Chapter 3 of “How to Live Forever”
In 1996, Larry Clark published the results of a randomized, double-blind, placebo-controlled trial in the Journal of the American Medical Association. He had enrolled 1,312 patients and given half of them 200 micrograms of selenium per day. The trial was stopped before its scheduled completion.
The reason a trial gets stopped early is that the results become so unambiguous that continuing to give the placebo group a placebo is considered unethical. Clark’s trial produced a 50% reduction in total cancer mortality and a 37% reduction in total cancer incidence. Prostate cancer incidence dropped 63%. Colorectal cancer incidence dropped 58%. Lung cancer incidence dropped 46%.
For a single mineral costing a few cents a day.
The medical establishment’s response to this result was, essentially, to ignore it. A decade later, the National Cancer Institute funded the Selenium and Vitamin E Cancer Prevention Trial — SELECT — to replicate Clark’s findings at scale. Life Extension Magazine predicted in 2008 that SELECT would fail. Not because selenium doesn’t work. Because SELECT was designed to use the wrong form of it.
Clark’s trial used selenized yeast, a form of selenium that closely resembles dietary selenium in its biological activity. SELECT used l-selenomethionine, a different chemical form with different bioavailability and different tissue distribution. SELECT also used all-racemic alpha-tocopherol as its vitamin E — a synthetic form that depletes gamma-tocopherol, the naturally occurring form that epidemiological data had already associated with dramatically lower prostate cancer risk. Men supplemented with alpha-tocopherol in SELECT experienced a 45-48% depletion of gamma-tocopherol by six months, sustained for the five-year duration of the trial.
SELECT found no benefit from either nutrient. This was presented to the public as proof that selenium and vitamin E do not prevent prostate cancer. The distinction between the form of selenium that worked and the form that was tested was mentioned nowhere in the press coverage. The 63% reduction in prostate cancer incidence from Clark’s trial was not mentioned either. A study using the wrong molecules produced a negative result that then served to erase a study using the right ones.
This is the pattern. Understanding it is prerequisite to reading the literature.
How to Build a Study That Fails
The pharmaceutical drug trial model was developed to evaluate a specific kind of intervention: a synthetic molecule at a specific dose, administered to a population with a specific condition, over a defined period, compared against a true placebo. This model has produced genuine medical advances and its methodological rigor is real.
It is also almost perfectly designed to produce negative results when applied to vitamins and minerals.
A drug has no baseline level in the human body. A vitamin does. You cannot design a true placebo group for a vitamin D trial because every participant already has some vitamin D in their blood from sun exposure and diet, and that baseline level varies enormously from person to person. The researchers can give placebo pills, but the “control condition” is not absence of the intervention. It is an unknown, varying, unmeasured quantity of the same intervention.
A drug acts through one or two specific molecular targets. A vitamin acts through dozens of simultaneous pathways, many of them interdependent. Designing a study to detect the effects of vitamin D without accounting for the subject’s magnesium status — which is required for vitamin D activation — will underestimate the effect of vitamin D in magnesium-deficient subjects, which in a typical American population is a substantial majority.
A drug’s effects can often be detected in weeks or months. The diseases that vitamins and minerals modulate — cancer, cardiovascular disease, neurodegeneration — develop over decades. A five-year trial of selenium supplementation will detect less effect than a twenty-year trial, not because selenium is less effective but because the cancers that selenium prevents had not yet appeared in the five-year window.
The researchers who conducted the United States Preventive Services Task Force analysis of multivitamins in 2013 acknowledged all of this in their own paper. They wrote that their methodology was “used primarily to evaluate drug therapy” and “might not be ideally suited to evaluating nutrients.” They admitted that “physiologic systems affected by vitamins and other antioxidant supplements are so complex that the effects of supplementing with only one or two components is generally ineffective.” They noted that “the best support for benefit of supplementation came from two multivitamin trials that used physiologic doses of a wider variety of agents.”
Having made these admissions, they proceeded to conclude that the evidence does not support vitamin and mineral supplementation for cancer or cardiovascular disease prevention.
