Around the Web; The Case of the Killer Vitamins

I’d like to thank Patrick Timpone for a very enjoyable interview on The Morning Show at One Radio Network. Here is the MP3; I’m on for the second half of the show. You can find a zip file at the archive for October 13. Patrick’s producer Sharon tells me that she’s already benefited from our book:

I was following The Primal Diet and since I read the book, I’ve been allowing myself potatoes and rice and doing very very well on them among doing some other things you recommend.

Also, I’d like to thank Jimmy Moore once more for hosting his highly entertaining “safe starch” symposium (Jimmy’s original post; my response, here and at Jimmy’s). It was great to get the opportunity to explain ourselves to so many people in the low-carb and Paleo movements.

Jimmy is planning to try our diet for a week in November, which will be a good occasion for us to publish a 7-day meal plan. We’ll invite anyone who’s curious to try the diet along with Jimmy, and compare notes.

[1] Interesting posts this week:

Angelo Coppola on Latest in Paleo wonders if Denmark’s saturated fat tax will apply to mother’s milk. If so, it’s bad news for unemployed infants! (He also discusses the “safe starch” debate.)

I once knew a French astronomer who died from snorting cocaine while observing at 14,500 feet. Emily Deans makes me wonder:  Did he have Crisco for dinner?

Stan the Heretic offers his mitochondrial dysfunction theory of diabetes. Peter Dobromylskyj and JS Stanton are also developing ideas along this line. Speaking of JS, his post this week has some great photos of Sierra wildflowers and reflections on the state of the Paleo community.

CarbSane partially confirms Dr. Ron Rosedale: eating carbs does raise leptin levels compared to eating fat, but it is a mild rise over an extended period of time, not a “spike.”

Beth Mazur explains why her bathroom door is always closed.

Chris Kresser discusses why chronic illness often generates a form of hypothyroidism, low T3 syndrome.

Joshua Newman knows how to flatter.

How solid is the case against Andrew Wakefield? Autism is certainly characterized by intestinal dysfunction, and Age of Autism notes that distinguished scientists are citing Wakefield’s work.

Richard Nikoley claims he doesn’t know the words to “Kumbayah.”

Seth Roberts points out that the Specific Carbohydrate Diet has been curing Crohn’s for 80 years, but still no clinical trial.

Jamie Scott, That Paleo Horse Doctor, asks: Why do horses get laminitis?

We’ve quoted vegetarian Dr. Michael Greger’s concerns about arsenic in eggs. I’m more concerned about soy protein in eggs.

Following Steve Jobs’s death, Tim asked for an opinion about the unconventional cancer therapies of Dr Mercola’s friend Nicholas Gonzalez. David Gorski, toward the end of a detailed examination of Jobs’s medical condition and treatment, links to his own claim that the Gonzalez protocol is “worse than useless.”

[2] Music to read by:

[3] Cute animal photo:

[4] Notable comments this week:

PeterC’s dad, who has diabetes, is doing well on our diet. Daniel’s stepdad had a similar experience.

Helen informs us that sweet potato intolerance may be due to raffinose.

Mario Iwakura gives us his infectious theory of diabetes. I think a lot of the cases of disrupted glucose regulation, where people get frequent hyperglycemic and hypoglycemic episodes, may be due to occult infections.

Dr Jacquie Kidd (who blogs at has gotten some great advice from Jamie Scott.

Ellen tells us of cases of iodine supplementation controlling diabetes.

Ned is looking for grass-fed cowbells.

[5] Do Vitamins Kill?: An analysis of the Iowa Women’s Health Study came out this week, and it purported to show that nearly all supplements except calcium and vitamin D increased mortality, with iron being the worst. Oskar asked us to look into it, so we did.

The study followed a large number of women in Iowa, and queried them several times about supplement use. In 1986, the baseline, the women had an average age of 62 (range of 55 to 69) and 66% were taking supplements. By 2004, the surviving women had an average age of 82 and 85% were taking supplements.

Here is the data on overall mortality vs supplement use:

“Cases” are instances of someone dying. “HR” or hazard ratio is the likelihood of dying if you supplement divided by the likelihood of dying if you don’t. Note that all the hazard ratio estimates are “adjusted.”

Unadjusted Hazard Ratios

The left columns of the table give us death statistics and allow us to calculate raw hazard ratios, with no adjustment whatsoever. Seven of the supplements have unadjusted HRs below 1.00, eight have unadjusted HRs above 1.00. The 15 HRs average to 1.01. Without copper, which has an unadjusted HR of 1.17, they average to 0.998. In short, death rates among supplementers were almost identical to death rates among non-supplementers.

