Announcing the Quantified Health Prize

by Zvi2 min read2nd Dec 2011144 comments

73

Bounties (closed)
Frontpage
We are giving away a $5000 prize for well-researched, well-reasoned presentations that answer the following question:
What are the best recommendations for what quantities adults (ages 20-60) should take the important dietary minerals in, and what are the costs and benefits of various amounts
Part of the question is figuring out which ones are important. You may exclude any minerals for which an otherwise reasonable diet will always fall into the right range, or any minerals whose effects are relatively trivial. 
If you have an excellent entry, even if you don’t win the grand prize, you can still win one of four additional cash prizes, you’ll be under consideration for a job as a researcher with our company Personalized Medicine, and you’ll get a leg up in the larger contest we plan to run after this one. You also get to help people get better nutrition and stay healthier. 
Proposal:
Most of us spend a good portion of our time and money trying to figure out what would be best for our health, and then trying to implement those findings. We ask ourselves how to eat, how to exercise, what drugs and supplements to take and what treatments to seek, but everywhere we turn we find different opinions. Even if one reads the primary research, one finds studies riddled with problems. Most studies have an agenda to sell a product or prove a pet theory. They are then filtered by publication bias.  When results are presented, many authors use framing to steer us to the conclusions they want us to draw.
We can and must do better.
We hereby challenge this community to do better.  We're always saying how great and effective rationality is. This is our chance to prove it, and  put those skills to the test. These problems badly need proper application of Less Wrong's favorite techniques, from Bayes' Theorem itself to the correction of a whole host of cognitive biases.
This contest is also a pilot for a larger contest; before we go and put a lot more money on the line and ask more questions, we want a chance to work the kinks out. 
Entries are due by the end of day on January 15, 2012. This is a change of the original deadline, but it will not change again and it will be strictly enforced.
Final judgment will be made by Personalized Medicine’s Chief Science Officer, based on finalists chosen by our expert reviewers. If necessary, Peer Review will first be used to reduce the number of entries to a manageable size. 
The contest page can be found  here, and the FAQ can be found  here.

73

141 comments, sorted by Highlighting new comments since Today at 2:37 AM
New Comment
Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings
[-][anonymous]10y 23

Sweet, gwern's modafinil fund was getting kind of low anyway.

7gwern10yI'm flattered (and I also see that I am becoming notorious for modafinil, I hope this doesn't cause me problems later in life!), but I'm not actually sure I want to participate. The deadline is fairly short compared to last time, and dietary minerals [https://en.wikipedia.org/wiki/Dietary_mineral] haven't been a previous focus of mine. When I look at the list in Wikipedia of essentials or 'other' as well, the only ones I've looked into are: 1. iodine Which as far as I know, while very important to human development and especially IQ, must be administered chronically and during said development - and so is fairly useless for 'adults (20-60)'. EDIT: in my further writeup on iodine ( http://www.gwern.net/Nootropics#iodine [http://www.gwern.net/Nootropics#iodine]), I was able to find 2 studies which used iodine on middle schoolers or college-age adults; in neither was any significant effect on IQ noticed. 2. Magnesium Which may be useful in acute doses for adults, but this is based on some highly speculative small mouse studies on magnesium l-threonate [https://en.wikipedia.org/wiki/Magnesium_L-threonate], and so any material on it would be awfully short. 3. Potassium User:Kevin [http://lesswrong.com/lw/7wh/rationality_drugs/4xr6] claims potassium in various forms has helped him a lot. This has a surface plausibility because of potassium's role in neurons, but is as or more speculative than the magnesium. I suppose I could package up the research on those 3 and hope for one of the smaller $500 prizes if it looks like, as last time, there will be few participants.
9Zvi10yWould you be more inclined to participate on an extended deadline?

For what little it's worth (since I'm not a potential participant here), the deadline struck me as a short one. Participants are going to do this in their spare time, and it's a busy time of year (holiday/family obligations for many people, and end-of-academic-term obligations for students and academics).

8gwern10yProbably, but with diminishing returns - an additional month is worth almost as much as 2 months. More important to me is how many people would be participating and what I could dig up on any minerals.
4Curiouskid10yNice pun.
7lukeprog10yI was going to make a comment about how gwern should do this, but I see you beat me to it!

DEADLINE CHANGE: We have decided based on feedback to extend the deadline to January 15, 2012, to allow people time to find the contest or take care of finals and holiday obligations, but it will not change again and will be strictly enforced.

Ohh, this is exactly my field of study (currently). I may have to consider a submission. A good excuse to motivate me to consolidate my research anyhow!

3Kevin10yAny suggestions on a useful textbook?
3Kevin10yhttps://www.metametrix.com/learning-center/books/2008/leifm [https://www.metametrix.com/learning-center/books/2008/leifm] is the best I found in a pretty comprehensive search. Functional Medicine is weird. It seems to be spun in all kinds of different ways by alternative medicine practicioners, but at the ivory tower level of Functional Medicine, it seems engrained in proper Bayesian paradigms. See this PDF which goes so far as to include a simple overview of AI techniques! http://www.functionalmedicine.org/ifm_ecommerce/Files/IFM%20TFM%20Chapter%205%20excerpt-128721172845367500.pdf [http://www.functionalmedicine.org/ifm_ecommerce/Files/IFM%20TFM%20Chapter%205%20excerpt-128721172845367500.pdf]
3gwern10yWhich ones have you been studying?

