Jul 30 2010

Calcium and the Law of Unintended Consequences

Biological systems are extremely complex. This nugget of wisdom may seem trivial but it is a lesson the scientific and medical communities have been learning over and over again for a couple of centuries. Every time we think we understand a biological system we find there is a deeper level of complexity, or another layer of interactions we had not previously taken into consideration.

This is why we need high quality clinical trials to feel confident about the net health effects of any intervention. A treatment may make sense based upon our current understanding of human biology, but that’s not enough. We need to know what it actually does to people.

This is a lesson, however, that the supplement industry has not learned (and does not appear interested in learning). I have nothing against supplements themselves, only how they are regulated and marketed. I prescribe supplements all the time, when they are evidence-based. But the supplement industry has pulled off a marketing coup – the holy grail of marketing: turning their hype into accepted conventional wisdom. For example, the public generally believes that “natural” is always better, that supplements are always safe, and that we all need to take supplements – the more the better. None of these things are true.

A more subtle marketing deception often employed by the supplement industry is the use of pre-clinical basic-science evidence to support clinical claims for their products. In other words, they cite studies in which some marker in a petri dish is increased in the presence of their supplement, then extrapolate several steps to an ultimate clinical application. One of the most common manifestations of this I encounter is looking at some marker of immune function, claiming that it is increased, then extrapolating from this that the supplement “boosts the immune system” and then further extrapolating that it will therefore help the body fight off cancer.

Antioxidants are another example – extrapolating wildly from pure basic science evidence that antioxidants reduce oxidative stress. When clinical studies started coming in showing no benefit the antioxidant sellers claimed that the doses weren’t high enough. Now we have data showing that high doses of antioxidants are associated with increased risk of heart disease. It turns out (gasp) that the biological system was more complex than we though. Oxygen free radicals are also used as signaling chemicals that trigger important cell functions, and suppressing them has – unintended consequences.

Now we potentially have another story to add to the law of unintended consequences – calcium supplements and heart disease. Osteoporosis – the thinning of bones through loss of calcium – is a serious medical problem for the geriatric population. After about age 50 (and post-menopause for women) men and women start to lose calcium from their bones. The best way to prevent this from becoming a problem is to maximize your bone density by age 50, and this is done through weight-bearing exercise. Exercise after age 50 will also help stave off bone loss to a lesser degree.

There is an ongoing controversy about the use of calcium supplements, calcium plus vitamin D, and the use of prescription medications to increase bone density and reduce osteoporosis. This is a complex area of medical evidence, but the quick version is that calcium supplements by themselves are likely not of any significant benefit in preventing and treating osteoporosis and reducing fractures. Calcium plus vitamin D has modest benefit for bone density. In some patients with osteoporosis and not responding to supplements and exercise alone, pharmacological treatment may be necessary to reduce the risk of fractures. It also is likely true that calcium and vitamin D supplements are only helpful if a person is deficient in one or both. If you are getting sufficient calcium and vitamin D from your diet, taking additional supplements is likely of no benefit.

This is a quick summary of a lot of evidence, and there are still many open questions that need to be addressed by further research. Meanwhile the pharmaceutical industry is doing what they can to push prescriptions for osteoporosis and osteopenia (a milder form of bone loss). And the supplement industry is pushing calcium and vitamin D supplements.

In addition to the fact that there is little evidence to support the use of these supplements in most people, there is the question of safety. Again – there is the common assumption that supplements can do no harm, so even if the evidence for efficacy is weak there is nothing to lose from taking supplement for nutritional “insurance.” This is not a good assumption.

A new study published in the BMJ, in fact, suggests that calcium supplements are associated with a 30% relative increased risk of heart attack (but not death, and only a tiny increase in stroke risk). This is a meta-analysis of previous studies, and the authors found:

Calcium supplements (without coadministered vitamin D) are associated with an increased risk of myocardial infarction. As calcium supplements are widely used these modest increases in risk of cardiovascular disease might translate into a large burden of disease in the population. A reassessment of the role of calcium supplements in the management of osteoporosis is warranted.

So calcium supplements are of questionable benefit, and may increase the risk of heart attacks. The authors point out that prior studies did not find the same increased risk from a diet high in calcium, so it only seems to apply to calcium supplements. This is yet more evidence for the general principle, supported by clinical evidence, that it is better to get your nutrition through a healthful varied diet than through supplements.