This is interesting because supplement usage rose rapidly with age. It was 66% at age 62 and 85% at age 82. Supplement users were, on average, older than non-supplement users. But mortality rises rapidly with age. So there should have been a lot more deaths among the supplement users, just because of their more advanced age.

The paper should have, but didn’t, report age-adjusted hazard ratios. Adjusting for age is very important, since mortality depends strongly on age, and so does supplement use. However, it’s obvious what the result of age-only adjustment would have been. Supplement usage would have shown a substantial reduction in the risk of dying.

Hazard Ratios Adjusted for Age and Energy Intake

The least-adjusted hazard ratios reported in the paper are adjusted for age and energy intake.

The energy intake adjustment is disappointing, because energy intake is affected by health: healthier people are more active and eat more, and obese people also eat more. Including indices of health as independent variables in a regression analysis will tend to mask the impact of the supplements on health, creating misleading results.

However, let’s go with what we have. Based on “Age and Energy Adjusted” hazard ratios, supplements generally decrease mortality. Nine of the fifteen supplements decreased mortality, five increased mortality. At the 95% confidence interval, five supplements decreased mortality, only one increased mortality.

Looking at the specific supplements, results are mostly consistent with our book analysis. Let’s start with the five that showed harm:

  • Folic acid and iron – two nutrients we regard as dangerous and recommend not supplementing – both elevate mortality, as we would expect. Iron is particularly harmful, and should generally be avoided by women once they have stopped menstruating.
  • Multivitamins slightly increase mortality, a result that has been found before and that we acknowledge in the book. This is probably due to (a) an excess of folic acid, (b) an excess of iron (if the women are taking iron-containing multis after menopause), (c) an excess of vitamin A (this is no longer the case – multi manufacturers have reduced the A content of vitamins in response to data – but in 1986-2004 most multis contained substantial amounts of A) which is harmful in women with vitamin D and/or K2 deficiencies (both extremely common, and D deficiency in this cohort is supported by the benefits of D and calcium in the study and the northerly latitude of Iowa) or (d) imbalances in other nutrients; for reasons of bulk multis tend to lack certain minerals, notably magnesium and calcium.
  • Vitamin B6 is an anomaly, as we wouldn’t expect B6 to be harmful in moderation. I’m guessing B6 would have been taken to reduce high homocysteine and for this purpose would often have been taken along with folic acid, a harmful supplement. Also, B6 should be balanced by vitamin B12 and biotin, and may not have been. Perhaps people with cancer were unaware that B6 promotes tumor growth; (UPDATE: See comments; I was misremembering studies, B12 and folic acid can promote tumor growth, but in other studies B6 looks protective against cancer) indeed, in the breakdown by cause of death in Table 3, B6 increases cancer mortality by 6%, but CVD mortality by only 1%. (Folic acid and vitamin A were other cancer-promoting supplements.) The harm from B6 was not statistically significant and I wouldn’t read much into it.
  • Copper is another anomalous result, but this was the least popular supplement, taken by only 229 women or 0.59%. Copper’s hazard ratios were dramatically affected by adjustment: in the raw data, mortality is only 17% higher among copper supplementers, but after age and energy adjustment it is 31% higher, and multivariable adjustment increases it substantially again. Clearly the effect of copper is highly sensitive to adjustment factors, indicating that copper was being taken by an unusual population. I think the hazard ratio for copper is impossible to interpret without knowing why these women were supplementing copper. If we knew their situation, there would probably be an appropriate adjustment that would make a huge difference in mortality. I would say the numbers are too small, the population too skewed, and the information too limited to draw any conclusion here.

Overall, I would interpret the nine that showed benefits as being highly supportive of micronutrient supplementation. The fact that vitamin A, vitamin B complex, vitamin C, vitamin D, vitamin E, calcium, magnesium, selenium, and zinc all reduced mortality suggests that a well-formulated multivitamin would likely have reduced mortality.

Hazard Ratios After Multivariable Adjustment

Now, what about the “Multivariable Adjusted” results, which were responsible for the headlines?

We have to keep in mind a famous aphorism from the mathematician John von Neumann:

With four parameters I can fit an elephant, and with five I can make him wiggle his trunk.

The multivariable adjustments use 11 parameters and 16 parameters respectively. Using so many parameters lets the investigators generate whatever results they want.

I don’t think it’s a coincidence that both multivariable adjustments substantially increased the hazard ratio of every single one of the 15 supplements. The 11-variable adjustment increased hazard ratios by an average of 7%, the 16-variable adjustment by an average of 8.2%.

Rest assured, it would have been easy enough to find multivariable adjustments that would have decreased hazard ratios for every single one of the 15 supplements.

