This site uses cookies to store information on your computer. Learn more x


Providing perspectives on recent research into vitamins and nutritionals


Vitamin D and Health: Still Accruing the Evidence

By Julia Bird

At TalkingNutrition, we identified vitamin D as the most researched vitamin in both 2012 and 2011, and even though we haven’t started our analyses yet for 2013, based on what has been coming through our daily searches, it will probably maintain its top position for this year. A systematic review published late yesterday in The Lancet Diabetes and Endocrinology by Autier, Boniol, Pizot and Mullie summarized results from a great number of vitamin D publications, both observational and intervention studies. They included the results from 290 prospective cohort studies and 172 randomized trials looking at a great number of endpoints.

The observational research showed a decrease in cardiovascular events, diabetes and the metabolic syndrome when the highest compared to lowest levels of vitamin D were compared. There were also modest decreases in some infectious diseases, and a fairly robust finding was an around 30% decrease in all-cause mortality in people in the highest group of vitamin D levels. When looking at individual health conditions, vitamin D supplementation did not appear to affect cardiovascular disease outcomes, diabetes risk factors, or cancer. There was a small, significant decrease in overall risk of all-cause mortality of around 5%. These trials were mostly performed in elderly women living in nursing homes and may reflect reduction in risk of osteoporosis, falls, fractures, and their associated outcomes.

The issue with observational research is that it is subject to confounding. In epidemiologists’ language, this is the association between a particular factor and the exposure of interest (in our case, vitamin D) and the outcome. Some known confounders for vitamin D status include: time spent outdoors (increases vitamin D levels due to sun exposure), health knowledge (decreases vitamin D status due to sun safe behaviors), age (reduction in time spent outside and reduced skin production), healthy diet (increases vitamin D levels due to consumption of vitamin-D rich foods such as oily fish, mushrooms, and fortified milk), use of dietary supplements (increases vitamin D status). These factors are all also linked to various health outcomes. That higher vitamin D levels are protective of various health outcomes could be actually a marker of greater outdoor recreation, better health or nutrition knowledge, or younger age. Epidemiologists normally attempt to correct for these factors, but it is always possible that something was overlooked or could not be measured.

This is why intervention studies are so important. A randomized, controlled design allows researchers to avoid the pitfall of confounding that is an issue in observational research. At the moment, it seems like vitamin D supplements in general could reduce all-cause mortality. To properly look at whether vitamin D supplements can affect important outcomes such as cardiovascular disease and cancer is very difficult as the biological changes that lead to these diseases often start decades before diagnosis. Large, long term studies are currently underway to collect more robust evidence.

In the meantime, what are the risks of trying to prevent vitamin D deficiency? Should people avoid supplements until the rest of the research comes in? Vitamin D is safe when it comes from a reliable source and is taken as indicated. If people can afford to take supplements, and are concerned about their vitamin D levels (for example, they do not get enough vitamin D from foods, spend a lot of time indoors, or live in an area in which the skin cannot produce vitamin D in the winter), there is a good reason to ensure that intakes of vitamin D are adequate.

Main citation:

Prof Philippe Autier MD,Prof Mathieu Boniol PhD,Cécile Pizot MSc,Prof Patrick Mullie PhD. Vitamin D status and ill health: a systematic review. The Lancet Diabetes & Endocrinology - 6 December 2013. DOI: 10.1016/S2213-8587(13)70165-7