Their analysis defined a “multivitamin” as a formulation containing as few as three nutrients. Out of 154 clinical trials with “multivitamin” appearing in the title in PubMed, they deemed only two methodologically sufficient for inclusion. A single trial on women was included, and the “multivitamin” in that trial contained five ingredients. The researchers themselves conceded that “it could be argued that there are no data on a true multivitamin in women included in this analysis.”
This is not a failure of rigor. It is rigor deployed in service of a predetermined conclusion.
What the Data Says Despite the Methodology
The more revealing exercise is to read what the USPSTF analysis actually found and buried.
The Physicians’ Health Study II enrolled male physicians, gave them a genuine multivitamin for 11.2 years, and found an 8% reduction in overall cancer incidence, a 12% reduction in cancer death, and a 39% reduction in fatal heart attack risk. These are not marginal results. A 39% reduction in fatal heart attack risk from a multivitamin taken for eleven years is a finding that, announced for a pharmaceutical drug, would generate front-page coverage and a Nobel Prize discussion.
The SU.VI.MAX trial found a 31% reduction in total cancer incidence in men over 7.5 years of multivitamin supplementation.
When SU.VI.MAX and PHS-II results were pooled across ten years of follow-up, cancer risk was reduced in both. The USPSTF researchers performed this pooled analysis and then did not present it prominently in their conclusions.
A single trial found a 58% reduction in cancer incidence for subjects supplementing with vitamin D plus calcium over four years. Another found a 45% reduction in overall cancer risk for calcium supplement users.
The European Prospective Investigation into Cancer and Nutrition — EPIC-Heidelberg — enrolled 23,943 subjects and followed them for an average of eleven years. Individuals who used antioxidant vitamin supplements were 48% less likely to die from cancer and 42% less likely to die from any cause compared to non-users.
All of this is inside the research literature. None of it reaches the public as a headline. What reaches the public is SELECT — the study that used the wrong molecules and found no benefit — rephrased as “Selenium and Vitamin E Found Ineffective.”
The asymmetry is not accidental. It is structural.
Follow the Incentive
A pharmaceutical company that patents a molecule can recover its research investment through exclusive market rights. The research funding, the clinical trials, the FDA approval process, the physician education campaigns, the direct-to-consumer advertising — all of it is financed by the margin between the cost of the molecule and the price the patent allows the company to charge.
Selenium is a mineral. It occurs in soil. No one can patent it. The same is true of vitamin D, vitamin C, magnesium, zinc, and every other micronutrient this chapter concerns. The economic model that funds pharmaceutical research does not exist for any of them.
What exists instead is a reverse incentive. The supplement market competes with the pharmaceutical market for the same patients and the same dollars. A patient who prevents cancer with selenium is not a patient who buys chemotherapy. A patient who maintains cardiovascular health with a multivitamin over twenty years is not a patient who takes statins for twenty years. The economic interest of the pharmaceutical industry is not served by research that produces positive results for cheap non-patentable compounds.
The USPSTF is funded by the Agency for Healthcare Research and Quality, which operates within the Department of Health and Human Services, which is staffed substantially by physicians trained within an educational system funded substantially by pharmaceutical companies. The connections are not conspiratorial. They do not need to be. Institutional bias does not require conspiracy. It requires only that the people making decisions about what to fund, what methodology to apply, and what conclusions to publicize share a set of professional assumptions about what counts as evidence and whose evidence counts.
I have been reading this literature since 1976. The positive findings accumulate steadily in the primary literature and vanish almost entirely before they reach the public. The negative findings, particularly the ones produced by methodologically compromised studies using wrong forms of nutrients at inadequate doses for insufficient durations, become headlines and policy statements.
The real crime seems to me to be not the pharmaceutical industry’s behavior, which is at least economically rational, but the behavior of the federal agencies whose statutory purpose is public health and whose demonstrated practice is the suppression of data that threatens their institutional relationships.