I believe it verges on the unethical that the variables chosen include dangerous health conditions: diabetes, high blood pressure, and obesity. These three health conditions just happen to be conditions that are often improved by supplementation.

Anyone familiar with how regression analyses work will immediately recognize the problem. The adjustment variables serve as competing explanations for changes in mortality. If supplementation decreases diabetes, high blood pressure, and obesity, and through these changes decreases mortality, the supplements will not get credit for the mortality reduction; rather the decreased diabetes, blood pressure, and obesity will get the credit.

Imagine we had a magic pill that completely eliminated diabetes, obesity, and high blood pressure, and reduced mortality by 20%, with no negative health effects under any circumstances. But if regression analysis showed that non-diabetic, non-obese, and non-hypertensive people had 25% less mortality, then a multivariable adjusted analysis would show that the magic pill increased mortality. Why? Because the elimination of diabetes, obesity, and hypertension should have decreased mortality by 25% (the regression analysis predicts), but mortality was only decreased 20%, so adjusted for diabetes, obesity, and hypertension the magic pill must be credited with the additional 5% dead. The multivariable adjusted HR for the magic pill becomes 0.8/0.75 = 1.067.

Of course, what ordinary people want to know is: Will this magic pill improve my health? The answer to that would be yes.

What (too many) scientists want to know is: Which methodology for analyzing this magic pill data will get me grant money? That depends on whether the funding authorities are positively or negatively disposed toward the magic pill industry. Once you know that, you search for the 16-variable multivariable regression that generates the hazard ratios the authorities would like to see.

My take? Judging by the data in Table 2 plus corroborating evidence from clinical trials reviewed in our book, I would say that a well-formulated supplement program, begun at age 62, may increase the odds of survival to age 82 by something on the order of 5% to 10%. Perhaps not a magic pill; but worthwhile.

[6] Not the weekly video: An exceptional magic show:

[7] Shou-Ching’s Photo Art:

[8] Weekly video: A new tool for stroke recovery:

Leave a comment ?


  1. Paul, the “Why Vitamins Kill” section of this is amazing … it probably should be its own post! Might be worth tagging this under supplements so folks can find in the future.

  2. Learning things left, right, and center, I am today! That was a specular analysis of the vitamindskillyoudead study. When hearing about it I immediately thought “why can’t we just look at what controlled trials tell us?” but oh no, some people aren’t having any of that. Why if diabetics take magnesium that must mean that magnesium causes diabetes, the statistics said so.

    I also appreciate the links to Carbsane and the article about soy protein in chicken and eggs. That finding echos my experiences. I’m somewhat less allergic to plants that want to kill me now, but I still avoid eating animals fed junk like soy and grains.

    Cheers and have a great weekend.

  3. Apropos Wakefield — sometimes we are right, sometimes we are wrong. Sometimes we tell the truth, sometimes we don’t. Wakefield may now be doing some work that stands up. This doesn’t make his previous work correct, or his conduct towards children in his studies ok.

    Here in New Zealand we are now are having outbreaks of measles in teens and young adults, thanks to the MMR scare whipped up by the media on the basis of Wakefield’s work. My wife and I are old enough to have relatives with permanent severe brain damage from measles-caused encephalitis. Given the mortality rate from measles, and the increasing loss of herd immunity as immunisation rates drop, Wakefield and his enablers deserve harsh scrutiny.

  4. I really like your comment on the vitamin study, especially the part about “competing explanations”. Good work!

    Correction: The Specific Carbohydrate Diet is 80 years old, yes, but it was originally invented to treat “celiac disease”. What that meant in 1930 I am not sure. Maybe it include Crohn’s, maybe not.

  5. Paul, the more read about studies, the more I’m convinced that most are dishonest and worthless. The saw about statistics can now be amended to add studies.

  6. Great post to point friends to, thanks. My widens on her 5th pregnancy, and apparently cannot take enough folic avoid and iron:) fortunately we’ve been primal/phd for a while that everything shows up satisfactory to the health folks.

  7. Hi Beth,

    The post is tagged. Good idea, thanks!

    Hi Stabby,

    Great point about the confusion in interpreting observational studies … I didn’t even bother to get to that issue.

    Hi Steven,

    Thanks for your point of view. I think the pro-Wakefield autism people support immunization, just advocate delaying immunizations until toddlers are slightly older and their immune systems better developed. But I confess I haven’t tried to keep up with the debate.

