LESSWRONG
LW

by ryan_greenblatt
2 min read
A
28

8

In Inadequate Equilibria Eliezer mentions that multivitamins may cause harm via excess manganese

Well, because I looked up previous ketogenic Soylent recipes, and they used standard multivitamin powders containing, e.g., way too much manganese and the wrong form of selenium. (You get all the manganese you need from ordinary drinking water, if it hasn’t been distilled or bottled. Excess amounts may be neurotoxic. One of the leading hypotheses for why multivitamins aren’t found to produce net health improvement, despite having many individual components found to be helpful, is that multivitamins contain 100% of the US RDA of manganese. Similarly, if a multivitamin includes sodium selenite instead of, e.g., se-methyl-selenocysteine, it’s the equivalent of handing you a lump of charcoal and saying, “You’re a carbon-based lifeform; this has carbon in it, right?”)

I started taking multivitamins somewhat recently to get additional iron, vitamin B12, and vitamin D (I'm vegan). Specifically I am taking this multivitamin. I also take creatine. Taking these supplements was partially informed by this post, though I have been taking creatine on and off for quite a while.

I picked multivitamins instead of individual supplements because it was cheaper and more convenient. I also thought that covering additional vitamin/minerals would almost certainly be either neutral or slightly positive. However I now place somewhat more probability in the these additional vitamins/minerals being harmful in net (via excess vitamin A, manganese, or similar). How concerned should I be?

At the moment, I think the additional manganese I am getting from this multivitamin (2.3 mg or 100% DV) is probably not harmful. Precisely, I estimate that the probability of taking the supplement causing modest negative health effects (for instance, reducing expected IQ by >0.5 points or reducing expected life span by >1 month) is 20%. This is primarily based on noting that many common foods have high manganese content and not seeing any arguments against this dosage of manganese after briefly googling. I have seen arguments against higher dosages of manganese in supplements; this may be what Eliezer is discussing.

More generally beyond just manganese, how should the expected value of the 'unknown unknowns' of taking a multivitamin be estimated? It isn't immediately clear to me what the sign would be. I am currently leaning toward slightly negative expected value due to general principles about side effects.

Additionally, are there any other supplement I should be taking? I think my diet is generally healthy; however, I may get too little protein occasionally.

(edit: used to be 'less than 20%', but this wasn't what I meant)

World Modeling
Frontpage

8

Questions about multivitamins, especially manganese
New Answer
New Comment

2 Answers sorted by
top scoring

ChristianKl

50

I started taking multivitamins somewhat recently to get additional iron, vitamin B12, and vitamin D (I'm vegan). 

In men iron more iron is generally associated with worse health outcomes, so why would you do that? (see Tim Ferriss The Four Hour body for more discussion where he recommends blood donations to reduce the iron content of the blood). 

Taking these supplements was partially informed by this post, though I have been taking creatine on and off for quite a while.

The post doesn't mention iron at all, so why?

Generally, we find that in people who eat a normal diet (some poor people don't count in that category) multivitamin supplement don't improve health outcomes. While there's a case for vegan's supplementing B12 (and for everyone who doesn't spend multiple hours outside every day with a lot of uncovered skin surface to take Vitamin D), there's no evidence that broader multivitamins are useful.

Aside from that multivitamins usually don't contain Vitamin D in amounts that the people who recommend Vitamin D usually recommend.

If we assume that some of the vitamins are indeed healthy to supplement that leads to the hypothesis that others are harmful.

One key post that layed out the thesis of chromium/manganese being unhealthy was Most multivitamins have these 2 toxic minerals in them and there's a good chance that's the source for Eliezer believing that those things are unhealthy. 

[-]gilch20

I'm so surprised by the complaint about iodine in that link that it makes it hard to take the rest seriously. Is this guy not a crank?

Reply
4PeterMcCluskey
The iodine comment is confusing, but not clearly wrong. I experienced low thyroid levels that were likely due to excess iodine from kelp powder. But I wouldn't worry about getting too much iodine from a multivitamin. I'm currently taking chromium supplements, due to a test result showing I had below average levels, plus the advice of someone whose opinion I trust a lot more than I trust Louie's opinion.
4ChristianKl
He was at the time one of the MIRI people, one of EY's co-workers so it's reasonable to believe that's where Eliezer's beliefs come from. Of course, the discussion might also happened first and then Louie wrote it up, but that's basically where the belief comes from. I'm uncertain to what extend the claim is actually true. There's some further LessWrong discussion that you could dig up from the achieves. As my memory goes, that further discussion also didn't clearly resolve the question for me.   My personal philosophy with supplements is "Only take supplements when you have a good reason to believe that they are good for you, or you have a way to query your intuition for whether you need it (with Magnesium supplements sometimes I feel like having one and other times I don't and consider that a signal based on which to make decisions)". 
[-]ryan_greenblatt10

The post doesn't mention iron at all, so why?

Iron deficiency is more common without animal sources. Given my diet, I think being iron deficient is considerably more likely than having too much iron. I haven't done a blood test. I also don't have any strong intuition if it is better to have too much or too little iron (my prior would be that too little is worse).

A related question, should manganese rich foods also be avoided? For example, just a few slices of whole wheat bread have a similar manganese content to a typical supplement (100% daily value or ... (read more)

Reply
2ChristianKl
Why do you believe that? 
1mad
You are almost certainly getting enough iron in your diet as a person who (presumably) doesn't menstruate. If you are not feeling fatigued or dizzy on the regular, you almost certainly have enough iron. If you are worried, get a blood test, don't just supplement willy-nilly.  I would recommend you do a food diary for 3 days and enter into chronometer or myfitnsespal. You are probably getting 10-15mg of iron a day and the RDI for adult men is 8mg. Yes, vegans get non-heme iron, but the iron in meat is something like 80% non-heme, so most people actually get a very small amount of heme iron, and diet tends to have no impact on iron intake. I speak generally as I don't know what your  diet looks like: if you eat pizza pockets and ben & jerrys for every meal then you have bigger problems than your multivitamin. I'm a vegan doing a nutrition degree (at an accredited university, not at one of those woo-woo online holistic centres). I also have low iron because I have dysmennorhea, which I assume you can't suffer from.  Definitely take b12 and take D if you don't get sun (you probably don't). Make sure you are getting vegan D if that's important to you, most D is not vegan. Also make sure the D is an appropriately high dose, not the very low doses that are in multivitamins. 

PeterMcCluskey

20

Too much calcium can be harmful, although that one doesn't have much. Harm from calcium can probably be avoided by getting enough vitamin K2 from fermented foods or a supplement. You're probably not getting K2 from the multivitamin.

Potassium deficiencies are common (but likely less so in vegans). Supplements aren't allowed to have much potassium, but it's easy to get potassium chloride and use that in place of salt.

Moderation Log
A
20