Unhelpful Answers (Ancestral Health Symposium 2013)

At the Ancestral Health Symposium, I went to a talk about food and the brain, a great interest of mine. The speaker said that flaxseed oil was ineffective because only a small fraction (5%) gets converted into DHA — a common claim.

During the question period, I objected.

Seth I found that after I ate some flaxseed oil capsules, my balance improved. Apparently flaxseed oil improved my brain function. This disagrees with what you said.

Speaker Everyone’s different.

A man in the audience said what I observed might have been a placebo effect. I said that couldn’t be true because the effect was a surprise. He disagreed. (The next day, in the lunch line, he spoke to a friend about getting in a kerfuffle with “an emeritus professor who wasn’t used to being disagreed with.”) I spoke to the speaker again:

Seth Is it possible that flaxseed oil is converted to DHA at a higher rate than you said?

Speaker Anything’s possible.

This reminded me of a public lecture by Danny Kahneman at UC Berkeley. During the question period, a man, who appeared to have some kind of impairment, asked a question that was hard to understand. Kahneman gave a very brief answer, something like “No.” 

Afterwards, a woman came over to me. Maybe flaxseed oil reduced inflammation, she said. Given that the brain is very high in omega-3, and so is flaxseed oil, this struck me as unlikely. I said I didn’t like how my question had been answered. I’ve been there, she said. Other members of her family were doctors, she said. She would object to what they said and they would respond in a dismissive way.

The speaker is/was a doctor. Her talk consisted of repeating what she had read, apparently. The possibility that something she read was wrong . . . well, anything’s possible.




15 Replies to “Unhelpful Answers (Ancestral Health Symposium 2013)”

  1. “Everyone’s different.”

    “Anything’s possible.”

    What a convenient philosophy. Mind you, the Second Law of Thermodynamics rules out the second one.

    Seth: I could have said “If everyone’s different, why do you believe what you just told us will help anyone?”

  2. But both could be right. Let’s pretend for a second that the speaker was talking about the results of a carefully done experiment. The outcome of the experiment applies *on average*. We rely on groups being comparable (exchangeable, conditionally independent from potential outcomes, etc) when making inferences about treatment effects. However, person i, let’s call him Seth, may very well react completely different than person j. Yes, anything is possible but the experiment demonstrated that on average some stuff is more possible than other.

    An experiment is not answering the question of what is the causal effect for i or j or any of the persons individually. Treatment heterogeneity is an important aspect of experiments.

    Seth: I wouldn’t say “both could be right”. She said that flaxseed oil was a poor way to get omega-3 to the brain. I found it worked fine. I believe flaxseed oil is better than fish oil because it produces a steadier flow of long-chain omega-3. The short-chain omega-3 in flaxseed oil is slowly converted to long-chain omega-3. whereas the long-chain omega-3 in fish oil produces a spike in long-chain omega-3. That spike might be why fish oil gave me a headache the one time I tried it.

  3. When I started taking flaxseed oil, it gave me headaches for the first couple of weeks. I was almost going to discontinue taking it. The headaches went away, but I still get indigestion from it once every ten days or so.

  4. Seth, I guess you haven’t done the experiment to see if DHA would improve your balance equally much as flaxseed oil? It would be interesting to know. If you could do it without getting headaches.

    Metacomment: I sense that most medical scientists are in denial about how difficult and complex their own fields are. They are deeply invested in there being easily discoverable, clearcut answers, and they greatly underestimate how little they know and how complicated the real answers are. (It’s not hard to understand why a career in science might ultimately make a person delude themselves this way.)

  5. I wonder if you might have gotten better reactions if you used questions instead of claims (both for the speaker as for the placebo-man). To ask them how their explanation fits into your results.

    For the speaker even simple questions like “How have you verified that Flaxseed oil is useless (or is this just a theoretical opinion)?” might work.

  6. fish oil makes my already heavy periods even heavier. for that reason, I cannot take it. ground flax seed does not have that affect on my periods. is this because the concentration of omega 3s in the ground flaxseed that my body absorbs is much less that that of the fish oil? it is a possibility.

  7. I imagine that at least some of the following probably went through the speaker’s head after your question:

    1. How exactly did you conclude your “balance” was better? Is this something you test in some consistent manner every morning? With as nebulous a term as “balance”, an operational definition is a must, particularly when presenting the subject in a scientific forum.

    2. How do you know it was the flax seed oil? Is every day of your life completely identical so that the only variable that changed was the consumption of the oil the night before? In a similar vein, since you had no expectation of any effect on your balance (as per your “refutation” of a placebo effect), I think we can safely conclude this wasn’t a prospective trial, but a retrospective review of your prior day’s activities. Why on earth did you single out the flax seed oil as the cause of your improvement??? This is one of the main reasons retrospective analyses are considered low quality evidence. Your sample size of 1 doesn’t help the case much either I’m afraid. Do you understand the purpose of controlled experiments?

