3 replies to this topic

[timar]

Recently I wrote some posts about the - in my opinion - remarkable benefits the often ridiculed multivitamin supplements have shown in some high quality studies, particulary with regard to cancer risk. This was before I found the recent report of the U.S. Preventive Services Task Force report on vitamin and mineral supplements published this November and freely available on PubMed. This 186 page magnum opus is the most comprehensive review on the evidence for vitamin and mineral supplementation ever published and although, because of its critical emphasis and negative conclusions regarding most single (high dose) vitamin or mineral supplements, it inevitably has attracted some scorn by the supplement industry as well as the usual, superficial "vitamins are a waste of money" media reports, it actually is a very thourough and nuanced review that carefully summarizes the available evidence from high quality studies. Hence, not surprisingly (well, to me at least ), its conclusions on multivitamins are much more promising than those on single vitamin/mineral supplements.

Please allow me to use this review as an opportunity to further emphasize the point I made and to quote my own posts on this subject which by now lie burried in a rather long-winding and rather digressive topic:

Now what to do, when we have a host of widely inconsistent, low quality studies? We should watch out for at least a single well done, methodologically sound study with sufficient statistical power to draw meaningfull conclusions from that study alone. In this case, we do now have such a study. It is the multivitamin RCT arm of the Physicians' Health Study II. Here we have a large (n=14,641) and at the same time very homogenous population that (by its profession) is exceptionally health conscious and well nourished, physically active and largely abstinent from smoking and exessive alcohol use. Moreover, we have a unequalled medium follow-up of 11.2 years, in which the participants took the multivitamin formula vs placebo. The results: a modest decrease in risk of cancer (HR: 0.92, P=0.04) and a non-significant reduction in all-cause mortality risk (HR: 0.94, P=0.13) - despite that the Centrum Silver formula used has been far from optimal, as it contained 4mg iron and 2mg copper.

If you add to that study the results of a very well done epidemiological study, taking into account the "sick-user bias" which severely distorted previous epidemiological studies on supplements, and this finding of increased telomere length in woman who took multivitamin supplements, I think we can make a pretty strong case for a comprehensive low to moderate dose multivitamin supplement.

PS. Regarding the case for multivitamins I made above: I forgot to add this recent study on multivitamins and breast cancer risk. Also, see the work of Bruce Ames for sound theoretical explanations why multivitamins may protect against cancer and other age-related diseaes and a nutrigenetic rationale for the intake of supra-RDA B vitamins.

We can add another two studies, which have just been published this month. The one is a finding from the above mentioned Pysicians' Health Study II RCT that the multivitamin significantly reduced the risk of cataract, the other an epidemiological finding from the Nurses' Health Study II that multivitamin use is siginificantly associated with lower risk of colorectal cancer. Thus the evidence from high quality studies - RTCs and epidemiological - mounts in favor of Ames' Triage Theory. This is exactly what is to be expected from that theory - a modest but significant reduction of cancer and other age related diseases in well-nourished populations.

I forgot to mention the SU.VI.MAX trial because it is an older trial and didn't test a multivitamin in the strict sense of the term. While it didn't include a "complete" multivitamin formula, it combined some important antioxidant nutrients in RDAish doses (this is what sets it apart from most other trials which included only one or two nutrients in much higher doses and which were mostly disappointing). Therefore, the task force report included it in its analysis of multivitamin trials. Let me quite some excerpts from the report here:

We identified four good-quality RCTs (n=28,607)^49-52 and one good-quality cohort study (n=72,337)⁵³ that evaluated the health effects of a multivitamin supplement (Table 8; Appendix E Tables 1 and 2); studies varied in which nutrients were included in the multivitamin formulation and their dosages (Table 9), as well as in the duration of supplement use and outcomes evaluated. No impact on all-cause mortality was found in any of the three RCTS reporting this outcome: the Supplementation in Vitamins and Mineral Antioxidants Study (SU.VI.MAX),⁴⁹ the Physician's Health Study II (PHS-II),⁵⁰ or the Roche European American Cataract Trial (REACT).⁵¹ In REACT, which included 297 elderly European men and women, there were more deaths reported in the intervention group (n=9) than in the control group (n=3) after 3 years, but this difference was not statistically significant (p=0.07). When these data were pooled, there was no effect on all-cause mortality (unadjusted RR, 0.95 [95% CI, 0.81 to 1.11]), even in a sensitivity analysis with REACT removed (unadjusted RR, 0.95 [95% CI, 0.89 to 1.01]; data not shown).

