Vitamin Studies: Rebuttal to Allegation That Certain Vitamins May Shorten Lifespan
It is almost inconceivable to think that for the greater part of the 20th century, a ?scientific? debate raged as to whether cigarette smoking was dangerous. An even longer running controversy focused on whether food choices (i.e. diet) had any effect on how long people lived. It was not until the later part of the 20th century that the FDA admitted that high saturated fat diets increase heart attack risk.
Today?s unanimous recommendation to eat lots of fresh fruits and vegetables is a relatively recent phenomenon. Back 50 years ago when mid-life heart attack rates were nearly triple what they are now, it was not unusual for people (especially men) to eat virtually no fruits or vegetables. Even now, the processed food industry heavily lobbies government agencies to not recommend disease-preventing foods to the public.
Just as companies that profited by selling cigarettes and processed foods egregiously misled the public in the past, today?s consumers are confronted with so-called ?scientific? reports that question the value of some dietary supplements.
A report released on April 16, 2008 went as far as to suggest that certain vitamins might shorten lifespan. All of this is reminiscent of the scientific charade perpetrated by tobacco companies who falsely claimed that cigarettes did not cause lethal disease. In this case, the economic beneficiaries will be pharmaceutical companies who can expect increased sales of prescription drugs to those who fall for the latest media hype.
As the Life Extension Foundation has done for the past 28 years, we will succinctly separate the facts from the fiction so our members can draw a rationale conclusion about this latest hoax.
Outlandishly low and high potencies evaluated
As has been the case of previous studies conducted by mainstream doctors, the potencies of the nutrients evaluated in this negative report are far different than what knowledgeable health conscious consumers take on a daily basis.
This report condemning supplements claimed that vitamin A caused a 16% increased risk of dying. The vitamin A potencies given to study subjects ranged from 1,333 IU to 200,000 IU a day, or taken every other day, or in one case, ingested just once! Yes, a trial used to vilify this nutrient gave the participants one dose of vitamin A. When these nursing home subjects did not show greater survival benefits, the authors of this negative report blamed the single dose of vitamin A for the deaths.
We don?t know of anyone who actually takes their vitamin A in these kinds of doses, but these were the very studies used to assassinate vitamin A.
This negative report also claimed a 4% increased risk of dying in those who took vitamin E. One might speculate that the failure to supplement with gamma tocopherol may have caused this mortality increase (we discuss more about this later).
The stark reality is that the potencies of alpha tocopherol vitamin E evaluated in the negative report ranged from 10 IU to 5000 IU, which of course has nothing to do with what informed consumers are actually taking. In fact, 10 IU of vitamin E does not even meet the daily minimum requirement for this nutrient. Nonetheless, the authors of this negative report included this study with a host of others to attack vitamin E.
For beta-carotene, the authors claimed a 7% increased risk of death. Doses evaluated ranged from 2000 IU to 83,350 IU. We don?t know of anyone taking this high dose of beta-carotene, but these studies were nonetheless used to further frighten the public against this supplement.
Study periods evaluated ranged from 28 days to 12 years. The authors apparently thought that study periods as short as 28-days were long enough to include in their attack against certain vitamins.
As you will read next, these details about potencies have little meaning when one learns of the study-selection bias that resulted in 91% of evaluable vitamin studies being omitted from the analysis.
Most vitamin studies excluded from report
Out of a total of 815 vitamin studies considered for evaluation, 748 studies were excluded from the analysis (only 67 trials were included in the final report). Here are the absurd reasons given for excluding these 748 studies:
405 trials out of the 748 were removed from the statistical analysis plan because there were no deaths reported in any of the treatment arms. Therefore, this meta-analysis excluded 405 trials that showed no increase in mortality risk.
245 studies out of the 748 were removed from the statistical analysis plan because the authors? inclusion criteria were not fulfilled. Double-blind, randomized, controlled intervention studies were excluded in the analysis for a number of reasons that can only be described as so exacting that many studies published in the New England Journal of Medicine, JAMA, and other top-tier medical journals would fail these criteria. Here were reasons given for excluding these 248 studies:
Specific allocation methods for study participants were not described in detail.
