Fixing A Broken Vitamin D Status...That's Actually Broken
In writing for TalkingNutrition, I routinely post blogs that say “if only this researcher did XYZ, they might have seen an effect!” (where XYZ is normally recruiting subjects with low nutrient status at baseline). It brings me great joy when I get to see a similar study to the one I just wrote the “if only!” blog about that actually did what I had wished for and then sees a positive effect. And it’s particularly exciting for me when both examples happen in the same week.
Earlier this week I wrote about a vitamin D intervention trial by Javed et al. where teens were provided high-dose (100,000 IU/month) supplements, and no results were seen – possibly because the average starting serum vitamin D concentrations for the majority of the subjects were above the definition of adequacy set by the Institute of Medicine (20 ng/mL). This week also saw the release of a new vitamin D intervention trial with a similar design, albeit a different population.
Houston and colleagues sought to determine whether providing vitamin D supplements (100,000 IU/month for 5 months) to homebound older adults via the Meals-on-Wheels program would reduce the incidence of falls. Sure enough, after 5 months, their analysis demonstrated a 58% risk lower risk of falls among those receiving vitamin D relative to those receiving placebo (RR = 0.42, 95% CI = 0.21-0.87). Of note, the mean baseline vitamin D concentration was 20.9 ng/mL, but the majority of their subjects (57%) had serum vitamin D concentrations below 20 ng/mL. Furthermore, at follow-up, 72% of subjects receiving placebo had serum vitamin D concentrations below 20 ng/mL, compared to just 1% of the subjects receiving vitamin D.
Now, is this a direct apples-to-apples comparison to the Javed et al. study? Of course not – these studies involved very different populations (obese teens vs. home-bound elderly) and different outcomes (flow-mediated dilation vs. fall incidence); however there is still a lesson to be learned by comparing these trials. Put simply: vitamin status matters. By ensuring that they recruited subjects with an inadequate vitamin D status, Houston and colleagues better modeled the phenomena seen in epidemiological literature linking vitamin D status to falls and fractures, and thus provided themselves with a greater chance of observing an effect in their trial. By examining a population where the majority of their subjects had adequate vitamin D status to start, Javed et al. effectively stacked the deck against themselves and made it more difficult to prove their hypothesis.
Remember, we shouldn’t be asking ourselves whether the mere act of using a dietary supplement is valuable. We should be asking ourselves whether having an inadequate nutrient status leads to a problem, and whether improving nutrient status fixes that problem. Dietary supplements, like vitamin D, are just one way to improve nutrient status and maintain health.
Houston DK, Tooze JA, Demons JL, et al. Delivery of a vitamin D intervention in homebound older adults using a Meals-On-Wheels program: a pilot study. J Am Geriatr Soc 2015; Aug 16 (epub ahead of print).
Javed A, Kullo IJ, Babu-Balagopal P, Kumar S. Effect of vitamin D treatment on endothelial function in obese adolescents. Pediatr Obes 2015; Aug 14 (epub ahead ofprint).