Medical interest in fish oil began with the observation that heart disease is extremely rare among the Inuit (Eskimo) people despite the fact that they consume a great deal of fat. Close investigation showed that the Inuit diet includes high amounts of an unusual type of fat found primarily in cold-water fish, seals, and whales.
This finding led to an enormous amount of scientific investigation into the potential health effects of what came to be called
. However, despite decades of research and hundreds of clinical trials, it has proved difficult to conclusively demonstrate that fish oil does in fact prevent heart disease.
The active ingredients in fish oil are two fats in the omega-3 category: eicosapentaenoic acid (EPA) and the other docosahexaenoic acid (DHA). Fairly solid evidence indicates that EPA and DHA can reduce levels of
. Triglycerides are a substance that is related to
and helps to create the artery clogging plaque.
Much weaker evidence hints that EPA and/or DHA might:
- Raise HDL ("good") cholesterol levels
- "Thin" the blood
- Prevent dangerous heart arrhythmias
- Slow heart rate
- Improve blood vessel tone
All of these effects would be expected to help protect against heart attacks and other forms of heart disease. Studies designed to examine the effects of fish, fish oil, or its component omega-3 fatty acids on the incidence of heart disease have failed to produce consistently positive findings. Some of have shown benefit; some have shown no effect; one even found evidence that fish oil could, in certain cases,make heart disease worse.
However, in 2007, the results of a gigantic study were published that appeared to settle the question at last. The heart-protective effects of fish oil—so it seemed—had at last been proven beyond a shadow of a doubt. Unfortunately, as we shall see, this study suffers from severe design flaws, and in fact proves little or nothing at all.
Over 18,000 people were enrolled in the trial. All had high cholesterol, and all were using standard drugs in the
family to lower their cholesterol. About half of the participants were additionally given 1800 mg of purified EPA daily while the other half were given no extra treatment. Researchers then followed the participants for about five years. What they wanted to know was whether the use of EPA on top of statins further reduced the rate of major cardiac events (broadly defined to include sudden cardiac death, fatal and nonfatal heart attack, unstable angina, angioplasty, stenting, or coronary artery bypass grafting.)
The outcome was quite positive. Participants in the group taking EPA plus statins showed a 19% reduction of major cardiac events as compared to those taking statins alone. Positive indeed, but, unfortunately, far short of proof. This study was an open trial, meaning that participants knew whether they were taking the treatment or not taking it. Open trials are notoriously unreliable. Only
can actually establish the effectiveness of a medical treatment.
In a double-blind, placebo-controlled study, some participants are given the real treatment while others are given an identical-appearing placebo, and both participants and researchers are kept in the dark (“blind”) as to which is which. This is a very complex subject, discussed in detail in the article,
"Why Does This Database Depend on Double-Blind studies?"
. Here, we will mention only one of the many possible problems that can occur when a study isn’t double-blind: the so-called “halo effect.”
As has been established in thousands of studies, when people know they are being given a treatment in a medical trial, they take better care of themselves in general. They may exercise more, eat a better diet, follow their doctor’s advice more closely and seek medical care for symptoms they might otherwise ignore. The net result can be dramatic improvements in health that have nothing whatsoever to do with the treatment under study.
This is not a merely theoretical problem; quite the contrary, the power of the halo effect is known to be substantial. In this particular study, the entire relative benefit could easily have been due, not to the EPA itself, but to the “halo” that surrounded the use of EPA.
The doubt always raised by lack of blinding was further worsened by a specific detail in the outcome of this study: the fact that no significant difference between the two groups was seen regarding death rate. It was only in the more subjective, elective “major cardiac events” that EPA showed relative benefit, such as in the frequency of angioplasty procedures. These are relatively subjective “outcomes” inherently susceptible to psychological influence.
The bottom line: While it continues to appear likely that fish oil or its constituents offer a heart protective effect, definitive evidence still remains to be presented.
Yokoyama M, Origasa H, Matsuzaki M, et al. Effects of eicosapentaenoic acid on major coronary events in hypercholesterolaemic patients (JELIS): a randomised open-label, blinded endpoint analysis.