Sawyer sat fawning over the recommendation of their house doctor, Richard Besser, MD, who said, “Unless your doctor tells you that you have a true nutritional deficiency, in my opinion, this is not a good use of people’s money.”
I’m thinking if Diane Sawyer was worth her salt, she would ask the easy follow-up question, “So, Dr. Besser, how many people actually have a nutritional deficiency in these essential fatty acids?”
Dr. Besser, if he really knew anything about the subject, would say that the recommended intake is about 500mg per day and the average American gets closer to 150 mg per day.
Oh, and for the researchers’ ultimate conclusion in this meta-analysis? “Our findings do not justify the use of omega-3 as a structured intervention in everyday clinical practice or guidelines supporting dietary omega-3 PUFA administration.”
This would be news to those organizations that have issued guidelines vouching for omega-3s: The World Health Organization, the American Heart Association and the U.S. National Academies of Science.
When even these staid old outfits are advocating omega-3 DHA and EPA, I take you back to my second point, above: One study—and certainly not a meta-analysis (see point 1)—does not a conclusion make.