You might want to post this over at the Immortality Institute (http://imminst.org/). They recently created a multivitamin using community input, so all of this has been discussed over and over again there already.

For a second there I thought this was you, personally, wanting an answer to this question, which was funny but not really ridiculous or out of character.

Good luck. Hope this gets good results!

3Zvi10yThank you. I take that as high praise!

The contest asks for "A recommendation list that tells people what they should do based on their situation, without any additional information or explanation. Keep it as short as possible, but no shorter."

Are there limits on the sorts of recommendations that are considered acceptable? For example, could one recommend no mineral supplementation? Alternatively, instead of recommending mineral quantities, could the paper recommend a procedure of personal experimentation saying, "take mineral A and monitor the results with process A', then adjust according to criteria A*, then take mineral B and monitor the results using criteria B'..."

5Zvi10yThere's a reason there was no explicit size requirement on the recommendations. They could be one sentence, or they could be a complicated set of conditional instructions, or anything in between. It's a real life question asking for a real life solution, and time/effort/money required of the patient is one of the trade-offs involved. Will add this to the FAQ in some form, to make sure it is clear.
0[anonymous]10yI asked a substantially similar question to the organizers, and the answer is basically yes. It seems like it should be obvious that the most correct answer for a company called "Personalized Medicine" is not something like "everyone takes the same multi-vitamin!"

If anyone entering this contest wants help from someone who develops and implements Bayesian statistical models for a living, let me know. I won't be entering the contest myself -- I don't have time for the required literature search -- but I can help out with some modeling and Bayesian analysis. One caveat -- my experience is with models of consumer choice used in marketing research, rather than analysis for medical research.

I plan to submit an entry to this. I hope that I don't get steamrolled by someone amazing.

I don't have much prior knowledge in this area. Do you think that will be a problem?

3Zvi10yThat shouldn't be a problem; I expect most who enter to not have much prior knowledge. Prior knowledge is doubtless an advantage, but my estimate is that it is not as much of one as you might think, mainly a way to save time (not that this is unimportant), but that's one of the questions I'm highly curious about.

But I’m just an undergraduate, and I’m not even studying biology. Am I qualified to enter this contest?

Absolutely! What we are looking for is in-depth analysis of the published data, not specialized knowledge of biology.

I imagine that prior knowledge of statistics is what's most helpful here, rather than medicine or biology.

I think the presentation needs to include how to tailor the information for your own personal situation. A person with a mineral deficiency can receive enormous benefits from supplements but a person who does not have such a deficiency may receive no benefit or worse. You don’t need to be a Bayesian to understand that just because a mineral supplement will help a specific individual with a specific individual condition, it does not follow that more of that mineral will provide any benefit for a generic human without that condition or deficiency.

A place... (read more)

and what are the costs and benefits of various amounts?

Does cost just refer to price, or does it include risks?

I'd like to see information on how a person can tell if they're deficient or oversupplied in a nutrient and what to expect if the problem is being remedied. For example, a person might have a blood test which shows a deficiency, but if they trouble absorbing the nutrient, just taking more may not help and (faint memory) they might even be deficient while having what seems to be enough in their blood.

To Less Wrong: what are your priors on nutrition? Not necessarily what you do, but what your prior life experience and research has led you to believe is actually correct.

I'll start:

Paleo diet is great, except some carbs and gluten are fine in moderation except in a minority (though possible a high minority, like 40%) of people with outlying genetic mutations that make them process carbs improperly (I admit my prior on this is partially determined by my current unwillingness to stop eating pizza and sandwiches). Intermittent fasting helps prevent cancer.

If... (read more)

4RichardKennaway10y"Eat food. Not too much. Mostly plants." -- Michael Pollan, In Defense of Food, a book which I have not actually read, but which pretty much describes my attitude to all of the diet fads around (including geek fads like paleo), and my actual eating habits. And paleo is "great"?

Pollan's book is horrible. It is actually against science per se in nutrition, continuously bringing up the supposed holistic irreducibility of diets and emphasizing "common sense", "tradition" and "what our grandparents ate" as primary guidelines. Pollan presents several cherry-picked past mistakes of nutrition science, and from that concludes that nutrition science in general is evil.

I am not fundamentally against heuristics derived from tradition and/or evolution, but Pollan seems to use such heuristics whimsically, mostly to push forward a personal agenda of vegetarianism, organic foods and an extremely warm and fuzzy philosophical baggage of food culture and lifestyle. Also, Pollan's arguments are almost exclusively based on affect (nature = good, artificial = evil, people selling artificial food = monstrous, etc.). Actually, looking a bit into the book to refresh my memories, Pollan doesn't even use traditions to make inferences about foods' healthiness; they are merely convenient sources of positive affect.

0JosephBuchignani10yPaleo is a big tent with many suboptimal stalls. Plant nutrition is more difficult to optimize than animal sources. Plants are not strictly necessary, and become completely optional once comfortable starch intake is achieved. Yes, paleo is great. No, it is not a fad, although it contains various fads. Fad: "a temporary fashion, notion, manner of conduct, etc., especially one followed enthusiastically by a group." Paleo is no more a fad than are herbal remedies. Neither are temporary phenomena.
-2JosephBuchignani10yActually, you can get all vitamins, minerals and micronutrients by eating scallops only. It's a whole animal and mineral rich due to inexhaustibility of ocean water (compared to soil mineral content). But you still need rice and fish protein for nutritional bulk and flavor.

Any citations or links?

