Does Omega-3 Work Miracles?

About twice a year I go on a health kick that lasts a few weeks. Typically this involves going for one-mile runs two or three times, doing as many push-ups as I can (about eight) every night, increasing the fiber in my diet, ramping up my carrot juice consumption, and taking whatever health pill is currently in vogue.

I’m right in the middle of my latest health kick right now. My pill of choice this time around is Omega-3 fish oil. Omega-3 is alleged to have all sorts of wonderful benefits including reducing heart disease and aiding mental function. Indeed, we have blogged about the beneficial impact of Omega-3 on prisoners twice (here and here) in the past (although because of my low Omega-3 intake prior to my latest health kick, my memory was so poor that it took Dubner to remind me of this fact).

I don’t feel any different today than I did two weeks ago when I started on Omega-3 — not that I expected to. The same cannot be said for Seth Roberts (author of the Shangri-La Diet, subject of a Freakonomics column in the New York Times, and self-experimenter extraordinaire). In two blog posts, Seth details profound and immediate impacts of Omega-3 on his balance. Every morning, for reasons I can’t pretend to understand, Seth climbs on top of his “balance-o-meter” and sees how long he can balance. After he increased his dose of Omega-3, the very next day he established a personal best on the balance-o-meter.

Could it really be the extra tablespoon of flaxseed oil (which is high in Omega-3) that did it? Seth thinks so. I’m more skeptical. An alternative explanation is that it is not the Omega-3 itself, but rather, Seth’s belief that Omega-3 will work that helps him balance. Of course, if all Seth cares about is good balance, it doesn’t really matter whether it is the Omega-3 or his belief in it.

D Brooks

While it's nice to hear that somebody is feeling better, such stories should always bring to mind the quote attributed to toxicologist Frank Kotsonis:

"The plural of anecdote is not data."


Great article you referenced, Coyotesqrl. It should be required reading for everyone in Our Obese Nation. It should be under magnets on every fridge, posted on every community bulletin board, nailed to every church door.

Oh, right, I forgot. It's the NYTimes, that lying liberal rag. Well, at least HALF of Our Obese Nation could read it.


Don't take your vitamins in pill form. There is no proof that they are equivalent to sources from food. Factory farming has removed (inadvertently) so many nutrients from our food. To get your Omega 3 and other less trendy and perhaps yet unnamed vitamins, eat eggs from chickens who are allowed to roam in fields and eat bugs. As an additional benefit the chickens have a nice life instead of being crammed four to a tiny cage, with nothing to do but peck their fellow torture victims. If you think that is hyperbole, watch a chicken taking a dust bath in the sun.


PollyQ beats me to the punch. It would be hard for Seth to prove (or disprove) on his own the effects of omega-3.

Reading Seth's blog a little more closely, he originally felt that omega-3 was improving his sleep and felt more balanced, so maybe there's the reason: he's more awake and thus more aware of his balance. Still hard to say that omega-3 was improving anything.

Steven also hits the point, of course, on placebo effects. Seth could have somebody give him oil "A" for a few weeks and oil "B" for another few weeks and then he should compare the data. Somehow you would have to make them taste the same, though.

Some people in his blog suggested using chess as a test. I imagine this test to be much harder than his balancing test to figure out any effects are from practice or from something else, not to mention you have to figure out effects from your opponent.

In general his balancing test is way too esoteric. He should also test something simple yet popular like free throw shooting. If he saw improvements there, you would see lots and lots of people trying and that would sure help prove or disprove any effects from omega-3. Hell even the pros would try it if somehow data trickled up to high school and then college levels. I'm not sure a free throw would be a good measurement to show that omega-3 is brain food, but his test should be something that simple and popular. Anyways my long way of saying: repeatability is important.

I still like Seth's thinking. He's testing hypotheses and making new ones depending on data, ("You know, not only I my sleep's improved, but I feel better balanced. Let me test that out to see if it's true.") I sure wish more people think this way.



There are several good arguments posted above; however, I think you have to look at the overall science of how omega-3 fatty acids work at a molecular level to make any of these assessments. Since I have worked on omega-3 supplement research, I can give a little of that insight.

