Medical Fads, Part I. Vitamin D Supplementation and the Pitfalls of Observational Studies
In their book Ending Medical Reversal: Improving Outcomes, Saving Lives (Johns Hopkins University Press, Baltimore, 2015) Vinayak Prasad, M.D. and Adam Cifu, M.D. write:
If you have followed the news about prostate cancer screening, mammography for women in their forties, hormone replacement, cholesterol-lowering medications [many trials found that lowering cholesterol levels had no impact on heart disease], and stents for coronary-artery disease, you might think doctors cannot get anything straight. These common practices were … found to be ineffective…. You might be worried that some medical practices are nothing more than fads. In some cases, you might be right.
This is the first of a series exposing medical fads. My aim is to inform medical consumers, thereby making better medical practitioners.
The association of low levels of vitamin D with many illnesses created a fad
As happened with vitamins C and E, vitamin D supplementation has become a fad. Megadose vitamin C supplementation was popularized by the brilliant biochemist Linus Pauling. The fad for megadose vitamin E supplementation in the 1970s was driven by the lay press and pharmaceutical industry.
Unlike those fads, vitamin D supplementation is driven by physicians and supported by substantial peer-reviewed literature, giving it a veneer of legitimacy lacking in those other fads. Physicians ordered more than ten million vitamin D levels on Medicare patients at a cost of $365 million in 2016, up 547% from 2007. Consumer spending on vitamin D rose from $248 million in 2008 to $1 billion annually in 2019.
As of 2020, vitamin D deficiency is recognized to cause only two diseases, rickets in children and osteomalacia in adults. These diseases are characterized by weak bones due to lack of calcium and develop only after prolonged vitamin deficiency.
In contrast, the Vitamin D Society website lists many diseases that they claim can be prevented to some extent by vitamin D supplementation, including many cancers, types 1 and 2 diabetes, and several types of cardiovascular disease. Vitamin D is also widely touted as decreasing the severity of COVID-19.
The vitamin D fad is based on observational studies
The idea that vitamin D supplementation prevents non-skeletal diseases is largely based on observations that people with these diseases have lower levels of vitamin D. Superficially compelling observations like these generate interest in researchers and medical consumers, starting a fad.
However, when those associations are found everywhere, i.e., in practically every investigation, involving diseases that have very different causes, then the significance of those associations is suspect. The associations are most likely due to unmeasured factors called “confounders.” The results were actually caused by a confounding influence that escaped the notice of the investigators.
For example, researchers reported that watching television is associated with inflammation, judging by an abnormal blood test. Obviously, a cause and effect relationship between these is implausible. It is more likely the cause of the association was more mundane: people who watch more television are more likely to have a sedentary lifestyle, be less fit and more obese, and that one of these confounders caused the abnormal blood test.
Confounders often hide in plain sight. They have a powerful influence on the susceptibility and severity of illness, but their effect is so subtle that they are not considered. Socioeconomic status and a sedentary lifestyle are common confounding influences.
The highest-quality evidence disproves the role of vitamin D in nonskeletal illness
The highest quality of scientific evidence comes from prospective, randomized, placebo-controlled studies. These have begun to refute the significance of disease associations with low levels of vitamin D. In January 2019, a study in the New England Journal of Medicine reported that vitamin D supplementation did not decrease the incidence of heart attacks or invasive cancer.
In August 2019, another prospective randomized placebo-controlled study in the New England Journal of Medicine showed that vitamin D supplementation did not decrease the risk of type 2 diabetes. Among the diseases on the Vitamin D Society’s website were cancer, heart attacks, and type 2 diabetes.
Confounding influences created associations between low vitamin D levels and disease
In hindsight, it is obvious that low levels of vitamin D are involved in few if any of the diseases in which it has been implicated by observational data. Because sunlight is required to make vitamin D, any disease which limits sun exposure will decrease levels. Humans also get vitamin D from their diet. Any disease which decreases appetite will also decrease levels of vitamin D.
Thus, low levels of vitamin D may be found in association with any chronic illness. Anyone who is sick is likely to have a lower level of vitamin D than a healthy person. The studies suggesting that supplementation with vitamin D will prevent diseases like cancer, heart disease, and COVID-19 are confounded by the health status of the study subjects.
The fundamental cause of medical fads is, to paraphrase the U.S. author Stephen Vincent Benét, “eating knowledge too fast.” In their zeal to be the most help to their patients, and perhaps be the most up-to-date practitioner in the community, physicians may unwittingly act on data prematurely. Prasad and Cifu write, “Reversals happen when we jump the gun, when doctors start acting based on the paradigm [i.e., received wisdom]rather than on the experimental results. They act as if the hypothesis is true, and the experiment is performed only years later, if at all.”
In the vitamin D supplementation fad, the paradigm is that low levels of vitamin D are widespread, and the hypothesis is that supplementation will impact a number of diseases. The experiments, prospective randomized placebo-controlled studies, have discredited that hypothesis.