0JosephBuchignani10yI went a month eating nothing but boiled rice, scallops and water. It was the highest energy/mood/libido diet I've ever tried, but I couldn't maintain weight because it tasted gross.
6PhilosophyTutor10yMy prior probability that any diet consisting solely of rice, water and any one X is a bad idea is very high. I'd want to see very strong evidence that scallops really do contain everything a human needs to remain healthy.
2JosephBuchignani10yMany of the other extreme elimination diets I tried showed obvious signs of micronutrient deficiency. For example, I tried bread and water, lean meat and water, etc. It's easy to recognize signs of deficiency - fatigue, cravings, etc. But I wouldn't say scallops have everything. I did lose weight. I think you'd have to add in some fish to get a complete diet. Not so much for vitamins or minerals, but for something related to satiety and macronutrient composition. It could be insufficient fats/oils, or maybe you can't get enough protein because it triggers satiety too fast due to impending overdose of some micronutrient. Or maybe it's just the taste. I've been going for many months on a diet of solely rice, fish, scallops, rice and shrimp. Since I'm still healthy and productive, there can't be any short or medium-term deficiency there. In general, however, a meat-only elimination diet works. See the Stefansson trial, and his study from living with Eskimos. Meat with sufficient fat on it is all that is needed to sustain human life, and it can even come from a single animal, as long as that animal isn't being grain fed from mineral depleted soil. Thus it's not so much surprising that rice, water, and any X meat is sufficient for health. Rather, the extreme positive biological reaction to micronutrient overload from high scallops intake is what's surprising.
0Kevin10yWhy do you think it is that supplements are so much more toxic than food? Eat a dozen oysters, you'll get 1200% of the RDA of zinc and feel great. Eat 12 zinc pills, and you'll probably vomit within 5 minutes. What about food buffers the effects of minerals? I've come to understand that the ratios of minerals is of supreme importance and of course food has much more complex ratios of minerals, but I'm sure there is something else going on here.
8wedrifid10yI'd like to test that theory. I'll be back in five minutes.
2Kevin10yfacepalm If I was betting on it, I would have said within 20 minutes and on an empty stomach. You should get nauseous, at least, but the contents of your stomach might be enough to hold it down.

If I was betting on it, I would have said within 20 minutes and on an empty stomach. You should get nauseous, at least, but the contents of your stomach might be enough to hold it down.

I haven't eaten anything in 4 hours. I had 12 pills in one swallow each of which apparently has two and a half times the RDI of zinc for males. I haven't experienced any nausea. If I head eaten a dozen oysters on the other hand I am almost certain I'd have vomited. But that says a lot more about my aversion to oysters than their zinc content.

More generally I can confirm that taking vitamins on an empty stomach most certainly does produce more acute side effects. For example I tried a pharmacological dose of niacin (1g) on an empty stomach and apart from the intense pain and flushing over my entire body I was extremely nauseous and nearly passed out. The same dose with a meal only gives me a mild, almost pleasant tingling sensation - albeit one that lasted longer.

Now... no more zinc for me for a couple of weeks.