This study also demonstrates the utility for researchers to keep going back and asking questions about interventions. Biological systems are complex, and we have to look at interventions from multiple angles before we get a handle on their net health effects. And this study is not the final word – it takes many studies over years of research for a consistent picture to emerge. All we can say at any point in time is the best interpretation of existing research. Right now for osteoporosis it seems that the best strategy for bone health is to have a healthful diet with sufficient calcium and to exercise regularly (funny how evidence seems to be coming back to that same conclusion). Regular doctor visits will also determine if this is insufficient and you need further targeted therapy.

27 responses so far

27 Responses to “Calcium and the Law of Unintended Consequences”

  1. Draalon 30 Jul 2010 at 10:06 am

    Did you come across soy supplements (i.e. isoflavones, particularly genistein) in part of your research on this topic?

  2. Draalon 30 Jul 2010 at 10:08 am

    And any of the calcium supplements studied include antacids? I’m wondering if they play a role in increased risk of heart attacks.

  3. ccbowerson 30 Jul 2010 at 10:49 am

    “calcium supplements studied include antacids”

    Tums and similar type antacids are calcium carbonate, which is the exact form of calcium typically used in studies (not that the calcium salt should matter). The study was a meta-analysis of randomized placebo controlled trials so the antacids thing should not be a factor.

    “only a tiny increase in stroke risk”

    I would not conclude any effect on stroke with a p value of 0.11 in a study like this.

    There are more and more studies like this all the time, that indicate some risk to all vitamins, supplements, and medications. It is hard to think of a single intervention that doesn’t carry some risk, and this is an important consideration when making decisions. Too often people take products with the hope that they will be good for their health (in some vague sense), with the assumption that there is no risk of harm. I can’t understand taking any supplement, vitamin or medication without clear evidence that it is helpful in a way that is relevant to the person taking the product. Dr Dean Edell has been promoting this idea for decades now. Why is the idea so hard for people to accept and incorporate into their lives? If nothing else it would save billions of dollars.

  4. ccbowerson 30 Jul 2010 at 11:08 am

    Part of the problem over time is the misapplication of data from studies to the real world. In addition to the idea Steve mentioned above (extrapolating basic research to clinical effects), there is also the flawed logic of extrapolating the results of studies looking at peoples’ diet to people taking supplements. For example a study may find that diets rich in vitamin X may be correlated with a reduction in a certain disease Y, but it does not follow that taking supplement X by itself is helpful for that same condition. In this made up example there is an assumption that vitamin X was the reason those diets were beneficial, when in fact it could have any number of things. In fact this same pattern has happened so many times we should know better by now.

  5. Draalon 30 Jul 2010 at 11:41 am

    “In fact this same pattern has happened so many times we should know better by now.”

    It’s the reductionist approach, trying to find the single variable that is responsible for the positive correlation. I see it at a direct application of the scientific principle. Plausibility is needed to justify the single variable experiments; you can’t just run a bazzillion experiments to test ever single possible variable right off the bat. It’s not practical.

  6. ccbowerson 30 Jul 2010 at 2:00 pm

    “It’s not practical”

    Well you have to understand the limitations of a given study. In these cases the conclusions that were made did not follow the evidence, because large assumptions were made that should have been recognized as assumptions.

    Its one thing to see an interesting correlation, its another to start taking (or prescribing) supplements based upon flimsy evidence. So far very few have really panned out when more rigorous follow up studies have been done. So when a study shows that eating a diet rich in X decreases your risk of condition Y, then it is reasonable to encourage that type of diet. Extrapolating to the specific nutrient X as a supplement (without further study) is not a good bet.

  7. Draalon 30 Jul 2010 at 2:19 pm

    “Extrapolating to the specific nutrient X as a supplement (without further study) is not a good bet.”

    I’m not disagree there. I’m saying that these studies play an important role; they provide a basis for justifying further study.

  8. Draalon 30 Jul 2010 at 2:21 pm

    And after re-reading what you said originally, I misinterpreted that you were discounting the value of these studies, but you were saying that misrepresentation of the data has happened so many times that “we” should know better.

  9. ccbowerson 30 Jul 2010 at 2:45 pm

    Yes, that is what I meant.

  10. Yoinkel Finkelblatton 30 Jul 2010 at 3:25 pm


    Soy sounds bogus as a treatment for Osteoporosis.

  11. SARAon 30 Jul 2010 at 5:10 pm

    “It is hard to think of a single intervention that doesn’t carry some risk, and this is an important consideration when making decisions.”
    This is of course true of all decisions in life, but we tend to ignore that reality.
    It goes to the same problem we all face – too much or too little information, poorly understood context. And humans tend to want quick easy solutions.

    All of it is made worse by marketers who know these human decision makers so well. But I’m not sure we wouldn’t be just as bad without the marketers misdirecting us. Without the ads, we would just rely all the more on Betty Sue at the water cooler.

