Should vitamin D supplements be recommended to prevent chronic diseases?
BMJ 2015; 350 doi: https://doi.org/10.1136/bmj.h321 (Published 29 January 2015) Cite this as: BMJ 2015;350:h321
- Haakon E Meyer, professor, senior medical officer12,
- Kristin Holvik, researcher1,
- Paul Lips, professor3
- 1Norwegian Institute of Public Health, Division of Epidemiology, Box 4404, Nydalen, 0403 Oslo, Norway
- 2University of Oslo, Department of Community Medicine, Oslo, Norway
- 3Endocrine Section, Department of Internal Medicine, VU University Medical Centre, Amsterdam, Netherlands
- Correspondence to: H E Meyer haakon.meyer{at}fhi.no
The bottom line
Do not recommend vitamin D supplements to prevent chronic disease because clear evidence of benefit does not currently exist and adverse effects cannot be excluded
Vitamin D supplements in doses of 600-800 IU (15-20 µg) per day combined with calcium (0-1000 mg/day, depending on current dairy intake) may be recommended to prevent fractures in elderly people
Vitamin D has gained much attention in research and clinical practice as a possible preventive factor for a wide array of chronic diseases, including cardiovascular disease, various cancers, type 2 diabetes, autoimmune diseases, and chronic obstructive pulmonary disease. Vitamin D3 (cholecalciferol) is a steroid hormone precursor and is synthesised when skin is exposed to ultraviolet B radiation. It is also found in a limited number of foods, especially oily fish. The other form of the vitamin, vitamin D2 (ergocalciferol), is found in dietary plant sterols exposed to ultraviolet B radiation and is somewhat less effective than vitamin D3.1 Vitamin D has well known effects on calcium metabolism and is traditionally linked to the prevention of rickets in children. It is also now clear that vitamin D deficiency causes bone loss through secondary hyperparathyroidism.2
Because vitamin D receptors are present in many organs and tissues, vitamin D may have extraskeletal effects.2 In addition, many observational studies have shown associations between 25-hydroxyvitamin D3 (25(OH)D3), the major circulating form of vitamin D, and the risk of chronic diseases.3 As a consequence of increased popularity, measurement of 25(OH)D3 to determine vitamin D status has become common, and vitamin D supplements, at doses far exceeding the recommended daily allowances or dietary reference values,4 5 6 are often given, despite limited evidence of an effect.
What is the evidence of the uncertainty?
What are the possible benefits?
On the basis of the existing evidence, we can conclude that vitamin D supplements combined with calcium decrease the incidence of fractures in elderly people.7 8 However, there is insufficient evidence that, in the doses tested, vitamin D supplements alone prevent fracture.3 7
It is biologically plausible that vitamin D can help prevent various chronic diseases, and this seems to be supported by observational studies.2 However, observational data might be biased by confounding, and diseases might also lead to low vitamin D rather than the other way around. According to meta-analyses and systematic reviews, clinical trials have shown no consistent effect of vitamin D supplements on the incidence of cardiovascular disease, cancer, chronic obstructive lung disease, or diabetes.3 4 5 9 A Cochrane review concluded that vitamin D3 supplementation for an average of 4.4 years slightly decreased all cause mortality (from 11.4% to 11.0%; relative risk 0.94, 95% confidence interval 0.91 to 0.98), and the result was also significant in those given less than 800 IU a day. However, these results were not considered to be robust enough to recommend widespread supplementation.1 2
Sources and selection criteria
Numerous systematic literature reviews and meta-analyses have been performed on this topic,1 4 5 9 including a recent umbrella review of systematic reviews and meta-analyses of observational studies and randomised controlled trials (RCTs).3 In addition to reviewing these sources, we searched Medline and Embase to identify the RCTs published after 2012 that were not included in these reviews. Several small trials have been published and we restricted our search to large trials with at least 1000 participants that aimed to assess the effect of vitamin D supplements on the incidence of chronic diseases (searching for chronic disease, cancer, cardiovascular diseases, myocardial infarction, cerebrovascular disorder, fractures, mortality, respiratory tract diseases, diabetes mellitus, and multiple sclerosis). No new RCTs were identified beyond those already reported in the published systematic reviews and meta-analyses.
What dosage?
The dose consistent with fracture prevention in randomised controlled trials (RCTs) was in general not higher than 800 IU (20 µg) a day when combined with calcium (1000-1200 mg/day in most trials).7 8 Updated dietary reference values for vitamin D from Europe and the United States, mainly based on its benefits for bone health, are 400-600 IU (10-15 µg) per day for adults and 800 IU (20 µg) per day for elderly people.4 5 6
Who would benefit?