Note that REACT, while it included several vitamins, was actually was a high-dose trial. When REACT was excluded, the combined effect on all-cause mortality from PHS-II and SU.VI.MAX approached statistical significance. SU.VI.MAX alone found a significant risk reduction in men (RR 0.63 [95% CI, 0.42-0.93]) but not in woman (RR 1.01). The RR in the male-only PHS-II has been 0.94 (P=0.13). This effect is explained to some extend by the significant decrease in cancer incidence observed in PHS-II and again, only in the male group of SU.VI.MAX:

The age-adjusted HR for total cancer was 0.92 (95% CI, 0.86 to 0.998) among 14,641 male physicians older than age 50 years taking Centrum Silver® (Pfizer, Kings Mountain, NC) after 11.2 years of followup.⁵⁰ In a general population of middle-aged adults (n=13,017), SU.VI.MAX indicated no impact on overall cancer from a five-component multivitamin formulation after an initial followup of 7.5 years and posttreatment followup of an additional 5 years.⁵⁴ However, based on a significant sex-by-treatment group interaction (p=0.02), a sex-specific subgroup analysis showed a protective effect on any cancer (RR, 0.69 [95% CI, 0.53 to 0.91]) among men, but not women. [...] When data from the two studies were pooled, the cancer incidence was significantly lower among men in the supplement groups (unadjusted RR, 0.93 [95% CI, 0.87 to 0.99]; I²=0.0%; data not shown).

Finally, from their discussion section:

In general, the results of vitamin supplementation trials have been disappointing at best, despite having a solid mechanistic basis.¹⁴¹ One explanation for this could be that the physiological systems affected by vitamins and other antioxidant supplements are complex, so the effects of superphysiological doses of only one or a few components are generally ineffective or actually do harm.¹⁴² This possibility is compatible with the finding in this review that the best data for any salutary effects was for the two multivitamin trials, which used more physiologic doses of a wider variety of agents.

So far I agree with them, but later they went on to state:

For most of the supplements addressed in this report, we found no evidence of an effect of nutritional doses on CVD, cancer, or mortality in healthy individuals with a generally adequate diet. In most cases there are insufficient data to draw any conclusion. One exception in this report are the two multivitamin studies, SU.VI.MAX and PHS-II, which both found lower overall cancer incidence in men. While these trials were both methodologically rated as good, the lack of effect (and limited data) on women in SU.VI.MAX, the modest effect in men in both trials, and the lack of any effect in either study on CVD makes it difficult to recommend that supplementation with any vitamin will provide protection against CVD or cancer.

OK, let me summarize: we know that multivitamins provide a siginificant reduction in cancer risk (at least for men more than 50 years of age). However, that reduction was only "modest" and anyway, there is no reduction in CVD risk, so they can't be recommended as a protective measure "against CVD or cancer". Huh!? Strange kind of logic, isn't it? Bike helmets protect against skull fracture, but because of their lack of any effect on hip fracture they can't be recommended as a protective measure against skull or hip fracture... Well, maybe I'm beginning to understand why some representatives of the supplement industry implied that the report would be biased against supplements.

When it comes to women though, the prospective studies on the association between multivitamin use and telomere length as well as breast cancer risk mentioned above provide some evidence that multivitamins may provide benefits for woman too.

I'm looking forward to hear your take on the report or on my own summary regarding multivitamins.

Edited by timar, 18 December 2013 - 01:12 AM.