  12. KGellingon 30 Jul 2010 at 5:27 pm

    “If you are getting sufficient … vitamin D from your diet, taking additional supplements is likely of no benefit.”

    True but glib and unhelpful as most people DO NOT get sufficient vitamin D from their diet (you’d need 2-3 portions of fish every day). “Most people meet their vitamin D needs through exposure to sunlight”, according to the NIH amongst others

  13. ianmcareyon 30 Jul 2010 at 5:46 pm

    I’m curious what we may find out about fish oil supplements in a few years now that they’re all the rage …

  14. BillyJoe7on 30 Jul 2010 at 5:49 pm

    I’m confused here. Do we get vitamin D from our diet or from sunlight. Or both?

    And two questions:

    Apparently calcium supplements can increase heart attacks, whereas calcium-vitamin D seems safe. What about vitamin D supplements on their own?

    Apparently if dietary calcium is insufficient, homeostatic mechanisms keep the blood calcium level normal by removing calcium from bone. So how do you determine if someone is calcium deficent?

  15. ccbowerson 30 Jul 2010 at 5:49 pm

    “Most people meet their vitamin D needs through exposure to sunlight”

    Which is another unintended consequence of the push to avoid sun exposure over the past few decades… vitamin D deficiency.

  16. tmac57on 30 Jul 2010 at 9:01 pm

    Well, as a layperson who follows this kind of thing quite a bit closer that the average person, all I can say is WTF!. Am I the only one who is getting frustrated by the mixed signals about what to put in your body? Just when I think that I have a reasonable grasp of what the story is,a new piece of science comes along and says “not so fast grasshopper”. Just ignore this rant, but do realize that my reaction is way more typical than that of the average educated scientist out there.

  17. Draalon 30 Jul 2010 at 10:16 pm

    Just saw a yogurt commercial touting the benefits of calcium and vitamin D to stop bone loss, both ingredients found in their yogurt. So…are they making a false claim now?

  18. kvsherryon 30 Jul 2010 at 11:20 pm

    What about those of us who are night shifters? Does this mean we should either be gorging ourselves on fish or taking supplements?

  19. ccbowerson 31 Jul 2010 at 1:01 am

    Saying that calcium w/ vit D will stop bone loss is a bit of an oversell, but the claim that they together can slow bone loss is reasonable (Steve links to one study about this above).

    For night shifters, my understanding is that the cause of the increase in health problems is not clear. There are often references to disruptions in circadian rhythms, but night shifters may have other bad habits related to the shift… there are often sleep problems combined with an increased consumption of convenient junk foods. Fish oil supplements have demonstrated some benefits for secondary prevention of cardiovascular disease (or high triglycerides) , but I’m not sure that the evidence is there for primary prevention with shift work. If you like fish, eat it. I think if you have to work overnights, focusing on getting enough good sleep and trying to eat healthy overnight are probably the things to focus on.

  20. Pinkyon 31 Jul 2010 at 9:19 am

    Interesting the way it was reported in Australia: http://www.theage.com.au/national/calcium-heart-risk-link-20100730-10zxa.html

  21. maudson 31 Jul 2010 at 1:22 pm

    Isn’t it ultimately a matter of people not knowing what is best for them, when they think they do?

    Thoughts are usually implanted by something/someone else, so yes marketer’s are evil, in a sense, but they do no worse than any of us; when we give advice, counsel, or order.

    If everyone had a proper diet for their circumstance (read size, locale, dietary restrictions), and had enough exercise, supplements would not be necessary. The issue is that most of us do not, and we know that innately. So we placate our shortcomings with things we think will help; whether that be “natural” supplements, illicit drugs, or anything else that gives us our edge.

    Science has only come so far, and we certainly won’t be finding the fountain of youth any time soon. So we take the next best thing, which is what we perceive to be a step in reducing what ails us. Most of the time it is wasteful, as in we didn’t “need” it to begin with, but sometimes it is something beneficial. Such as the man who never goes outside taking a vitamin D supplement. Or the elderly lady with osteoporosis taking calcium. We don’t know exactly how it helps, but the idea that it is helping is often good enough. Long live the placebo, and also the real.

  22. BillyJoe7on 31 Jul 2010 at 3:37 pm

    From the above link:

    “Endocrinologist Peter Ebeling, who is medical director of Osteoporosis Australia, said the patients in the analysis had total calcium intakes of up to 2400 milligrams a day – far higher than the recommended intake of 1000 to 1300 milligrams.”

