Published on October 3, 2017

In April 2016, GrassrootsHealth published a research paper which focused on vitamin D levels achieved and cancer incidence.

Why is this important? Because we have found, through our D*action members, that there is a six-fold variability in achieved serum levels. Really? Yes! That means that two people, both taking 5,000 IU/day, could be vastly different. Our study data has found that at 5,000 IU/day one person achieved a paltry 20 ng/ml (50 nmol/L) and another a whopping 120 ng/ml (300 nmol/L).

Who was involved with this paper?

GrassrootsHealth has four researchers on staff capable of analyzing and publishing study data. We can use data from our three population cohorts – D*action, Breast Cancer Prevention, and Protect our Children NOW! and have partnerships with leading researchers who allow us access to their study data. For this paper, we partnered with leading vitamin D and cancer researchers – Dr’s Garland, Gorham, and Lappe as well as late GrassrootsHealth Research Director, Dr. Heaney.


Was this a new study? Where did the data come from?


For this analysis we combined data for women 55 and older from our GrassrootsHealth cohort (N = 1,135, median serum level = 48 ng/ml) and the cohort of a previously published randomized controlled trial (RCT) of vitamin D and calcium supplementation with respect to cancer (Read Lappe RCT paper; N= 1,169, median serum level = 30 ng/ml) and used the pooled cohort to investigate cancer incidence over time (median = 3.9 years).  Combining cohorts gave us a wider range of serum levels, more data, and thus improved statistical power. We looked at all invasive cancers combined, excluding skin cancer.

Lappe-GRH 2016 paper Cancer Curve by serum level


Using statistical analysis, we found that women with vitamin D serum levels greater than or equal to 40 ng/ml had a 71% lower risk of cancer than women with serum levels less than 20 ng/ml. In the graph above you can see that the greatest decrease in risk occurred between about 10 – 40 ng/ml. Calcium did not have any significant effect on cancer risk in this analysis.


Same data, another view

A different way to look at the data is through what a Kaplan-Meier plot which shows the proportion of cancer-free participants (so you want to be close to 1.0, or fully cancer-free) over time by their vitamin D level. Notice the group of participants at or above 40 ng/ml (100 nmol/L) are in a line significantly above the less than 20 ng/ml (50 nmol/L) line.


GRH Lappe cancer fig 3 kaplan meier plot

Future Studies


In order to change the standard of care and have consistent results, future studies should base their analysis on vitamin D status. They should analyze by vitamin D serum level, and not limit analysis to a treatment group or dosage and they should try to effect change over the complete response range (with overall cancer this seems to be between 10-40 ng/ml).



Serum 25-Hydroxyvitamin D Concentrations ≥40 ng/ml are Associated with >65% Lower Cancer Risk: Pooled Analysis of Randomized Trial and Prospective Cohort Study
Sharon L. McDonnell, MPH et al.
April 2016
Read Paper