Why Omega-3 Studies Fall Short: Problems With Analysis

A recent article has been published stating that EPA is superior to combinations of EPA and DHA in treating cardiovascular disease (1). The conclusion was drawn based on analyzing cardiovascular outcomes in various published trials from 1995 to 2020. The findings were based on using meta-analysis, a type of statistical analysis that combines results from several scientific studies. This paper is an excellent example of why meta-analysis is often not useful for reaching meaningful scientific conclusions.

Problem Number 1: Omega-3 Fatty Acid Dosage

The first problem in analyzing published trials from over 25 years is that virtually none used the same omega-3 fatty acids, amounts, or even the identical placebos (1). This lack of consistency between these reported trials makes it virtually impossible to come to meaningful conclusions.

Problem Number 2: Blood Markers

The second problem is you need to demonstrate that a specific blood marker can be related to consuming the omega-3 fatty acids and the final clinical outcome. For statins, changes in LDL cholesterol have been the standard blood marker used to link the statin dose to a successful clinical result. For omega-3 fatty acids and cardiovascular disease, the changes in omega-3 fatty acids in the blood should be the linking factor relating the dose to the clinical outcome. Such a blood marker could either be the omega-3 index or the AA/EPA ratio. Without using an internal blood marker that is replicable, the results from one trial to the next simply cannot be compared. The need for an internal blood marker is illustrated in the two studies where EPA was used as monotherapy. Based on previous trials, it can be estimated that the AA/EPA ratio in the REDUCE-IT study probably decreased from approximately 14 to 1.4. (2,3). In the REDUCE-IT trial, this decrease in the AA/EPA ratio was associated with a 25 percent reduction in cardiovascular events in for those taking 3.9 grams of EPA per day (2). The patients in the JELIS study using the same EPA preparation started with an AA/EPA ratio of 1.6 and using 1.8 grams of EPA per day further decreased the AA/EPA ratio to 0.8 (4). This further decrease in the AA/EPA ratio was associated with an additional 20 percent reduction in cardiovascular events. Therefore, a simple assumption might be that it would require approximately 6 grams of EPA per day to reduce the AA/EPA ratio found in Americans and Europeans to reach an AA/EPA ratio of 0.8 found in the JELIS study. The same simple assumption would suggest obtaining an AA/EPA ratio of 0.8 would provide an approximate reduction in cardiovascular events of 45 percent. That would represent a quantum leap in cardiovascular therapy compared to any known drug. Furthermore, using a blood marker such as the AA/EPA ratio also allows comparisons of various doses of the omega-3 fatty acids to distinguish between a placebo dose and a therapeutic dose (5). Without titrating to the same blood marker (such as the AA/EPA ratio), comparing the results from one trial to another becomes virtually worthless.

Problem Number 3: Omega-3 Quality / Amount of EPA and DHA

The third problem is the quality of the omega-3 fatty acid preparations used. Some studies used only EPA, other studies used different combinations of EPA and DHA, and no study looked at DHA alone. It should be noted that those studies using only EPA alone had higher levels of atrial fibrillation compared to studies that used combinations of EPA and DHA (1). The most likely reason may be that many powerful pro-resolution hormones are derived from DHA. Published data has indicated that using EPA alone in the absence of DHA did not increase DHA and may have even decreased DHA levels (3). Thus, the lack of any increased DHA may be an underlying cause of increased atrial fibrillation (6,7).

Furthermore, there are differences in the structures of the omega-3 fatty acids used. Some were ethyl esters, some used free fatty acids, and none of the studies used triglycerides. This problem is particularly relevant for the free fatty acid form of the omega-3 fatty acids used in the STRENGTH study. This form of omega-3 fatty acids is highly prone to oxidation, eliminating any possible benefit of the omega-3 fatty acid supplementation (8).