    But here is the analysis from the actual study”

    “When the cohort was divided by fifths of dietary calcium intake rounded to the nearest 100 mg/day, the respective hazard ratios (95% confidence intervals) for the effect of calcium treatment on myocardial infarction were 1.18 (0.70 to 2.00) for <500 mg/day, 0.68 (0.39 to 1.18) for 500-699 mg/day, 2.28 (1.26 to 4.15) for 700-899 mg/day, 1.81 (0.97 to 3.41) for 900-1099 mg/day, and 1.41 (0.81 to 2.48) for ≥1100 mg/day;”

    Although the confidence intervals are fairly wide and extend into the insignificant range, the effect is greater at normal dose ranges.

  23. KGellingon 02 Aug 2010 at 8:26 am

    I think it is important to view the results in absolute terms too. Of the 20,000 participants (in the selected studies), the risk of a myocardial infarcation in the placebo group was 24 / 1000 and the risk in the calcium group was 30 / 1000. So there was a 6 / 1000 (0.6%) increase in absolute risk. It’s a significant increase but a lot less than the headline 30%.

    I suspect a incorrect logical leap has been made which is common when basing clinical decision on epidemiological evidence alone – “low calcium levels = take more calcium” (similar to “high cholesterol = eat less cholesterol” or “fat person = low-fat diet”), but the body’s physiology often turns out to be more complicated than that.

    Also, the results say little about calcium + vitamin D supplements (although Steve appears to be implying the opposite). In fact calcium + vitamin D trials because were excluded from the meta-analysis because”vitamin D deficiency has been associated with increased risk of cardiovascular disease and vitamin D supplementation with decreased mortality”.

    High levels of parathyroid hormone (PTH) promotes osteoclasts which strip calcium from bones. Vitamin D increases calcium absorption in the gut and promotes osteoblasts which put calcium on bones. Vitamin D [25(OH)D] levels are inversely correlated to reduced PTH, i.e. vitamin D goes up and PTH goes down. PTH levels off when 25(OH)D levels reach 40ng/ml.

    The International Osteoporosis Foundation recommends older adults need 800-2000 IU/day of vitamin D3 supplementation. They do not recommend calcium supplementation.

  24. Calli Arcaleon 02 Aug 2010 at 11:46 am

    I seem to recall reading about early attempts to treat the sequelae of acid reflux disease (long before the introduction of Zantac and other drugs) by using massive doses of calcium carbonate for a couple of weeks. The idea was to alter the biochemistry of the stomach. And it worked — but several participants died of what amounts to a calcium overdose. I wonder if this finding might be related to those early experiments in treating erosive esophagitis caused by GERD.

    If you are getting sufficient calcium and vitamin D from your diet, taking additional supplements is likely of no benefit.

    Well there’s the rub, isn’t it? What is “sufficient”, and how do we know how much a person is actually consuming and successfully absorbing? How do the various nutrients interact with one another and other things in the food and what form of them is ideal (if indeed it’s only one that’s ideal)? I’m not asking this as a criticism, but more as a general observation. In most cases, we really don’t know the answers to these questions beyond the really obvious. Yet the whole philosophy behind supplementation (especially in the healthy) rests on an unreasonable expectation of precision on our answers to those questions.

  25. mbon 02 Aug 2010 at 6:23 pm


    “I would not conclude any effect on stroke with a p value of 0.11 in a study like this.”

    So, if the hazard ratio remained 1.20 and P=0.04, what would you conclude?

    In this study, is it that there is a real 20% increased risk of stroke and the study is not powered to detect a statistically signficant difference between groups, or, as you imply, there is no risk or one can’t draw conclusions (wasn’t sure which you favored).

  26. mbon 02 Aug 2010 at 6:47 pm


    “I think it is important to view the results in absolute terms too.”

    The authors seemed to have tried to. In the paper, it says:

    “The number needed to treat (NNT) with calcium for five years to cause one incident event was 69 for myocardial infarction, 100 for stroke, 61 for any of myocardial infarction, stroke, or sudden death, and 77 for death.


    “Incorporating the results from the current analysis of studies contributing patient-level data, treatment of 1,000 people with calcium for 5 years would cause an additional 14 MIs, 10 strokes, 13 deaths, and prevent 26 fractures.

  27. BillyJoe7on 03 Aug 2010 at 7:06 am

    Calli Arcale,

    “Well there’s the rub, isn’t it? What is “sufficient””

    I already asked that question:

    “Apparently if dietary calcium is insufficient, homeostatic mechanisms keep the blood calcium level normal by removing calcium from bone. So how do you determine if someone is calcium deficent?”

    I don’t think we are going to get an answer apart from some silent head scratching. 😉

Trackback URI | Comments RSS

Leave a Reply

You must be logged in to post a comment.