2Kevin10ySurprising to me. A single pill of zinc causes nausea in me on an empty stomach. Any chance you were suffering from an acute zinc defiency? Do you disagree with the hypothesis that food is better than supplements? Potassium is probably the best example as a supplement that can be quite dangerous. Doctors will warn you in fear about taking 100% of the RDA of potassium via supplements without using an extended release pill, but would happily encourage you to drink 4 cans of coconut water.
2wedrifid10yI seem to have a fairly strong stomach with respect to anything that doesn't squick me out. No, basically not. I've been supplementing zinc semi-regularly only because my hair tissue sample put my copper levels at somewhat elevated. Zinc helps with that. Until such time as someone provides a supplement with a sufficiently formula including both macro and micronutrients as well as sufficient filler to delay the digestion to desirable rates, yes. But then I suppose you would just call that "artificial food". There are certainly things that really do need to be diluted or, better, taken with food. Given how important potassium balance is in the regulation of cell membranes and whatnot it is not surprising that putting a whole bunch in one place (ie. a tablet) can kill cells! Even 4 cans of water with potassium added would be fine - even though it would probably taste vile.
0Kevin10yHow did you pick a hair testing lab? It does seem like hair testing should be the best way of testing for zinc, but reportedly there are wide variances between results at different labs, such that mainstream medicine seems to mostly discourage hair testing. Or skeptics can tear into it. http://www.quackwatch.com/01QuackeryRelatedTopics/hair.html [http://www.quackwatch.com/01QuackeryRelatedTopics/hair.html]
0wedrifid10yJust via a doctor. I'm not sure how much I trust it but there was little harm in playing along. I was mostly interested in checking for chronic metal poisoning - which I returned rather excessively positive for (Aluminium).
1PhilGoetz10yHow much did it cost, were you able to get insurance to pay for it, and if so, how?
1wedrifid10yCost $100 if I recall. It's one of the few medical services I actually had to pay for here (Australia) - it's not quite mainstream enough to be covered by medicare.
0Douglas_Knight10yDoctors tell you not to take 100% RDA potassium via supplements, but you go around offering people 70% RDA in the form of 1 tsp KCl? Have you seen any adverse effects?
1Apprentice10yMy friend mentioned having a greenish tinge to his urine after taking high dozes of zinc for some time. Do you observe anything like that?
1wedrifid10yGreen? That's fascinating. And a little creepy. I usually expect excessive vitamin consumption to produce urine that looks pretty much like berocca [http://4.bp.blogspot.com/-S0Adoe4t_bw/Tsj0woBTbCI/AAAAAAAAAtE/VzkvarAVMXs/s1600/DSC_0520.JPG] - and give or take some urea probably has just about the same mineral concentration. I shall be on the lookout for greenish tinges.
2dlthomas10yYou wrote elsewhere that increased zinc helps with excess copper. If that's through increased excretion of the copper, it seems like that might contribute...
1Apprentice10yI asked him about this again and apparently you nailed it. He said he'd been worried that he might have excess copper (he felt tired all the time and had some other symptoms - he had been a vegetarian for a long time) and that's why he was taking the zinc to begin with. When he saw the green tinge he figured the treatment was working and apparently his symptoms went away at the same time. There might have been something else going on, I don't know. It seems at least mildly surprising that you could excrete enough copper to visibly affect urine color.
0wedrifid10yInteresting thought. I didn't have the impression that the competition between copper and zinc was quite that vigorous but who knows? I do know that if I had the amount of zinc I had today every day that I'd be guaranteed to wind up with a copper deficiency soon enough (and probably an iron deficiency too) - unless I supplemented with those too to keep a balance.
0Curiouskid10yPictures or it didn't happen.
5wedrifid10yThat's an impressive feat! Unfortunately it tells us only a little about whether scallops give you all the vitamins and minerals you need. From what I understand you will not experience much in the way of scurvy after just one month of being deprived of vitamin C, let alone get a clear picture of the long term results of any lack in the more subtle deficiencies! I'd love to see some links too. I'd perhaps take for granted that scallops could give the basic minerals we require (seawater has most of the same salts that we need) but the vitamins I'd have expected to be a different story!
2JosephBuchignani10yI was driven to it by necessity, I have intraheptic cholestasis. You aren't very familiar with the paleo literature. Conclusive evidence exists that human beings can consume just about any animal monotonically without suffering nutrient deficiency. Exceptions might be extremely simple animals like snails or maybe starfish that don't share enough similarity. But most seafood and land animals will work. The major exception to this rule is that some animals don't have enough fat to sustain life, which leads to protein poisoning. The solution is to either eat more carbs or more fat from another source. Vitamin C deficiency is impossible to contract while eating fresh meat. See polar expeditions for details. This is a problem with the "daily value" theory. Vitamin C is present in meat in far smaller amounts than in plants, but because of greater bioavailability, it is actually a better source than plant sources. As long as the meat is fresh. I may put together all this at some point, with links etc. It's buried in my notes.
0[anonymous]10yhttp://en.wikipedia.org/wiki/Rabbit_starvation [http://en.wikipedia.org/wiki/Rabbit_starvation]
0MixedNuts10yThat sounds extremely expensive. (And I know where to get cheap frozen scallops.) What's the next best cheap diet?
-2JosephBuchignani10yYes, although hunger decreased dramatically, so I didn't eat nearly as much scallops as you would think. Scallops have a major impact on food satiety and cravings, I've found. I suspect we tend to overconsume food to compensate for low density of key micronutrients. My long term stable diet is 1. scallops daily, one package; 2. unlimited white rice; 3. lean fish - cod, perch or pollock; 4. shrimp for flavor/texture. Diluting the scallop content brings down cost. The enhancement effect isn't quite as extreme, but it's still very good. This diet has almost no fat or any other difficult to digest substances due to my intraheptic cholestasis. If you are digestively normal, you could fill out the rest of the diet with any paleo ingredients, as long as you eat scallops daily. I believe this delivers superior performance to the typical paleo grass-fed organ meat route, but I cannot personally test this due to my limitations. I think humans are well adapted to a shoreline diet due to bottleneck event(s) caused by some natural catastrophe that rendered extinct those humans without access to shorelines. If you've watched Survivor, you know that any bipedal idiot can gather shellfish on the beach. Ancient shellfish middens indicate that they were a major food source.
3[anonymous]10yI think you're overreaching the idea of adaptation there. Scallops and other bivalve molluscs contain a lot of dietary protein, and also a lot of nutrients we need; part of the reason we need many of those specific nutrients is that unlike even many of our closest relatives, we can't synthesize them internally. So it shouldn't be surprising that a food rich in those things, with very few "empty" calories and which is not too calorically dense would be beneficial...
-3JosephBuchignani10ySometimes it's better to ask than to assume someone has already exhausted his case. http://eatingoffthefoodgrid.blogspot.com/2009/10/pleistocene-diet.html [http://eatingoffthefoodgrid.blogspot.com/2009/10/pleistocene-diet.html] http://www.ted.com/talks/elaine_morgan_says_we_evolved_from_aquatic_apes.html [http://www.ted.com/talks/elaine_morgan_says_we_evolved_from_aquatic_apes.html]
0[anonymous]10y.
0JosephBuchignani10yYes, unfortunately
4[anonymous]10yI love scallops, but I've never known someone to eat the whole animal. Unless you collect them yourself, I don't think you even could; I've only ever seen the adductor muscles for sale.
3JosephBuchignani10yYes, you're right, I was wrong about that
1Alicorn10yAre there other foods with this property? I hate scallops. (And shrimp.)
2JosephBuchignani10yIn theory, any shellfish should do it. Shrimp don't have this property. Shrimp are scavengers, whereas shellfish filter water. The latter activity is what creates the high mineral content.
1[anonymous]10ydouble posted
-3Alicorn10yClams?
0JosephBuchignani10yYep, should work
-2[anonymous]10yAny bivalve, really -- clams, oysters, mussels.
-9[anonymous]10y