Problem Number 4: Knowing the Root Cause of Heart Disease

The final problem is addressing the actual cause of heart disease. Is it caused by LDL cholesterol, triglycerides, or inflammation? As the Roman philosopher, Seneca said nearly two thousand years ago, “if you don’t know which port you are sailing to, then no wind is favorable.” In other words, if you don’t know what the primary cause of a disease is, then you are only guessing about a treatment. Cardiovascular disease is now appreciated as primarily an inflammatory disease, not a disease of LDL cholesterol (9). This new understanding of the underlying cause of heart disease is why omega-3 fatty acids offer a breath of fresh air into the future treatment of cardiovascular disease as long as you have clinical trials that seek to reduce inflammation instead of focusing solely on lipids. The CANTOS trial confirmed this by demonstrating a significant reduction in cardiovascular events using an anti-inflammatory monoclonal antibody without any change in LDL cholesterol levels (10).


In summary, the role of omega-3 fatty acids has exciting potential in the treatment of cardiovascular disease. However, that promise will only be fulfilled if clinical trials are well designed so they can be replicated. Throwing together ill-defined clinical trials together that can’t be appropriately compared and then coming up with a conclusion is equivalent to making scientific sausage.



  1. Khan SU et al. Effect of omega-3 fatty acids on cardiovascular outcomes: A systematic review and meta-analysis. EClinicalMedicine   doi.org/10.1016/j.eclinmed.2021.100997
  2. Bhatt DL et al. Cardiovascular risk reduction with Icosapent Ethyl for hypertriglyceridemia. N Engl J Med. 2019 380:11-22.
  3. Ballantyne CM et al. Icosapent Ethyl effects on fatty acid profiles in statin-treated patients with high triglycerides: The randomized, placebo-controlled ANCHOR study. Cardiol Ther. 2019 8:79-90.
  4. Yokoyama M et al. Effects of eicosapentaenoic acid on major coronary events in hypercholesterolaemic patients (JELIS): A randomised open-label, blinded endpoint analysis. Lancet. 2007 369:1090-8.
  5. Sears B. Omega-3 fatty acids and cardiovascular disease: Do placebo doses give placebo results? CellR4. 2017 5: e2302.
  6. Ramadeen A et al. Docosahexaenoic acid, but not eicosapentaenoic acid, supplementation reduces vulnerability to atrial fibrillation. Circ Arrhythm Electrophysiol. 2012 5:978-83.
  7. Tribulova N et al. Omega-3 index and anti-arrhythmic potential of omega-3 PUFAs. Nutrients. 2017 9:1191.
  8. Nicholls SJ et al. Effect of high-dose omega-3 fatty acids vs corn oil on major adverse cardiovascular events in patients at high cardiovascular risk: The STRENGTH randomized clinical trial. JAMA. 2020 324:2268-2280.
  9. Geovanini GR et al. Atherosclerosis and inflammation: Overview and updates. Clin Sci. 2018 132:1243-1252.
  10. Ridker PM et al. Anti-inflammatory therapy with canakinumab for atherosclerotic disease. N Engl J Med. 2017 377(12):1119-1131.
2 replies
  1. George says:

    I’ve been listening to nutritionists recite studies that omega 3 supplements, cause atrial fibrillation and inflammation long-term. Is there any serious fesr for this? Thanks a pot for your time doctor!!!

  2. Barry Sears says:

    Although there is an association of omega-3 fatty acid consumption and A-fib, there is an opposite correlation between CHD mortality and omega-3 fatty acid blood levels. This is most like due to the reduced by the ability of omega-3 fatty acids to reduce clot formation that can is a primary concern with increased A-fib. This is discussed in greater detail in a recent review (http://dx.doi.org/10.1136/openhrt-2018-000904).

    The best test to determine the future CHD mortality is the AA/EPA ratio. If your AA/EPA ratio is between 1.5 and 3,, then the CHD mortality rates are significantly lower than a higher AA/EPA ratio.

    There are no studies indicating that fish oil causes inflammation.


Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *

More to explore

Ozempic Weight Loss Fever

Americans are increasingly desperate to lose weight as our national obesity rate increases.  The current mania for injectable weight loss drugs like

© Copyright - Dr. Sears.com