    Hi Seth,


    It would be interesting to see when the various diagnostic classifications developed. One of my favorite books is The Story of San Michelle by Axel Munthe. He was at one time the youngest man ever to become a doctor in Paris (at age 23 in 1880), and became a fashionable society doctor patronized by European royalty. In his time “colitis” became an extremely fashionable illness, and his most frequent diagnosis. Munthe himself suggested it was as much a psychiatric as digestive condition; he diagnosed it in part because his patients felt better having a diagnosis, and liked having a decisive doctor who could make diagnoses.

    Hi erp,

    Sadly, I don’t do paper analyses often, because it would get depressing.

    Hi Gordon,

    There’s some evidence that mothers become progressively more deficient in nutrients with successive pregnancies, and their babies have poorer health, especially if the pregnancies are closely spaced. Make sure she eats well and supplements!

    Best, Paul

  8. Loved the Benjamin Franklin quote!

    Observational study of one: I just helped a lovely and spunky lady celebrate her 100th birthday. For many of those 100 years, she’s taken a daily multivitamin.

  9. Paul,

    In reference to B6 causing tumor growth.. Do you have a source for that? Is it in the book? I would reference back, but I have lent my copy out.

    This is definitely news to me! Thanks for the info.

  10. Hi Paul – I understand what you’re saying about folic acid, but what about folate supplementation? Is the jury still out on that? I supplement B6, B12 & folate due to moderately high homocysteine. Would you recommend that I continue with the folate or just drop it all together, and stick with just the B6 & B12? Thanks!

  11. Hi Marilyn,

    May she enjoy many more!

    Hi Lindsay,

    I guess I have to retract that.

    I was remembering in vitro studies but upon looking into the literature the in vitro studies seem rather mixed to promising for B6, eg Perhaps I was confusing it with B12.

    In observational studies B6 is strongly preventive of cancers. Eg ovarian cancer:, breast cancer:, colorectal cancer:,, lung cancer:

    Hi Ron,

    Folate is sure to be better than folic acid. I don’t know that there’s enough information to go on to guide you. I guess I’d be willing to take 400 mcg folate, 50 mg B6 and 500 mcg B12.

  12. It would be interesting to see Jimmy Moore follow your plan for more than a week..

  13. Hi Paul,

    The vitamin supplement study is very interesting. I was curious on your take about copper supplementation. As you mentioned, the subject sample size was very small and perhaps these subjects may be a unique subset with other variables playing into the higher mortality rates.

    I have read that copper can act as a pro-oxidant even though it may be used as a cofactor for some of our anti-oxidant enzymes. It seems that a deficiency and getting too much copper can be problematic. I know the UL is 10 mg for an adult. Unless one eats liver or shellfish, most people will never reach the UL. It seems like copper acts as a double edged sword.

    I have been tracking my daily dietary copper consumption and most days I easily meet the DRI of 1.3 mg. I noticed that nuts, seafood and dark chocolate contain a good amount of copper.

    Since my LDL cholesterol has been on the high side lately, I have been taking 2 mg of copper per day from a supplement based on your copper and LDL post. I am also trying out your recommended supplements and eating more safe carbs to see if that lowers my high LDL.

    Do you think 4 mg of copper from the diet and supplementation is too much?

    Here are my other supplement amounts:

    500 mg VitC
    2000 Vit D
    15 mg Zinc
    2 mg Copper
    100 mg Magnesium
    500 micrograms of Iodine from Kelp
    90 micrograms of K2 (MK7)
    100 micrograms of Selenium (brazil nuts)

    Does this look okay?

    I am also adding in more safe starches and fruit. I plan to have another blood lipid test at the end of the month and one in four months from now. I saw my doctor last week and he wanted to put me on statins. I told him that I plan to lower my cholesterol by nutritional means and do not plan on taking them. I am curious to see what happens and will keep you informed.

    Thanks for your time and help.


    • Erik,

      How has your response been to your diet and supplement changes? Notice any impact on LDL?

      Would love to compare notes.


  14. Hi Sarah,

    I would love that of course, but since he thinks starches are deadly it’s great that he’s willing to try for a week.

    Hi Erik,

    We think 2 to 4 mg copper per day from all sources, with say 30 to 50 mg zinc per day, is optimal.

    So 4 is verging on too much. As you’ve noticed, you can easily get into the optimal range with food, especially if you eat beef liver. I personally eat beef liver once a week and so don’t supplement copper at all.

    Are you taking a multi? You might find the “Zinc Balance” formula from Jarrow which has 15 mg zinc and 1 mg copper to be a more moderate way of getting copper.

    Other supplements to consider: vitamin C; epsom salt, for sulfur and more magnesium; more K2 including the MK-4 form; switching to elemental iodine from kelp; switching to supplemental selenium from Brazil nuts (to get better forms than selenomethionine).