I've posted a critique of the contest's internal assumptions here, along with part of my answer:

http://www.aquatic-ape-diet.com/blog/2011/12/02/critiquing-the-lesswrong-mineral-supplementation-5000-contest/

5Zvi10yThis is a problem with asking practically any question about health, if you have the point of view espoused in your article/website regarding health. Not only will it be the response to any question on nutrition, most questions about other things will get an answer starting with "First, fix your diet!" I know this first hand, since several of my friends have similar points of view. Not that there's anything wrong with that! What you say is a valid hypothesis, and if that is what you believe than that is what you should answer with your own, different kind of research. It will not be automatically dismissed, and if you convince us that you're right, you'll win. I'd be disappointed if no one tried this in some form. Also note that even if everything you think about nutrition is correct, that doesn't mean that you can convince a given patient to follow your advice, so recommendations for the rest of us are still highly useful, and even if modern medicine was hopelessly broken a decision must still be made on what to do.
4JosephBuchignani10y"It will not be automatically dismissed, " I'm more than satisfied with that. And I agree with the rest of your comment. As I said in the article, I think the company is a good idea.

I'm actively interested in optimizing my health, and I take a number of supplements to that end. The survey would seem most interesting if its goal was to find how to optimize your health via supplements. As it turns out, none of the ones I take qualify as "minerals". If it turns out in fact that taking Vitamin XYZ is the single best thing you can do to tweak your diet, then this survey's conclusions, whatever they turn out to be (eg. that Calcium is better than Selenium) will be misleading. Maybe that's the next survey.

FYI, I'm taking: vitamin C, green tea extract, acetyl-l carnitine, vitamin D-3, fish oil, ubiquinol, and alpha lipoic acid. I've stopped taking vitamin E and aspirin.

This concept is fantastic - I'm going to publicize it to some friends.

I'll be doing this - might as well put my notes to use.

Their website does not appear to be working. Is Personalized Medicine defunct?

0Rhwawn9yI think they may be. Weren't they planning on holding some more contests? It's been a while.

Anyone else giving this a serious shot? I'm curious to trade strategies, wondering if mine is the dominant one and this is a competition of research or a competition of strategies.

My strategy: recommending basically no supplemental minerals, except where indicated by testing.

1gwern10yI've given up on it. I'm deeply unimpressed by what mineral research I've looked at - it all seems like my original comment, that they are useful chronically during development, but not acutely after you're a grown adult (except in rare cases of major deficiency like iron deficiency for pica or anemia, which is equally useless).
6Kevin10yRight, the evidence of supplementation isn't great, but I have plenty of cites for dietary mineral intake reducing all or specific cause mortality compared to mineral deficient diets, in adults. As of now my hypothesis is that mineral supplementation is probably good in the case of deficiencies, but still not nearly as good as optimizing diet to fix deficiency, because the human body is a sublimely complex system and that there are many important functional nutrition ratios important beyond the widely accepted ones such as Zinc/Copper, Sodium/Potassium, etc. When we supplement a mineral without taking these unknown ratios into account, we mess things up. Natural foods seem to contain minerals in amounts that are almost never actively harmful. Consider potassium, considered one of the most dangerous supplements by medical science. Compare to coconut water, extremely high in potassium, and quite difficult to overdose on. Right now I'm supplementing with Concentrace, which is distilled minerals from the Great Salt Lake with the sodium removed. Mostly magnesium, which I suspect I am deficient in. I trust this magnesium supplement because it comes with a highly complex natural ratio of other minerals. That I'm not unknowingly displacing my vanadium or something because I'm getting a naturally tiny dose of vanadium (among other things). Consider: Most of the Vitamin E on people's shelves is almost literally poison. Vitamin E supplements may increase all-cause mortality. [http://www.ncbi.nlm.nih.gov/pubmed/15537682] However, the studies showing Vitamin E increasing all cause mortality mostly define Vitamin E as alpha-tocopherol [http://en.wikipedia.org/wiki/Alpha-Tocopherol], but this is incorrect. Vitamin E is eight different compounds [http://en.wikipedia.org/wiki/Vitamin_E], with gamma-tocopherol being the main dietary source. If you take Vitamin E supplements, you should go check the label. If it's only alpha-tocopherol, you should throw it out. Vitamin E supple
1Vladimir_Nesov10yThat current state of medical knowledge doesn't allow to make any well-supported recommendation should be a possible valid conclusion. It has nontrivial predictive power about possible recommendations, namely that any recommendations are expected to be bogus. It suggests that the current best recommendation is to take no dietary mineral supplements.
0Kevin10yI think it strongly suggests that you should try to eat food that maximizes the likelihood that you will test in vague healthy ranges for nutritional diagnostic testing. Of course, evidence on nutritional diagnostic testing is in just as bad of research shape as nutrition itself, but still useful when the goal is actually to improve people's health.
0curiousepic10yYou should turn what you've done into a post at least.
0gwern10yI didn't keep terribly good notes, so I'm not going to bother; people can just see what the contestants produce.

I've been struggling for some time with the issue of how to know what and whom to believe when it comes to lifestyle medicine (the effects of nutrition, supplements, exercise, etc. on health on longevity)... and it has occurred to me that I'm ignoring the elephant in the living room. As I understand it, the one non-obvious lifestyle change for which there appears to be good evidence of a very strong effect on longevity is caloric restriction.