1.) It is true that the type of omega-3 fatty acid does change the biological effect. The type of omega-3 in flaxseed oil is alpha linolenic acid (ALA). It is a shorter omega-3. Omega-3s get elongated in the body through a series of metabolic reactions to the longer omega-3s, EPA and DHA. EPA and DHA are the forms that are in fish, fish oil and most dietary omega-3 supplements. They are much more bio-available (meaning your body can use more of it faster for the “omega-3 benefit”). If you are taking flaxseed oil as your source of omega-3s, then you are not going to be getting much of the “omega benefit.”
2.) As noted above, the balance between omega-3 and 6 is essential. The American diet is saturated with omega-6 fatty acids that contribute to an inflammatory response (inflammation in the body can lead to diseases like diabetes, IBS, hypertension, heart disease, and possibly even depression). By getting the right balance of omega-6 to omega-3 (a good ratio would be about a 4:1, most Americans are at ~ 12:1), one is combating the inflammation that causes the above diseases. That is why omega-3 is also used to treat other inflammatory diseases like arthritis.

So I guess the bottom line is this: Seth is not taking a very good source of omega-3s and reducing inflammation doesn't really play a large role in balance, therefore, flaxseed oil is not giving him the benefit – it's science ?



Perhaps you're still getting too much omega-6.

Michael Pollan had a piece in the Times Magazine a few weeks back. His executive summary? "Eat food. Not too much. Mostly plants."


... or maybe he's just getting better because he's practicing every day.


PollyQ is today's winner.


Thanks, Steve, for writing about this. Here's why I think the balance improvements I've noticed are unlikely to be due to expectations:

1. I first noticed the effect putting on my shoes the morning after I started taking flaxseed oil. I had been putting on my shoes standing up for two years; until that morning, I had always had trouble. Every morning. (I had expected it to get much easier -- practice effect -- but it didn't.) The sudden improvement was a complete surprise. I had never heard of such an effect. I had hoped that flaxseed oil would improve my sleep.

2. The sudden improvement I saw when I switched from 2 tablespoons/day to 3 tablespoons/day was also a surprise, although I realize this may be harder to believe.

3. When I switched from flaxseed oil and walnut oil to sesame oil, I expected my balance to get worse. It did, but not when I expected. (It took 2 days to see a change; I expected to see it on the first day.)

Which is not to say I'm sure. If the effects I've seen are repeatable, I'll test myself not knowing what oil I've ingested.



Well, if it was just his desire for better balance that explains the results, why wouldn't any of the other substances he's taken have had the same effect? Presumably he hasn't gotten on his balance-o-meter only after the omega-3; I'd guess that he's tried this experiment many times before. I suppose he could have had extra enthusiasm for omega-3, or suspected that it would aid balance in a way other things he's taken wouldn't, but I doubt it.


Each of us has a different biochemistry so it only makes sense that each of us might be affected differently by various biochemicals. Perhaps Seth Roberts was low in Omega-3, and therefore got an immediate boost in balance. That doesn't mean that I will.
In fact, the real lesson from Seth Roberts is not what works for him, but his methodology to find out what works for each of us.


This is my first time posting, but I'm studying to become a registered dietitian and I enjoy economics, so this post caught my eye. Obviously, Seth's conclusion has absolutely no data to back it up. It seems feasible because ...well it seems to make sense, but I could use the same logic in another scenario. For example, for my entire adult life I've been wearing deodorant every day, and I've never been attacked by a lion; therefore, my deodorant must also act as a lion repellant. Now as an experiment, I will not wear deodorant tomorrow, and hopefully I won't get mauled by a lion, but if I do, then we all know that my deodorant is a lion repellant also, or maybe it would be the ultimate in absurd coincidences. Seth probably set an all time best after using the Omega-3 because he has been practicing on his balance-o-meter. This is my long-winded way to say that I agree with PollyQ.


One thing that Seth Roberts does which is very important, is he logs every detail about what he is measuring. When you say you are not experiencing any benefit, you mean you are not *aware* of any benefit. Improved brain function (or prevention of loss) is hard to measure anyway. So, you should have had taken a memory test before and after and recorded the results. This would help you see if you are doing better.
Study after study has found that self-reporting is amazingly unreliable (though not as bad as gamblers reporting their winnings/losses ;-)
Also, as mentioned above, omega-3 to omega-6 balance is important, as is the type of omega-3. It is still not fully clear, but the short vs. long strand omega-3s may have quite different effects. Fish oil caps provide a completely different blend than veg oils. See for details of different sources.
But, also be aware that your ratio of omega-3 to omega-6 will be a bigger indicator than anything else (and be sure not to go overboard with omega-3, as that has known problems too).



just for clarification, self-reporting is not always unreliable. Self-reported health has proven to be a quite reliable predictor of mortality (although not across different cultural groups).


Steven's point about it not mattering whether it's placebo or substance that's causing the better balance is the most interesting aspect I think.