  15. Paul, as a life-long hypothyroid, now improving my function through diet, i feel one of the more important nutrients i supplement IS iron. when i miss more than a couple of days, i get more hair loss and find myself unintentionally breathing more deeply, as though my blood weren’t moving enough oxygen. my observations are “validated” through this doctor’s discussion: — any thoughts?


  16. Hi tess,

    Definitely it’s bad to be iron deficient, especially for the thyroid.

    I think the issue is: why are you iron deficient? Do you eat red meat, liver, shellfish, seaweed, spinach, and other iron rich foods? Then ordinarily you should obtain enough.

    Internal bleeding is a possible cause of iron deficiency. Another is inability to absorb iron due to some intestinal dysfunction, like celiac disease. Many pathogens use iron and the immune response sequesters iron, so these can be factors.

    If you are eating sufficient iron, it’s usually worth having a doctor check out possible reasons for iron deficiencies.

    So I would supplement since it’s helping, but also look for causes. Also, be careful: you can pass from deficiency to excess very quickly.

  17. Hi Paul,

    Thank you for the reply. I forgot to mention that I bought centrum silver without iron. Do you think the folic acid and vit A is an issue in the supplement? I consume a lot of sweet potatoes, greens like chard and egg yolks for vitA.

    Oddly, I haven’t been able to find beef liver in our local stores. I would rather consume liver than supplement copper. I purchased the twin labs copper supplement containing 2mg per pill.

    Many thanks,


  18. Hi Erik,

    There’s not a lot of vitamin A in centrum so I’d say it’s desirable to eat liver to get some. Folic acid at 400 mcg I think is tolerable.

    Since grass-fed liver tastes better, you might want to look for local farmers to see if you can get some of their meat. Around Boston, organ meats are very inexpensive from local farmers.

  19. Hi Paul,

    Great stuff as usual. Love the supplement section. I certainly supplement to buoy my health and recommend supplementation to my friends, however I am much more wary about copper. I agree that it is needed however I think some people carry too much. First you have to consider the copper overload disease Wilson’s which is devastating. It presents with neurological, psychiatric and liver disease and diagnosis can take along time. And while Wilson’s disease is rare there is increasing research suggesting copper causes organ damage in diabetics, and been linked to Alzheimer’s and cognitive decline. And take it with a grain of salt but some orthomolecular psychiatrists have linked copper to psychiatric symptoms in the absence of Wilson’s. So I take copper consumption very seriously. I use filters to eliminate elemental copper from my drinking water(piping can be a source, and sometimes an extremely high source of copper), take no supplemental copper, and thus rely on food for my copper intake. I also take zinc to increase expression of metallothionein to bind copper, and other heavy metals. Caveat, though I do not have Wilson’s I had copper overload, so I’ve got a bit of a bias 🙂 And to be careful I watch my copper levels through testing.

    On another note, in the soy link unfortunately the reporter is very very confused. He seems to think soy “protein” and “isoflavones” are interchangeable. Clearly if you read the science described in the article it is isoflavones reaching the yolks. I’d bet money there is no soy protein reaching the yolks. However, I’m more concerned about feminization by these xenoestogens then eating soy protein. So even though the reporter has no clue I now I have to rethink my egg consumption. As some one on a budget yolks have always been a go to source of fat, protein, lecithin etc.


  20. Hi Eric,

    Thanks, great comment.

    Copper is definitely something you want in the right amount. Both deficiency and excess are catastrophic.

    Yet the amounts in our foods are almost uncontrolled. As you say, copper in pipes or cooking vessels can easily create an excess.

    Deficiencies are fairly common among Americans, but maybe we shouldn’t have emphasized that as much in the book as we did, because deficiencies would probably be rare on a Paleo-type diet that emphasizes seafood and liver.

    Thanks for sorting out the egg yolk issue. The low quality of our food supply is frustrating. It’s not clear there’s a good substitute for egg yolks.

    Best, Paul

  21. Paul,
    Thanks for linking to my comment about my stepdad. I made it bc there are so many ppl out there with T2D and his story needs to be told. With all the misinformation being spread around by the profiteers, any exposure to the truth is crucial. In many cases it may mean the difference between life and death. Or thriving and surviving. Bottom line is that everyone deserves to be healthy.

  22. Hello Paul, congratulations for your work on the diet and well advice. I am from Buenos Aires , Argentina. Here the Centrum Silver offered has a slightly different formula. It´s 5000 IU Vitamin A(1000 IU as acetate and the rest as betacarotene) and 3.5 mg of iron.Also the prospect awares not to consume the pills further than two months. I presume this should disuade me from consume it , but maybe the iron and vit A amounts could not be really so much. What do you think?