When I first heard about CR over twenty years ago there were already decades of research demonstrating its apparentl... (read more)

0JosephBuchignani10y1. Longevity is not the only factor of interest. 2. There's CR and CR. A paleo lifestyle will greatly increase natural tolerance to fasting, leading to longer periods without meals, up to one day at times. Deliberate CR is something different. 3. CR doesn't show up among blue zones or the world's oldest people. Rather, the opposite - enjoyment of life. 4. I read a chimp study that showed CR chimps lived longer but had terrible quality of life compared to the fat happy sly contented ad libitum eaters. That suggests it's a tradeoff between living longer slowly and living faster richly. 5. Longevity is extremely hard to study in humans and there are many better-established effects on health from altering biological inputs than anything related to longevity. 6. Most importantly, cages and unnatural diets may tend to exaggerate the positive effects of CR on animals. Now interestingly, many humans live in the modern equivalent of cages and eat highly unnatural diets... Therefore, I reject your thesis that rejecting CR and pursuing supplementation is misguided.
6gwern10yOr living at all: http://junkfoodscience.blogspot.com/2009/07/calorie-restrictive-eating-for-longer.html [http://junkfoodscience.blogspot.com/2009/07/calorie-restrictive-eating-for-longer.html]
0ericm10yThe article is not entirely accurate. There are numerous claims that are easily refuted.
2Vaniver10yOkay. Is the central point correct?
0ericm10yThe central point of that review? No, I don't think so. It is apparent from the ongoing rhesus monkey study that the calorie-restricted animals are much healthier and suffer fewer age-related frailties than the ad-libs. Regarding CR showing up among blue zones or among the world's oldest people, I think the Okinawans are a fine example of mild-moderate CR in practice.
0gwern10yThen why wasn't total mortality lower? Were. http://articles.sfgate.com/2004-04-04/news/17420824_1_urasoe-japan-naha-okinawa [http://articles.sfgate.com/2004-04-04/news/17420824_1_urasoe-japan-naha-okinawa] : More recent; "An Age of Centenarians ? Lifelong Learning Policies and Ageing" [http://ghrforum.org/design/ko/pdf/\(TA-4\]PETER_JARVIS.pdf) slide 18
0Vaniver10yWell, it was. 21/38 control animals and 14/38 experimental animals had died when they published in 2009. The statistical significance of that is .16. (Of course, if you extend the study, mortality becomes 100%. Why they measure this as mortality percentages rather than lifespan distributions is beyond me.) [EDIT] It looks like they do actually plan to report lifespan distributions, once all the subjects are dead.
0gwern10ySo it doesn't make even the too-weak significance of p=.05.
2dlthomas10yRight, but I think that that's predominately because the sample size is small. We certainly can't conclude, from this evidence, that it does not affect mortality.
0Vaniver10yRight, but that's always the case- you can never accept the null hypothesis.
3dlthomas10yThat it's always the case doesn't mean it's not relevant. The evidence showed us a small effect that may or may not be a real effect. Gwern demanded to know why there was no effect. These are not the same thing.
0Vaniver10yRight, but that strikes me as a failure of n. (They started this study with 30 monkeys; what the hell were they thinking? They expanded it to 76, but it's still too small.) It is noteworthy that the hazard rate of 3 for age-related deaths becomes only 1.5 when you look at all deaths. But what hazard rate do they expect controls to have relative to CR? If it's around 1.5, then they should have expected from the beginning that even if this study gave the modal result, it would not be statistically significant. For comparison, smoking has a hazard rate in the neighborhood of 2- which this study would have been too small to detect most of the time (as p=.03 for the age-related death hazard rate of 3, just below the .05 level). When you look at the individual age-related causes, the differences are dramatic. 5/38 of the controls were diagnosed with diabetes, and 11/38 were diagnosed as pre-diabetic. 0 of the experimental animals developed diabetes (and going from 40% to 0% is a big jump!). Heart disease and cancer were both reduced by 50%. Age at first age-related diagnosis was significantly later in experimental animals (p=.008). 20% of the control group had been diagnosed with an age-related condition by the time the first experimental subject was diagnosed, and at age 30, when half of the control group was dead, 66% of the experimental group had not been diagnosed with an age-related condition, compared to 23% of the control group.* Even age-related lean muscle mass deterioration was less among the CR group. It would be nice to know about energy levels, fragility, and so on, but from reading the study it seems pretty clear something is better about the CR group than the control group, and reasonable to suspect CR is better overall than neutral or worse. It's actually not even clear the CR group was more susceptible to injury, because the higher rates of non-age death among CR monkeys (9 instead of 7) could just be due to there being more CR monkeys (since more of th
0gwern10yThe age-related diseases was encouraging, yes, but we could've said as much just by pointing to the human study of CRs which has thus found various heart and other improvements. Energy levels and fragility are huge questions, since they are the obvious answer to 'what is the tradeoff for fewer age-related diseases through CR' - they make one fragile or reduce one's quality of life, and additional deaths manifest some other way. We, or at least I, are not interested in CR (as compared to gorging ourselves ad libitum) as a method of reaching our life expectancy of 77 or whatever in good shape; I'm barely even interested in it if it gains me a decade or so. (It's a serious hit to self-control and quality of life, I think.) I'm interested in its potential for serious life-extension and breaking the usual 120-year-old limit to one's lifespan. If it can't deliver that, I will be disappointed.
0Vaniver10yAre there people who hit 120 who were not in good shape at 77? If not, then it seems like improving your status at 77 is a necessary step. Beyond that, if you wouldn't do it for an EV of 1 decade, but would do it because it gives you a chance at hitting 120, I think our priorities are very different. (Indeed, referring to 120 as the 'usual' limit seems odd to me. It's five years longer than any man has survived, and there are only 6 men above 110. I wouldn't call that usual- indeed, the 'usual' limit seems like it would be the median age at death. But I count my remaining lifespan up from 0, which is not a universal approach.)
0gwern10yNecessary, but not sufficient, even if one grants the claim that CR results in an overall improvement (rather than a dramatic improvement on some age-related diseases and unknown compensating penalties). Yes. A treatment that gives another 20 years and breaks the 120 barrier is many times more interesting and full of potential than a treatment that only gives 10 years and respects the old longevity barrier.
0Vaniver10yI agree that 20 more years is more interesting than 10 more years. What I find confusing is the probability on living past 100 without CR that's implied by the longevity barrier being relevant.
-1Vaniver10yHow do you measure health, and what do you mean by "very clear"? I think there's general agreement that CR makes age-related deaths less likely. The concern is that it might do that by making other forms of death more likely. If I'm less likely to develop diabetes but more likely to fall / injure myself when I fall, I'm looking at a careful cost-benefit analysis, not a "woo less diabetes." In that light, the lack of statistical significance for mortality taking all causes into consideration is significant. (I was going to comment on the monkeys living easy compared to humans, but it looks like 2/3 of the control group that have died have done so for age-related reasons, which is the same estimate I've seen for humans.) Overall, it looks like the monkeys benefit from CR, but with the tiny sample it's hard to say how much, and without a discussion of their lifestyle / disease burden it's hard to say how it will generalize to humans living in the wild. [EDIT]I had not read the full study when I wrote this comment. Now that I have, I am no longer worried about the lack of statistical significance. The study was so small that it couldn't have reliably detected a hazard the size of smoking. The arrows from this study all point towards CR being better, but they're very fuzzy arrows, and so we can't draw any firm conclusions.
0ksvanhorn10yThanks for the link. That's information I hadn't encountered before. Sounds like the case for CR isn't as strong as I thought.
2Kevin10yIt also sounds like most or all of the benefits of CR can be had via intermittent fasting.