A combination of dates and/ or ?admittance? numbers was used to randomize participants rather than complex study participant identifier codes.
Methods used to hide the method of vitamin allocation were not described in detail.
The specific method of blinding was not described in detail
There were no drop-outs or withdrawals but the exact reason(s) for the lack of drop-outs or withdrawals were not described in detail
In trials with participants that dropped out or withdrew, the exact number and/ or specific reason for drop-out or study withdrawal was not described in detail
So while the media was running headlines like ?Taking Vitamins May Shorten Your Life?, the report the headline was based on omitted most of the vitamin studies that should have been included. This obvious bias rendered the findings meaningless, and we applaud certain news networks who withdrew this defective report from their websites within hours of posting it. One news network even followed with a positive report on the value of antioxidants.
Flawed statistical analysis
The authors used a variety of statistical tools and models to manipulate the data used in the negative report. One way they did this was to use both a ?random effect? model, and a ?fixed effect? model.
Much of what we are going to say here will be confusing to lay people. The reason we have to include it is that this entire report was based on statistical models chosen by the report?s authors. As many of you know, when attempting to conduct a meta-analysis this extensive, statistical methods can be used to create conclusions that may differ considerably from what the underlying data reveal.
The ?random effect? model can be used to identify for difference of effect of antioxidant vitamins on separate patient populations (for example, to assess for the effect of vitamins on cancer patients as against heart disease patients). The ?random effect? model can also be used to help determine if different doses of an antioxidant vitamin have different effects (for example, whether a high dose of vitamin A is associated with mortality risk in contrast to a low dose of vitamin A).
For this meta-analysis, the ?random effect? model can be used to evaluate if different doses and/ or single or combined antioxidants and/ or interventions significantly affects mortality. Out of the 67 studies included in the meta-analysis, antioxidants were administered alone, or in combination with other interventions (e.g. drugs, minerals, other antioxidants) daily or on alternate days for 28 days to 12 years at a very wide range of dosages, such as 10 IU to 5000 IU for vitamin E.
In this meta-analysis, the authors found that the ?random effect? model was not significant, meaning it failed to show an increase in mortality. So they used instead the ?fixed effect? model to show a 4% increased risk.
In essence, the results of this meta-analysis suggest that different doses of vitamins, different patient populations, and single vs. combined antioxidants had absolutely no effect upon mortality risk when the initial statistical model is used, but that vitamins increase mortality risk regardless of dose or patient population when a different statistical model is substituted. This is patently absurd, and calls into question the entire legitimacy of this meta-analysis.
In point of fact, the authors of this flawed meta-analysis were asked to only include the ?random effects? model in a prior presentation of this data in the Journal of the American Medical Association (JAMA) (Bjelakovic 2007a). By using the ?fixed effect?? statistical model and excluding fully 91% of the eligible studies for the analysis, the authors succeeded in achieving headlines by selectively emphasizing models that achieved statistical significance.
Most of the analyzed studies were done on sick people
This negative report attacking certain supplements recommends that healthy people should not take these nutrients, yet 46 out of the 67 studies that were evaluated were conducted on subjects that were diagnosed with disease.
The studies cannot relate mortality differences based on these vitamins
More than 60% of Americans regularly ingest some sort of dietary supplement. The question is, are the 60% of Americans taking supplements going to live longer than those who don?t? An analysis of the scientific literature indicates they probably won't.
The reason is that few Americans are taking enough of the proper nutrients to duplicate the clinical studies showing that the diseases of aging may be preventable.* Twenty-eight years ago, the Life Extension Foundation began a systematic review of published scientific findings relating to the prevention of degenerative disease and aging. The results of this painstaking investigation provided convincing evidence that the killer diseases of aging could be largely prevented by the proper intake of nutrients, hormones, certain drugs, and lifestyle changes.*
ik neem er omdat ik er GEEN genoeg van heb in m'n lichaam tenzij ik ze supplementeer... en dit absoluut zonder gevaar, want indien ik een overschot zou hebben plas ik dit gewoon uit
