Intended for healthcare professionals

Letters Vitamin D and risk of cause specific death

Vitamin D has a greater impact on cancer mortality rates than on cancer incidence rates

BMJ 2014; 348 doi: (Published 29 April 2014) Cite this as: BMJ 2014;348:g2862
  1. William B Grant, researcher1,
  2. Cedric F Garland, adjunct professor2
  1. 1Sunlight, Nutrition and Health Research Center, San Francisco, CA 94164-1603, USA
  2. 2Division of Epidemiology, Department of Family and Preventive Medicine, University of California San Diego, La Jolla, CA, USA
  1. wbgrant{at}

Chowdhury and colleagues’ meta-analysis of cancer specific incidence and mortality rates with respect to bottom versus top third of baseline 25-hydroxyvitamin D levels found higher relative risks for those with cancer at time of enrolment than those without.1 The implication of this finding is that vitamin D has a much stronger impact on survival after developing cancer than on reducing the risk of developing cancer.

Findings of reduced risk of cancer from ecological studies are much stronger than those of observational studies, partly because of larger numbers of cases, and partly because of the stronger effect of vitamin D on cancer mortality rates than on cancer incidence rates.2 It is reasonable that vitamin D might have a greater effect on cancer mortality than on incidence because there are many risk factors for cancer incidence but only a few natural mechanisms in the body, such as vitamin D, that reduce the progression and metastasis of cancer.2

Clinical trials are therefore needed to investigate the effect of eliminating vitamin D deficiency on survival in patients with cancer. Guidelines for vitamin D randomised controlled trials were recently proposed.3 In addition, a committee of the National Institute of Sciences-Institute of Medicine and a guideline of the Endocrine Society of the United States have recommended vitamin D doses for adults from 2000-10 000 IU/day.4 5 On the basis of these guidelines, measurement of serum and supplementation (if needed) should be considered for patients with cancer with 25-hydroxyvitamin D levels <100 nmol/L. The dose should be tailored to individual patients on the basis of baseline serum 25-hydroxyvitamin D level before vitamin D is begun.


Cite this as: BMJ 2014;348:g2862


  • Competing interests: WBG receives funding from Bio-Tech Pharmacal (Fayetteville, AR), the Sunlight Research Forum (Veldhoven), and the UV Foundation (McLean, VA). CFG has none to declare.

  • Full response at:


View Abstract