[hav]

Here's the direct quote on their cancer incident observations:

During a median (interquartile range) follow-up of 11.2 (10.7 to 13.3) years, there were 2,669 men with confirmed cancer, including 1,373 cases of prostate cancer and 210 cases of colorectal cancer. Compared with placebo, men taking a daily multivitamin had a statistically significant reduction in the incidence of total cancer (active and placebo multivitamin groups, 17.0 and 18.3 events, respectively, per 1,000 person-years; hazard ratio [ HR ], 0.92; 95% confidence interval [CI], 0.86–0.998; P=0.044). There was no significant effect of a daily multivitamin on prostate cancer (HR, 0.98; 95% CI, 0.88–1.09; P=0.76), colorectal cancer (HR, 0.89; 95% CI, 0.68–1.17; P=0.39), or other site-specific cancers There was a lower risk of cancer mortality that did not reach statistical significance (HR, 0.88; 95% CI, 0.77–1.01; P=0.07). Daily multivitamin use was associated with a reduction in total cancer among 1,312 men with a baseline history of cancer (HR, 0.73; 95% CI, 0.56–0.96; P=0.022), but this did not differ significantly from that among 13,329 men initially free of cancer (HR, 0.94; 95% CI, 0.87–1.02; P=0.15) (P, interaction = 0.07).

Sounds like the beneficial effect on total cancer incidence barely reached statistical significance. While zeroing in on specific cancers or the overall cancer mortality rate yielded no statistically significant effects. Suggesting that the multivitamin mix they tested is a weak cancer deterrent at best.

Also note that their tested mix consisted of Centrum Silver plus some extras: Vitamin C, Vitamin E, and beta-carotene. So I don't see the logic of how these results can be attributed to any of those components individually or multivitamins generally. It only speaks to the combination they tested in the administration schedule they employed which included discontinuance of some of the extras partway through.

On another note, also consider the recruitment criteria for these studies, which intentionally excluded large parts of even the physician population... here's the phase I selection criteria:

In 1981, the PHS team sent out invitation letters, consent forms, and enrollment questionnaires to all 261,248 male physicians between 40 and 84 years of age who lived in the United States and who were registered with the American Medical Association. Almost half responded to the invitation. Of the 59,285 who were willing to participate in the trial, 26,062 were told they could not because they reported a history of myocardial infarction, stroke, or transient ischemic attack; cancer (except non-melanoma skin cancer); current renal or liver disease; peptic ulcer; gout; or contraindication to or current use of either aspirin or beta-carotene.

Phase I was an aspirin/beta-carotene study with 4 groups, one of which was a full-control. Then they whittled down all the above participants, excluding those who developed the original excluded criteria during phase I, to n=14,641 and re-randomized them for phase II. Wonder if they fully tracked what every participant took. Be interesting to see how the full-controls and/or to aspirin-only controls as a subset from phase I did in phase II. As well as the beta-carotene takers from phase I; after all, beta-carotene is the only common ingredient that spans both phases.

Howard

[timar]

I feel honored and humbled that some of the world's leading nutritional scientist actually came to very similar conclusions as I did in this topic in their letter to the Annals regarding the Task Force report and the accompanying editorial:

 

Ann Intern Med. 2014 Jun 3;160(11):80

 

Letter

Enough Is Enough

 

Guallar and colleagues (1) concluded that “the case is closed—supplementing the diet of well-nourished adults with (most) mineral or vitamin supplements has no clear benefit and might even be harmful.” However, they ignored decades of nutrition research and diet monitoring of the U.S. population to reach this misleading conclusion. Although a well-balanced diet is the best way to get one’s essential nutrients (except vitamin D and, for vegans and many older adults, vitamin B12), few persons in the United States follow the Dietary Guidelines for Americans.

Consequently, most persons in the United States are not “well-nourished” and do not meet the Institute of Medicine’s recommendations for the dietary intake of all vitamins and essential minerals. More than 93%, 61%, and approximately 50% of adults in the United States do not get the Estimated Average Requirement of vitamins D and E, magnesium, and vitamin A and calcium, respectively, from their diet, including enriched and fortified foods (2). Further, 98% and 71% do not meet the Adequate Intake of potassium and vitamin K, respectively (2). Many of these percentages are even higher among subpopulations with increased micronutrient needs, including older adults, African Americans, and obese persons.

Conversely, persons taking a daily multivitamin and mineral supplement formulated at approximately the Daily Value do fill many of these nutritional gaps effectively, safely, and at low cost: A high-quality multivitamin and mineral supplement costs as little as 3 cents per day (3), and long-term use is not associated with any adverse health effects (2, 4).