The Undebatable Cases
Setting aside the contested literature, certain vitamin and mineral findings meet any reasonable standard of evidence.
Vitamin D deficiency is associated with increased all-cause mortality in every large population study that has examined it. The VITAL trial — 25,000 participants, five years, rigorous RCT — found that vitamin D3 supplementation reduced cancer mortality by 25% overall and by significantly more in normal-weight individuals. This is a large, well-powered, recent, double-blind trial. There is no methodological objection to it that does not equally invalidate pharmaceutical drug trials.
Selenium supplementation in the Clark trial produced results so robust the trial was stopped early. The ethical threshold for stopping a trial early is a high one. The data has to be so clear that continuing would be unconscionable. That threshold was crossed.
Magnesium deficiency is present in an estimated 50-60% of Americans and is associated with cardiovascular disease, type 2 diabetes, hypertension, and neurological dysfunction. Magnesium is required for over 300 enzymatic reactions including the activation of vitamin D. A population deficient in magnesium is a population that will show attenuated results in vitamin D trials. This is rarely controlled for.
Zinc deficiency impairs testosterone synthesis, immune function, and wound healing. Vegetarians and older adults are at elevated risk due to reduced absorption from plant-based diets and declining gastric acid production respectively. Zinc status is rarely measured in clinical practice.
These are not disputed. They appear in standard biochemistry textbooks. The medical establishment’s response to them is to note them in specialty literature and ignore them in clinical practice, because there is no drug to prescribe and no reimbursement code to bill.
What I Have Been Doing Since 1980
In 1980, I began supplementing with a comprehensive vitamin and mineral protocol based on reading the primary literature directly rather than waiting for institutional consensus. The institutional consensus, as this chapter has documented, tends to arrive either never or inverted.
What I read then was the precursor literature to what Clark would publish in 1996: the epidemiological associations between selenium status and cancer incidence, the mechanistic research on antioxidant enzymes, the early vitamin D research on immune modulation, the zinc-testosterone connection, the magnesium-enzyme cofactor work. The conclusions were already visible in the data. They simply had not been packaged into headlines.
Four decades later the headlines have occasionally caught up. The VITAL trial. The NPC trial before it. The EPIC-Heidelberg data. The epidemiological literature on vitamin D is now substantial enough that even the USPSTF has partially reversed its earlier position.
The data did not change. The data was always there. What changed is that the accumulation became too large to ignore with a straight face.
The chapter on combining interventions will address the question nobody has answered: what happens when you add up forty years of these individually validated effects across independent biological pathways. The arithmetic is not in the literature. It is in a life.
Chapter 4: Caloric Restriction — What the 138-Pound Years Taught Me
Key Studies Referenced
Clark LC, et al. “Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin.” JAMA 276(24):1957-1963. 1996.
Klein EA, et al. “Vitamin E and the risk of prostate cancer: the Selenium and Vitamin E Cancer Prevention Trial (SELECT).” JAMA 306(14):1549-1556. 2011.
Sesso HD, et al. “Multivitamins in the prevention of cardiovascular disease in men: the Physicians’ Health Study II.” JAMA 308(17):1751-1760. 2012.
Hercberg S, et al. “The SU.VI.MAX Study: a randomized, placebo-controlled trial of the health effects of antioxidant vitamins and minerals.” Archives of Internal Medicine 164(21):2335-2342. 2004.
Li K, et al. “Vitamin/mineral supplementation and cancer, cardiovascular, and all-cause mortality in a German prospective cohort (EPIC-Heidelberg).” European Journal of Nutrition 51(4):407-413. 2012.
Manson JE, et al. “Vitamin D supplements and prevention of cancer and cardiovascular disease.” New England Journal of Medicine 380:33-44. 2019. (VITAL trial)
Fortmann SP, et al. “Vitamin and mineral supplements in the primary prevention of cardiovascular disease and cancer: an updated systematic evidence review for the US Preventive Services Task Force.” Annals of Internal Medicine 159(12):824-834. 2013.
Leave a Reply