It is plausible that most gain can be obtained by increasing concentrations in people with the lowest baseline values. In a recent meta-analysis of observational studies,10 the inverse association between 25(OH)D3 and total mortality was non-linear, with the largest difference in mortality between the groups with the lowest and second lowest 25(OH)D3 values, a finding that concurs with several other disease outcomes, such as cardiovascular disease and colon cancer.2 Whether optimal serum concentrations of 25(OH)D3 should be 50 nmol/L5 or 75 nmol/L is under debate,11 and this is not helped by standardisation of laboratory assays for 25(OH)D3 being less than ideal.2
Most studies on the health effects of vitamin D have been performed in white populations. The evidence is even more scant in other ethnic groups. Non-Western immigrants living in Western countries often have widespread vitamin D deficiency (serum 25(OH)D3 <25 nmol/L), and their 25(OH)D3 concentration is lower than that of people in their country of origin.12 In the Women’s Health Initiative Observational Study, low 25(OH)D3 was associated with an increased risk of fracture in white women, whereas a high concentration was associated with increased risk of fracture in black and Asian women.13 The Multi-Ethnic Study of Atherosclerosis found that white and Chinese participants with higher concentrations of 25(OH)D3 had a lower risk of coronary heart disease, whereas no association was found in black and Hispanic participants.14
Given all these uncertainties in observational data, it makes sense to limit measurement of 25(OH)D3 to situations where it is unclear whether or not to supplement.
What are the potential hazards and costs?
In a recent Cochrane review, combined vitamin D and calcium supplementation moderately increased the risk of nephrolithiasis (relative risk 1.17, 1.02 to 1.34), whereas vitamin D2 and vitamin D3 had no significant effect on other adverse events.1 Although the risk of vitamin D toxicity is low, side effects from high doses cannot be excluded. In two double blind RCTs that gave annual megadoses of vitamin D, the incidence of falls and fractures increased after vitamin D3 treatment in one,15 and in the other the risk of hip fracture increased after treatment with vitamin D2.16
The likely costs of large scale high dose supplementation and concomitant monitoring would first need to be justified by evidence of benefit.
Is ongoing research likely to provide relevant evidence?
We searched the US ClinicalTrials.gov database (www.clinicaltrials.gov) for trials with at least 1000 participants that tested the effects of vitamin D supplements on the incidence of chronic diseases and mortality in the general population. We identified five ongoing randomised controlled trials and are aware of three others (table ⇓). Many previous trials are limited by the fact that they were performed in elderly people; fracture prevention was the primary outcome, with chronic disease (such as cardiovascular disease) being analysedretrospectively; low doses of vitamin D were given, often in combination with calcium; and follow-up was relatively short. New trials use much higher doses than most of the earlier large trials, and they aim to assess whether high dose vitamin D supplementation without concomitant calcium can prevent chronic disease in healthy populations. However, these trials may not completely answer this question. For example, the VITAL trial (table) was planned with a statistical power to detect risk reductions of 15-40% (depending on the outcome),17 but the true preventive effect of vitamin D on chronic diseases could be less than 15%. The suggested effect of vitamin D supplements on total mortality was small according to the Cochrane review (relative risk 0.94).1 An effect of this size might be regarded as negligible at an individual patient level but might be relevant at the population level. In addition, owing to the inclusion criteria, a large proportion of participants in new trials will probably not have a low initial 25(OH)D3 concentration, and the statistical power to study subgroups with a low vitamin D status at baseline will be smaller. As well as the ongoing studies in the table, our suggestions for further trials are in the recommendations box.
Recommendations for further research
Population: People over 60 years with low vitamin D status (serum 25-hydroxyvitamin D3 <50 nmol/L)
Intervention: Vitamin D3 800 IU/day or vitamin D3 2000 IU/day.
Control: Placebo
Outcome: Cardiovascular disease, type 2 diabetes, respiratory diseases, cancer, mortality
Ongoing and planned randomised controlled trials with at least 1000 participants that will study the effect of vitamin D on the incidence of chronic diseases, fractures, and mortality in adults
What should we do in the light of the uncertainty?
We need a balanced view on vitamin D, with not too little and not too much. Because clear evidence of benefit over harm for vitamin D has not been proved, we should not recommend vitamin D supplements for prevention of chronic diseases (such as cardiovascular disease, cancer, chronic obstructive lung disease, or diabetes) until more definitive further research evidence is available.
Vitamin D deficiency (25(OH)D3 <30 nmol/L) should of course be treated to prevent skeletal complications. Vitamin D supplementation at doses of 600-800 IU (15-20 µg) per day combined with calcium may be recommended to prevent fractures in elderly people, according to evidence from a Cochrane review and updated nutritional recommendations for vitamin D intake in the general population. 4 5 6 7
Notes
Cite this as: BMJ 2015;350:h321
Footnotes
This is one of a series of occasional articles that highlight areas of practice where management lacks convincing supporting evidence. The series adviser is David Tovey, editor in chief, the Cochrane Library. To suggest a topic for this series, please email us at uncertainties{at}bmj.com.
The authors would like to thank Julie Whittle Johansen for proofreading the manuscript.
Contributors: HEM wrote the manuscript draft, assisted by KH and PL. HEM conducted the literature review. All authors approved the final version of the manuscript. HEM is guarantor.
Competing interests: We have read and understood BMJ policy on declaration of interests and declare the following interests: KH received payment for performing independent food safety assessments concerning vitamin D as a member of the Norwegian Scientific Committee for Food Safety’s panel on nutrition, dietetic products, novel food and allergy. She also received an honorarium from Renapharma for a lecture in an educational symposium for general practitioners in November 2012. The funding company had no influence on the content of the lecture. PL and HEM have none.
Provenance and peer review: Commissioned; externally peer reviewed.