We know placebo has effects in every therapy ranging from 20% on the low end to 40% on the high end. For many treatments, including many types of drug treatment, we'd be shouting from the rooftops if double-blind controlled experiments showed the treatment itself was that effective.

So given a situation in which placebo can account for the majority of the measured response, we should be looking to find treatments which have zero or close to zero negative side effects, even if they can't be shown to have *any* positive effects by themselves.

Chemotherapy, for instance, has horrible side effects, and in some cases it's not clear that the chemo has any positive benefit not attributable to placebo. Some researchers now believe that most current applications of chemo are actually accelerating the progression of metastatic cancers that they are trying to eradicate. Which leads one to conclude that perhaps placebo is the only thing saving some patients from the double whammy of cancer and chemo.

In medicine shouldn't we be going at it differently than we are today? Shouldn't we be assuming a 30% placebo and considering any trials that show less than that to be net harmful to the patient? And shouldn't we be looking for treatments that show at least 40% effectiveness with zero negative side effects? What am I missing here?



Flaxseed? I'd been entertaining a wonderful idea that the powers that be had put fish extract into bread, milk, etc...


Not sure about the flaxseed...some studies have shown flaxseed increases a man's risk of prostrate cancer.

I stick with the fish oils, I leave the flaxseed for my wife.


M.C., although I agree with you regarding Omega-3, I think your question regarding what Seth is doing deserves a reply.
It is "Science" as long as he is clear about his method and model (it is not double or even single blind, and the sample size is 1) and how one can treat the outcome. Since many scientific ideas come from data mining, small sample results, noted (side) effects, and unexpected results, there is nothing wrong with what he is doing as a 1st step (many grant applications are based on such results). He is not suggesting that his results are then validated or applicable to a large data set, but he is using self-experimentation as a way to formulate and refine ideas well in advance of postulating them as theories; many ideas are never explored because of the high cost of proper full scale experiments/trials. He uses rigorous recording/logging and plots his results (and applies regression analysis on them to try to catch personal observation bias), he is then using results to guide further experiments. In some cases, he has others try to replicate his results. The outcome of those "trials" (also not even single blind) is used to further guide the analysis (or perhaps show that he experienced localized effects only). In some cases, this allows him to correlate his results to theories already in play (and based on other kinds of standard experimentation). His diet model is correlated back to animal studies and connected to the set-point theory of weight regulation.
My guess is that because he noticed this unexpected result from a diet change, he is trying to explain it. Whether his results are reproduced by larger groups is unclear, and he may find at the end of his experiments that some other factor (or no particular one) is responsible. But, if he feels he is getting strongly correlated results, then others may choose to try to reproduce using proper double-blind tests.
Much of science historically has operated this way. We are more reticent about admitting that nowadays due to funding issues and politics.



Pkimelma, you are correct about the methodology in determining what is and what isn't a good design for a study. It is true that this is not even a single blind study and we are all aware that it would take several randomized double blind clinical studies to make these results fact. It would be good, as you pointed out, to have a few small pilot studies to determine if this should even be pursued in light of saving time and money (especially, as you noted, since grant money is rather hard to come by these days).

I must point out to you though, that my original post did not address study methodology at all. I was speaking on a purely molecular analysis level. I am guessing that we are two different types of scientists. I am a molecular biologist; I would guess that you have more of a clinical type roll. All this to say that I was not trying to address the relevancy of the study, but the plausibility of omega-3 fatty acids being able to deliver a balance benefit at the molecular level. Now, if balance does indeed have something to do with inflammatory regulation, then by all means run the pilots (I would suggest doing inflammatory cytokine profiles on people to see if they were down-regulated in people who have good balance and up-regulated in those who don't) and follow them with a randomized double blind clinical. However, if at a molecular level, there is no way that omega-3s could deliver this benefit, then do not even waste your money and time with any study.



Hi MC. Thanks for the clarification. I do admit that I am dubious about the effect of Omega-3 or Omega-3/Omega-6 ratio in balance. But, I also note that balance degenerates with age, as well as being impacted by a number of brain disorders, most notably Encephalitis, Stroke, MS, and Parkinsons. Note that the former 2 are very much inflammation related. I do not think much is understood about how age degrades balance, but I found an older study connecting it to ischemia on one hand and inflammation on the other. So, there could be a connection - does not mean there is of course.
So, I agree that one would want to understand what causal connection *could* exist at the molecular/physiological level before pouring money into this. My point was that his approach is a valid 1st step. It is unlikely anyone would pursue that avenue on molecular grounds alone (without some insight to suggest further research), so he is *perhaps* providing grounds for such research.