It's clear to me that there is a great deal of nutritional advice grounded in weak or worthless evidence. A few reasons that the evidence is weak: Long-durations RCTs are very hard to run (they would ideally run for decades, it's generally hard to blind people to what they are eating, it's hard to make people stick for long periods to diets they haven't chosen themselves). Observational studies are badly confounded and you can't control perfectly for all confounders. The usual "Why Most Published Research Findings Are False" problems are exacerbated by the existence of food producers with big budgets and huge incentives to influence the findings.

Is there any trustworthy guide to well-grounded nutritional science that leads to actionable advice? (beyond, don't eat things that are poisonous, try to avoid deficiencies in a range of vitamins and minerals that are frankly difficult to avoid getting enough of if you live in a rich country and eat a vaguely normal diet) Or do I have to sift through it by myself? I've found it hard to find any trustworthy consensus on questions as basic as, is there any health consequence to eating lots of saturated fat, or, does drinking a modest amount of alcohol provide strong health benefits compared to not drinking alcohol?

I thought this piece made a good argument about nutritional pseudoscience:

New Comment
16 comments, sorted by Click to highlight new comments since: Today at 10:27 PM

This is one of those cases where satisficing not optimizing is a rational approach. Eat a variety of not overly processed foods, consume in moderation, avoid overindulging regularly in eating and drinking, pay attention to your personal allergies and sensitivities. There is no good evidence-based nutritional advice beyond this for a good reason: once you satisfice, any extra tinkering is swamped by the noise of factors unrelated to nutrition. Of course, there are always outliers, and if you find or suspect yourself to be one of the rare ones for whom the basics don't cut it, you indeed have to tinker. But in this case asking for general "rationalist guides" on nutrition will not get you what you want.

I'm a little bit surprised by your answer. Do you consider fixing nutritional deficiencies a part of a healthy diet? There is some good evidence that iron deficiency is bad for you here

Definitely worth fixing nutritional deficiencies! The Pareto principle applies: most people will get no nutritional deficiencies by satisficing, not optimizing. As I said, "if you find or suspect yourself to be one of the rare ones for whom the basics don't cut it, you indeed have to tinker".

What do you mean by the most? How likely it is that you have no nutritional deficiencies?

it's not a probability question. Blood tests exist.

Two responses:

  1. It's unclear to me what would make diet advice "rationalist".
  2. The only novel, actionable consensus I've seen seems to be "avoid eating or drinking things with sugar added to them, especially if there's a lot of added sugar".

It's unclear to me what would make diet advice "rationalist".


This probably just means 'put together by members of this community'. Which is reasonable, because this community often does a better job of taking rational, evidence based approaches to things than the world at large.

Nutritionists are not dumb
Let's not be too cynical here. While, yes, nutrition science is short on definite conclusions, it still remains a science. If you want to figure out how to eat healthy, you would find this out the same way you would check whether aspirin prevents cardiovascular disease in certain subgroups or whether paracetamol extends the duration of symptomatic respiratory tract infections.

Step 1: Is there a consensus statement from a reputable professional society? Do different organizations and groups agree? If yes, here is your conclusion. Most sources agree that saturated fat is unhealthy. This is not controversial in nutritional science.

Step 2: Lacking consensus, what do up-to-date reviews and meta-analyses in reputable journals say? Maybe the data is so new that no consensus has emerged or maybe it is controversial for a reason. I find that a good review often presents both sides to the argument. This would be the case with moderate alcohol consumption. Last I checked, there is no consensus and both sides have good arguments.

Step 3: What are the implications if something were true or wrong? How do I balance my own time, money and quality of life against the promise of extended healthspan? Now here you will need a bit of statistical knowledge or intuition as well as a general understanding of biology. In the case of alcohol, given the doses and effect sizes involved, the harm or benefit of either side being correct would be very small.

More importantly, the healthier you are, the less you will benefit from optimizing your diet. Nutrition is an extreme example of diminishing returns. This is because the most important paradox you have never heard of, Taeuber's paradox, clearly shows that any improvement in healthspan (without slowing the aging rate) runs into tremendous diminishing returns.

To be worth your time, promising nutritional interventions above and beyond the basics must have certain properties, i.e. they must slow aging, potentially slow aging, improve non-health related quality of life, or address multiple health-outcomes at once. In this regard, all-cause mortality is the surrogate outcome worth paying the most attention to - although it is still imperfect.

I don't think you're being cynical enough. Nutritional science is a science, technically, I guess, but it's one of the worst in terms of the quality of evidence it's practical to produce.

And there certainy is not as much consensus as you suggest- I don't agree that there's any at all on saturated fats.

Now with epigenetic clocks we can see how dietary modifications impact health more broadly and they do show that nutritionist consensus (something like a mediterranean diet / plant based diet) is in the right direction.  Your skepticism isn't well founded, in my opinion.

I don't believe this is true, at all. I don't believe any part of this is true, actually.

I don't think there is a nutritionist consensus. And I don't think that there is wide agreement, even, about a plant based diet being superior. And I don't think there's any way of reliably demonstrating how dietary modifications impact health. If there were, we wouldn't still need to be doing all these terrible studies in a desperate attempt to know anything at all.

If you can provide support for any of these assertions I'd be interested.

For what it's worth, Gwern says not to worry about it so much.

"(I wouldn’t take ‘health’ too seriously as a criterion. Diet/​nutrition research is one of the worst fields in all medicine. Don’t sacrifice your quality of life now for some small late-late QALYs which may not exist at all.)"

Not sure is they would qualify as "rationalist" but I’m really fond of the Stronger by Science website/podcast

They’re quite knowledgeable and cautious in their advice, and I find their explanations very clear

If you want to go deeper they have a subscription for a very serious monthly research review, and a diet app that seems very carefully designed

[+][comment deleted]1y20

New to LessWrong?