    Kind regards,


  23. Regarding folic acid:

    1) Can you suggest a good multivitamin that does NOT contain folic acid.

    2) In the absence of eating too much processed wheat product, is the folic acide in a centrum silver (400 mg?) a problem. A quick look at food labels suggest:

    a) processed pirogies (no folic acid)
    b) deccoco linguine (folic supplmentation)
    c) random organic pasta from WF (no folic acid)

    3) in what other foods in folic acid added?

  24. Hi Robert,

    I think the 400 mcg is probably OK because 98% is turned to folate and 400 mcg folate shouldn’t be excessive. Lower might be better for a multi, but I won’t worry much about 400 mcg. You can’t easily find lower, this is the RDA.

    However, you can find “natural foods” multis that provide folate instead of folic acid. We’ll look for higher quality multis when we get time and add some to our recommended supplements page as alternatives.

    I think if you don’t supplement more than 400 mcg and avoid the fortified foods — mostly cereal grains — then you should be OK.

  25. @paul; thanks. Mostly convience for multivitams — both consuming and buying them!

    I love my bread. Go against the grain here, I know. But the folic acid story is a terrible indictement against the SAD and the amount of wheat we eat.

  26. Hi Robert,

    Pure Encapusulations has a multi with Metafolin instead of Folic Acid. Not sure if this is better though.

    Hi Paul,

    If in addition to Brazil Nuts, one were eating liver twice a week, would that be satisfactory in terms of getting the better type of selenium or would you still look at supplements?

  27. Ok, more confusion.

    Is folic acid added to all “bread”, to “wheat” or just “bread in a bag.”

    More product samples:

    Oatmeal (Silver Pallete): none
    Actual box of wheat: none
    cake mix (minimal)

    To me, the easiest step is just add it it the wheat. That means processed foods have it, even if it isn’t listed.

  28. Hi CPM, thanks. We recommend 1/4 to 1/2 lb beef liver a week, that works out to 40 to 80 mcg selenium. I would say that that’s a good dose for that day, but wouldn’t provide so much selenium as to make up for lower doses on other days. If you eat shellfish and other good selenium sources, then you would be OK.

    Hi Robert,

    I believe folic acid is added to flour generally. However, there’s no obligation to report it. Rather, they’re allowed to say the food is “enriched” in folic acid if the content is above a certain level.

    Interesting history of the folic acid issue here:

    The current folic acid recs are all about optimizing the rate of neural tube defects in pregnant women, when folate needs are increased; and the only risks considered were masking B12 deficiency. They were considering reducing the RDA to 180 mcg in the 1990s.

    If you’re a man, the optimal intake is going to be lower than the optimum for pregnant women. However, would have to hunt for a vitamin that was as low as 180 mcg.

  29. Paul,

    *Terrific* as usual.

    I had a quick question related to Vitamin C supplementation. What would the benefit/hazards be of the different types of Vitamin C out there (ascorbic acid, sodium ascorbate, etc.), and delivery systems (powdered, crystalline, etc.)?


  30. Hi Kenny,

    I think they’re equivalent in the body, so the difference would be mainly the effect on the digestive tract. This would only matter at very high doses.

  31. What to Read This Week | - pingback on October 18, 2011 at 8:33 am
  32. Paul, as always, thanks for the link.

    Great example, much like iron, that food supplmentation benefical to one group (pregnant women) isn’t helpful to another.

    And it also shows how much the offical model on how we eat differs from the reality. We do eat to much wheat!

  33. Since I haven’t seen this directly addressed, do you have an opinion on folic acid vs. folate vs. tetrahydrofolate vs. just eating plenty of chicken liver and greens during pregnancy? I’m planning on entering childbearing years soon and (hopefully) staying in that mode for some time…do you think it’d be wise to supplement folic acid long-term or just before conceiving & during early pregnancy?

    I’d be interested in any other pregnancy-specific recommendations you have, though I realize one can’t specialize in everything… 🙂

  34. Hi Amelia,

    The chicken liver and greens strategy is probably the best strategy for folate, but also be sure to get plenty of choline (either by supplementing or eating liver and egg yolks), B6, and maybe some B12 (not too much). The better your choline status, the less folate you need.

    I would be fine with supplementing up to 400 mcg folic acid as in regular multis. I think the 800 mcg in the prenatals is too much. Folate-containing supplements are better than folic acid but might be worse than food.

  35. how do i subscribe to your newsletter/blog?

  36. Hi Gregory,

    Use the RSS “Entries feed”.

  37. Hi Paul,

    What form of B9 would you recommend? Did you look at the form linked in Pure encapsulations multi vitamin?