    3. Balance is a widely distributed function and is dependent on the integrity of multiple bodily systems (musculoskeletal, muscle spindle, peripheral nerves, inner ear/semicircular canals, cardiovascular, vestibulocochlear nerve, vestibular nucleus, visual system, etc.), not all of which exist in the brain. Thus, “improved balance” does in no way equate to “improved brain function” (once again, “brain function” is a bit nebulous a term for an academic venue).

    In clinical practice, the most common causes of balance disruption are related to hypovolemia, inner ear disorders, and peripheral neuropathy, none of which are “brain functions.”

    4. Even if your balance did improve, and even if we were to concede (quite charitably!) that it was related to the flax seed oil you consumed the night before, why would we automatically conclude that conversion of flax seed oil to DHA was the responsible mechanism???

    5. Are you high? Did those words really just come out of your mouth?! Who let you in here???

    And on and on and on….Suffice to say that there are many many layers of problems with your statement. And taken apart, the odds that the sentence “my balance was better one morning due to consumption of flax seed oil the night prior thanks to enhanced production of DHA” is true is roughly 86 gazillion to one.

    Hence the response “anything’s possible.”

    I think you owe her a huge thank you.

    Seth: I carefully measured my balance. I carefully confirmed that flaxseed oil was the cause of the improvement. I found that flaxseed oil improved other measures of brain function. As for whether conversion of the omega-3 in flaxseed to DHA was responsible, can you propose a plausible alternative? When your patients or friends or family say something that disagrees with what you believe, how do you respond?

  8. And there has been some discussion about the oxidation potential of some fish oils, pre-capsulation, in the factories, whereas flax seed oil can be ground and consumed quickly under the recipients control.

  9. I used to have horrible earwax problems. My ears would clog up and I couldn’t hear. I read on the internet somewhere that omega-3 could help. I took some capsules of fish oil, and it went to work immediately. I could actually hear my ears unclogging. I wasn’t testing my balance at the time, but I’m certain that if I was, it would have improved dramatically.

    Since then, I lost a lot of weight and I don’t have the earwax problems anymore. I believe this is in part due to reduced inflammation, and eating more nutrient-dense foods.

    Anyway, it seems likely that several of the answers you’ve received might be on to something. It very well could be that the flax seed oil is anti-inflammatory and improved your balance by opening up your ear canals. If that’s true, it’s also possible that it opened up other tubes in your body, such as the blood vessels in your brain. So it might have led to improved brain function, and it might have led to better balance, but the fact that you had better balance might not be due specifically to better brain function.

    I do believe that your general question is valid. There are millions of healthy, intelligent human beings on the planet who have never let an oily fish get anywhere near their mouth. So, while oily fish might be a very concentrated source of omega-3, it’s certainly possible or even likely that the body can get sufficient amounts by other means.

    Personally, I sense that sometimes, your body is better off with a less potent form of a nutrient. Polar Bear liver, for example, is such a potent source of vitamin A that eating it can kill you. The vitamin A found in vegetables, on the other hand, is weaker, less potent. Your body has to convert it to a usable form. So, it probably takes what it actually needs and ditches the rest.

  10. I happened to attend the presentation you are referring to, and heard both your questions and the responses offered. I actually thought that the speaker did an excellent job of responding to your personal experience, given that it seems, at least at first, to disagree with much of what the current published research would suggest about the usefulness of flaxseed as an omega-3 source. The speaker could not discount your personal experience, and was not the one that suggested that your experience may have been placebo-influenced (that was a loudmouthed attendee). The speaker, instead, chose to acknowledge your experience, but was unwilling to pit your personal experience against the majority of the published research in a fight to the death, a decision that I respect and appreciate. As the speaker noted, there can be a large degree of interindividual variability, and your experience does not invalidate the published data, nor vice versa. By phrasing your question in the “is it possible?” format, the speaker had two possible answers: yes or no. If she chose no, she’d be invalidating your perspective and shrugging off n=1 data as well as speaking with an unwarranted scientific hubris, which I suspect we all would be disappointed with as a response. If the speaker said yes, which they did, they were simply acknowledging that your personal experience AND the published data could be correct, as they are not mutually exclusive and may share a common mechanism other than just EPA and DHA concentration at the cell membrane level. I respect the answer supplied, and wonder what possible response would have pleased you more.

    Seth: When I said “is it possible that . . . ” I meant “what about the possibility that . . . ” That the speaker would take the question literally did not occur to me. If the speaker had tried to be helpful in her answer, that would have pleased me more. For example, she could have explained the basis for what she’d said. Why she believes it.

  11. I’ve noticed this effect in many areas of scientific investigation (nutrition, global warming, etc). Perhaps this phenomenon is caused by a natural human trait to strive to understand and categorize something unexplained. Then, once what was a mystery is first categorized by someone (whether that categorization is correct or not), it is no where near as fun to attempt to duplicate/reaffirm/confirm that categorization than it is to come up with a whole new categorization. So, that first categorization is just left unconfirmed and is just used by others in attempting to find other new categorizations.

    Just a crazy thought 🙂

Comments are closed.