Is there a way to hide banned comments?

0[anonymous]10yApparently.

Would it be wrong to suppose that the body is more active in metabolism than given credit for? Every chemical induced regularly over a period of time is not guaranteed- but still subject to requiring a higher threshold than was previously necessary. My proposal is not empirical whatsoever, minus the observation of my own habits and metabolism. Given, I am young, and still have a metabolism that acts like ants on a cookie, i still fully support the idea that the body changes over time, and supplements and dietary habits must adapt as well.

Consider that some... (read more)

3Foxy10yI will also put it out there that I was taken to a strange report from a Yahoo news article (The credibility is now dipping below zero, I know) that spoke on a report that showed that pill supplements, in a double blind study over a few decades, shows that people who take vitamins daily die a few years sooner than their au-natural counterparts. Though, it is likely that this data has something to do with their outstanding health conditions. Meaning, the vitamin enthusiasts likely already had a reason to be concerned with their health. Just saying. More supplement for thought.
3Prismattic10yAlternatively, consuming massive doses of fat-soluble vitamins (particularly Vitamin E) is a really bad idea. Can't remember where I read it, but I assume it was the same study, and the multivitamin quantities involved were many times the RDA.
0Foxy10yThis isn't the original article I was talking about, but its something along the same lines, in case someone wants a look. http://www.realnatural.org/2011/10/11/vitamin-study-tells-different-tale/# [http://www.realnatural.org/2011/10/11/vitamin-study-tells-different-tale/#]
0gwern10yAnother fun link on vitamins & anti-oxidants: http://www.sciencebasedmedicine.org/index.php/antioxidants-and-exercise-more-harm-than-good/ [http://www.sciencebasedmedicine.org/index.php/antioxidants-and-exercise-more-harm-than-good/]
2Curiouskid10yupvoted for witticism.

should take the important dietary minerals in

A normal diet provides easily enough calcium to meet most people's needs [EDIT: That is completely wrong, but still works for the purposes of my question]. Is the question asking about how much we should consume in total, or how much of the various minerals we should consume from supplements? In the first case, the answer is "about as much as you are currently consuming", while in the second case, the answer is "probably none."