    Folic acid is given during and before pregnancy mostly, but it also given to people who have Thalassemia (genetic anemia with defective hemoglobins). I was recently diagnosed with Thalassemia minor and the doctor prescribed 2.4mg folic acid per day. Do you think it might be harmful for people that require more folic acid due to the higher red cell turnover? I am worried about the big amount prescribed, yet it is supposed to help with my condition.

  38. Hi celeste,

    I think in your case I would supplement in the amount your doctor recommends, but look to see if natural folate supplements can fit in my budget. Folic acid is much cheaper and that is a consideration.

    While folic acid looks problematic, the problems take 80 years to show up at the few percent level in mortality. The thalassemia iron issues are a much more pressing concern.

    Be sure to take B12, B6, and choline with it. B12 is especially important with folic acid supplementation because folic acid can mask B12 deficiency. Choline may reduce your folic acid needs (maybe 1.6 mg would be sufficient?).

    Here is a review:

    Best, Paul

  39. Paul,

    you’re a true thinker & scientist. i was asked by a physicist friend about the “vitamin kills you” study. so i’m forwarding your article/

    however, you got me worry about B6 & cancer also soy in chicken eggs. (i try to get pasture eggs but it is not always possible or convenient). it is a little depressing.


  40. Hi Pam,

    Well, it seems I was wrong about the B6, and I wouldn’t worry too much about soy in chicken eggs. The chickens have already had one pass at detoxing them, and our gut and liver will take another, so it should be much safer than eating the soy directly.

  41. I was thinking you might want to edit your post to correct the B6 thing because some people may not read all the comments.

    I read somewhere that more than 500mg Vitamin C per day could cause kidney stones. Have you heard of that?

  42. Thanks, syd, I’ve updated it with a correction.

    I did a post on kidney stones: Although vitamin C degradation products contribute to stones, good antioxidant status prevents vitamin C from degrading. So vitamin C itself is kind of mixed in effect, it can help or hurt, but other antioxidants that recycle C will help unambiguously — notably glutathione. So I recommend improving glutathione status, and continuing to supplement C.

  43. Questions about Dietary Supplements | Denver Fitness Journal - pingback on November 27, 2011 at 5:26 pm
  44. Hi Paul,

    Thank you so much for all of this terrific information. I am just starting out on your program and planning to purchase the supplements you recommend.

    I have been using milk of magnesia for the past 2 years as an underarm deodorant. Do you know if I can rely on absorption through the skin as a source of supplemental magnesium?

    I would get my level checked, but I try to avoid doctors . . .

    Many many thanks for all you are doing to promote wellness.


  45. Hi Kay,

    I don’t know what it’s absorption is through the skin, but Wikipedia says absorption is mediocre through the intestine (

    Mg2+ is poorly absorbed from the intestinal tract, so it draws water from the surrounding tissue by osmosis. Not only does this increase in water content soften the feces, it also increases the volume of feces in the intestine (intraluminal volume) which naturally stimulates intestinal motility….
    Only a small amount of the magnesium from milk of magnesia is usually absorbed from a person’s intestine (unless the person is deficient in magnesium).

    I wouldn’t expect it to be better through the skin.

  46. Hi Paul and others,

    Sorry first of all, because I’m going to post about something I already posted about on other threads here… it’s the case of the unresolved dizziness/balance issue (it’s not vertigo). I have had this symptom for probably almost 3 months now. I have seen my doctor and had bloodwork done a few weeks ago and nothing ‘appears’ out of the ordinary. I do have an appt with ear, nose, throat doctor scheduled for second week in March just to rule that out for sure. Other possible causes of course are neuropathy, pinched nerve somewhere, etc…

    Over the period of the last month and a half, I have been eliminating suspect foods one at a time, but have come up with nothing- no culprit found there (potatoes- with their solanine; milk products; eggs; coconut oil, and others). These are the foods that I started eating more of in the past few months, so I thought one of them could be causing a problem. I’m glad they’re not because I love all of those!

    So, my latest epiphany was just this morning. I had just eaten breakfast and taken my vitamins. Shortly after that, the strange sort-of dizzy feeling came on strong (prior to breakfast I had not yet noticed feeling it at all). Suddenly I thought, “the cheap multivitamin!!” that has to be it! So, I looked it up (I don’t want to tell you the brand because I’m ashamed; I already should have known better than to buy a multi at Wal-Mart; nothing against Wal-mart). Of course this is all just speculation, but in looking up some things online, one possible culprit could be the iron in the multi (I am in 30’s, so not postmenopausal). I read that vitamin C helps absorption of iron. Guess what I always take with my multi– 500mg or 1000mg extra vit C. I also read that black tea and phytates in nuts, etc. help block absorption of iron. So, quickly, hoping it wasnt too late (I know I’m quick to jump to a conclusion here) I ate one raw brazil nut, handful each of raw walnuts and raw almonds, hoping to block any excess iron. And guess what I’m drinking? Yep, black tea. I don’t know, maybe I’m silly.