 

Guallar and colleagues state that “β-carotene, vitamin E, and possibly high doses of vitamin A supplements increase mortality” (1). Only approximately 0.1% of adults in the United States exceed the Tolerable Upper Intake Level (UL) of vitamin E because of high-dose supplement use, and approximately 1.1% exceed the UL of vitamin A (2). It is well known that vitamin A should not be consumed long term in amounts exceeding the UL because it may cause hypervitaminosis A and birth defects and that smokers should avoid β-carotene supplements because of an increased risk for lung cancer. The meta-analysis of randomized, controlled trials (RCTs) reporting that high-dose vitamin E supplements increase mortality (1) has been refuted by several, more comprehensive meta-analyses, such as that of Abner and associates (5).

The known biological functions of micronutrients are to maintain normal cell and tissue function, metabolism, growth, and development by serving as essential cofactors or structural components of thousands of enzymes and other biomolecules, among other means. For example, vitamins A and D, iron, and zinc play critical roles in innate and adaptive immunity and folate is required for normal neurologic development. A multivitamin and mineral supplement containing folic acid dramatically decreases the risk for neural tube defects and is recommended for women of childbearing age.

Multivitamin and mineral supplements also may help decrease risk for chronic disease. The largest and longest RCT of a multivitamin and mineral supplement conducted to date, the PHS II (Physicians’ Health Study II), found a statistically significant 8% reduction in total cancer incidence in male physicians (12% when excluding prostate cancer) and a statistically significant 9% and 13% reduction in total and nuclear cataract, respectively (4). These findings are consistent with those of several other RCTs and are even more impressive given that conventional RCT designs have limited ability to reveal benefits of nutrients—in contrast to drugs—for chronic diseases (6). Therefore, taking a daily multivitamin and mineral supplement not only helps fill known nutritional gaps in the diet of most persons in the United States (thereby assuring normal body function and supporting good health) but also may have the added benefit of helping to reduce the risk for some chronic diseases.

To call the case closed; deny the value of further research; and label multivitamin and mineral supplements useless, harmful, and a waste of money (1) is wrong, is not based on the established science for their primary indication, and misinforms the public and the medical community.

 

Balz Frei, PhD

Linus Pauling Institute, Oregon State University; Corvallis, Oregon

 

Bruce N. Ames, PhD

Children's Hospital Oakland Research Institute; Oakland, California

 

Jeffrey B. Blumberg, PhD

Friedman School of Nutrition Science and Policy, Tufts University; Boston, Massachusetts

 

Walter C. Willett, MD, DrPH

Harvard School of Public Health; Boston, Massachuse

 

References

 

1. Guallar E, Stranges S, Mulrow C, Appel LJ, Miller ER 3rd. Enough is enough: Stop wasting money on vitamin and mineral supplements [Editorial]. Ann Intern Med. 2013;159:850-1. [PMID: 24490268]

2. Fulgoni VL 3rd, Keast DR, Bailey RL, Dwyer J. Foods, fortificants, and supplements: Where do Americans get their nutrients? J Nutr. 2011;141: 1847-54. [PMID: 21865568]

3. ConsumerLab.com. 2014. Accessed at https://www.consumer.../multivitamins/ on April 27, 2014

4. Gaziano JM, Sesso HD, Christen WG, Bubes V, Smith JP, MacFadyen J, et al. Multivitamins in the prevention of cancer in men: the Physicians' Health Study II randomized controlled trial. JAMA. 2012;308:1871-80. [PMID: 23162860]

5. Abner EL, Schmitt FA, Mendiondo MS, Marcum JL, Kryscio RJ. Vitamin E and all-cause mortality: a meta-analysis. Curr Aging Sci. 2011;4: 158-70 [PMID: 21235492]

6. Blumberg J, Heaney RP, Huncharek M, Scholl T, Stampfer M, Vieth R, et al. Evidence-based criteria in the nutritional context. Nutr Rev. 2010;68:478-84. [PMID: 20646225]

 

(This letter was made available publicly by the authors on several websites, so I think it is OK to post it here in full length).

 

Rhonda Patrick who runs the website foundmyfitness.com, and works at Bruce Ames' lab, apparently was so infuriated by that editoral that she made a video debunking it. I agree with her analysis:

 

Edited by timar, 03 September 2014 - 04:01 PM.

[niner]

Excellent debunking, but the lie has already made it half way around the world before the truth got its pants on, to (slightly mis-) quote Mark Twain. How many lazy medical professionals will have heard the headline, think to themselves "just as I suspected..." then never see the debunking? Probably a lot.