8wedrifid10yWhile prevalent enough that most consider the whole "your natural diet is about perfect" myth to be mostly harmless or even virtuous I'm going to declare it toxic. It would be completely bizarre if it turned out that our arbitrary, drastically variable diets - which are often produced from heavily farmed soil - just happened to give us the optimal amount of minerals.
5Curiouskid10y"natural" is of course a common semantic stopsign.
5PhilosophyTutor10yIt might turn out that for all or most micronutrients there is no advantage at all in consuming more than a certain minimum requirement, and hence that the "optimal" amount of minerals is anything between the daily minimum and a toxic dose. If that were the case then it would not be bizarre at all if it turned out that most people in the developed world were getting an optimal amount of minerals, given that our dietary staples are fortified with micronutrients for the specific purpose of preventing micronutrient deficiency and that micronutrient deficiency as a public health issue is something we've been aware of and working to reduce for quite a while now. If you eat breakfast cereal with milk in the morning and a sandwich made from commercially-sold bread and meat for lunch I suspect you would have to try very hard to develop any micronutrient deficiency. It would be a lot more interesting if there was some sort of dose/effect curve for these micronutrients where we could find all the maximum values and hit them, as opposed to a boring dose/effect plateau, but just because it would be interesting doesn't make it true.
5[anonymous]10yI didn't research it for more than five minutes, but I suspect someone on such a diet could easily develop a vitamin K deficiency. The DRI for males/females is 120/90 mcg/day, and I found no breakfast cereal, milk or bread that had anything close to that. There are too many lunchmeats to check, but the highest values I saw were in the 3 mcg / 28 oz range.
3PhilosophyTutor10yThis source [http://www.fao.org/docrep/004/Y2809E/y2809e0g.htm#bm16.5] indicates that vitamin k uptake is very poorly understood, but that vitamin k deficiency is rare in practice and that you have to put people on a very low vitamin k diet for an extended period to show any negative effect at all. This would be very weird except that our gut flora is known to produce vitamin k, so we're probably producing a fair chunk of what we need internally. From that article: Based on that article there's certainly grounds for adding some green salad to that sandwich to be on the safe side, but it seems likely that your gut flora would keep you going regardless. I was pretty sure that vitamin k deficiency was practically unknown (and a vitamin k overdose relatively dangerous), and so for that reason making vitamin k supplements available was generally seen as a bad idea, but I learned a bit more about the whys of that because you raised that question, so thank you.
0wedrifid10yOn the plus side you'll probably never need warfrin!
1wedrifid10yFor most part actually being true [http://en.wikipedia.org/wiki/Micronutrient] does very little to convince people that it is true either. Fortunately we are reasonably robust when it comes to adverse conditions. We can survive with sub-optimal nutrition - our perfomance just degrades gracefully. A few IQ points here, some osteoperosis there, maybe some mood imbalance or fatigue. For most it falls short of being a blatantly diagnosable condition but that doesn't make it optimal. We can also get by without exercise while subjecting ourselves to chronic stressors, being sleep deprived, smoking and binge drinking. I don't recommend those either.
2PhilosophyTutor10yThe wikipedia link you provided makes it clear that micronutrient deficiency is a serious problem in the developing world but I could not see any factual support in that link for the claim that micronutrients in general have an interesting dose/response curve, or that micronutrient deficiency in the developed world is any kind of problem for the majority of people. You might want to edit that link to reflect the fact that it does not support the claim "actually being true". There is some expert support [http://www.guardian.co.uk/science/blog/2009/sep/10/optimum-vitamin-dose-supplements] for your view but there is also excellent reason for caution [http://www.skepdic.com/vitacon.html#1] with regard to any claims you might hear about micronutrient supplementation. This is a very popular area with scammers, cranks and the deluded. I'd be interested in seeing any properly blinded and controlled studies that show that micronutrient intake over the daily requirements has meaningful benefits - if I can gain a few IQ points by eating more fruit and vegies I'll take that deal.
3wedrifid10yOn both sides. I'll note that the caution here, as is so often the case [http://lesswrong.com/lw/20i/the_scientists_hammer/], is with respect to people given (or who themselves decide to take) toxic levels of vitamins A and E. Yes, overdosing on most fat soluble vitamins is a terrible idea. Also, don't drink mercury. What are these "daily requirements" that you speak of? The numbers you read on the back of the cereal packet? ... But before I get distracted by that can of worms I'll remind myself that the matter discussed was not regarding higher-than-RDA level vitamin consumption. The claim that Solvent made (then wisely recanted [http://lesswrong.com/lw/8nx/announcing_the_quantified_health_prize/5dp5]) was that we shouldn't consume any minerals from supplements. Your breakfast cereal and meat sandwich [http://lesswrong.com/lw/8nx/announcing_the_quantified_health_prize/5dpi] diet claim was along these lines too. While I do happen to assert that for certain vitamins (most notably D and much of the B group) the RDI is poorly calibrated the potential for improvement there is comparatively small. The low hanging fruit (so to speak) is in correcting the all too common chronic but mild deficiencies that even the RDI can tell you are way off.
2JosephBuchignani10yI have experienced cognitive gains that would almost certainly show up on IQ tests by eating better animal sources of micronutrients. Studies would be great.
2John_Maxwell9yWhat sources specifically?
1Douglas_Knight10yHow big cognitive gains are you talking about? IQ tests have poor test-retest reliability. Have you ever taken an IQ test? eg, SATs. If you did on your prior diet, you could take it again. Do you think that your peak performance has improved or just average?
0JosephBuchignani10yI scored 1560 / 34 on SAT / ACT and 99th percentile on GMAT as well, if I recall correctly. I've never taken an IQ test. I was born in 1984, so by the time I took them the SAT's were less g-loaded. I would say average, peak and trough performance all greatly improved, but I can't quantify it. I felt like a genius, relative to where I had been, and much quicker mentally. I have no way of returning to my previous diet right now, so I can't rigorously test this.
5Solvent10yI have become aware of this while researching. I agree.
6wedrifid10y50 Bayes points!
1Zvi10ySounds like you believe the answer is "about as much as you are currently consuming" for most people, but presumably not for all people...
0Solvent10yOkay, thanks for the clarification.
0Kevin10yDoes your normal diet include cow milk? I don't drink cow milk and assumed I didn't get enough calcium.
1Solvent10yUm, I was actually completely wrong about that. Most people don't get enough calcium. I think I mean some other mineral. Ooops.