    In quick researching over the past 10minutes, I stumbled across a type of diabetes that I had never heard of (well, not that its a different type, but that its cause is different from probable other causes of diabetes).. it’s called Bronze Diabetes. Has anyone ever heard of this? When I read it, my ears did perk up because I am type 1 diabetic. And I have one older brother who is also type 1 diabetic. Apparently this disorder is related to iron toxicity which can damage organs, and if it damages the pancreas, you can end up with diabetes. I don’t know much else about it yet, still reading. Here are some sites I skimmed over:

    Anyway, anyone have any thoughts or comments on iron toxicity? My hematocrit was within normal range when bloodwork was done a few weeks ago (it was 42 about a month ago and 41.7 about a week after that). Ferretin wasn’t tested– is that what would be a better indicator of iron toxicity?

    I’m going to stop supplementing for at least a week and see if there’s any improvement.
    Just wish I could figure it out…

    Sorry this is so long! I’ll let everyone know when I crack this case!

    Many thanks!! 🙂

  47. Hi KH,

    Terrific detective work! Looks like you should ask your doctor to evaluate you for hemochromatosis or iron excess. Yes, ferritin should be tested. I don’t think hematocrit is an indicator of iron status.

    Best, Paul

  48. Hi Paul,

    Many of us are finding we have genetic snps like MTHFR and MTR, MTRR which supposedly require bio-available forms of folate and b12, respectively. This would be the L-methyltetrahydrofolate form (end product) of “folate” and methylb12. Dr Amy Yasko, a pioneer in nutrigenomics for methylation and developing protocols using supplements to “by-pass) genetic snp polymorphisms recommends about 300mcg of L-methylTHF and about 100mcg of THF (folinic) in supplement form. Some methylation “experts” recommend even higher doses. MTR and MTRR play a role in utilization and recycling of b12. She recommends pretty high doses of hydroxocobalmin and methylcobalmin (depending on COMT status) for these two snp defects, anywhere from I think 2mg to maybe 6mg. Her protocol is much more involved than this. There is speculation that high levels of FA in serum to inability to absorb and utilize this form in the cells, and possibly literature which points to synthetic form of FA blocking receptor sites and “not allowing” the bioactive forms to get into cells and “work” in the methylation cycle. It may be lack of proper methylation that may be linked to (some) cancer. Food sources do contain both the THF and MethylTHF forms. But I have not heard of one methylation “expert” recommending that the folate be obtained from food.. not sure why. There may be a concern for getting enough nutrient, esp with digestion issues being so common? Dr Quadros at Tufts is exploring folate receptor blocking and binding antibodies as CSF deficiencies of folate. The treatment is avoidance of all dairy (makes antibodies go higher) and very high doses of THF (tetrahydrofolate/folinic). I think there may be exploration in using methylTHF and lower doses, but this is not confirmed. If you look at the Deplin site I will link, there is a chart that shows genes that play a role in processes FA. One defective gene (heterozygous (30-60% efficiency or homozygous 10-30% efficiency) could significantly slow the process of breaking down FA, even THF/folinic, into the bioavailable form and contribute to high serum levels of unusable forms: (DHFR, MTHFD1 and MTHFR all play roles along the way)


    • Hi R,

      Thank you for sharing all that information. These are very interesting issues.

      Studies of how supplements interact with SNPs and specific genetic alleles are mostly in a relatively early stage of knowledge, and I don’t think there’s settled knowledge about how to optimally supplement for a given SNP. Keep in mind that the significance of a SNP may be highly dependent on the context of the rest of the genome and the diet and lifestyle, so it is non-trivial to estimate how a SNP may affect nutrient needs. However, we can say in general that common SNPs can’t harm fitness much and so they won’t require heroic supplementation.

      In the case of folate, there is a major interaction with choline status and with B12 and B6. Choline relieves folate deficiency and most people are severely choline deficient. I would tend to recommend improving choline and B12 ahead of folate. It is true that folic acid can inhibit MTHF utilization, so some people should take other forms of folate. PHD foods are relatively folate rich (organ meats, seafood), so food may be a good source even for people with SNPs that increase folate needs.

Leave a Comment

NOTE - You can use these HTML tags and attributes:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

This site uses Akismet to reduce spam. Learn how your comment data is processed.