expert opinion

Why do studies not consider dose–response relationships?

September 1, 2012


Robert Heaney, Ph.D., Professor of Medicine, Creighton University, and Jeffrey Blumberg, Ph.D., C.N.S., Director, Antioxidants Research Laboratory, Tufts University, USA

“Unfortunately, most of the randomized, controlled trials of vitamin D that have been published to date have paid little attention to baseline status, i.e. the blood vitamin D concentrations of the participants at the beginning of the study. Among the 31,022 patients whose results were analyzed, for example, by a recent meta-analysis (1), data on baseline concentrations of 25-hydroxyvitamin D were available for only 4,383 patients (barely 14%). Instead, the studies focused almost exclusively on the assigned dose. The dose–response relationship that vitamin D shares with most nutrients makes it clear that giving additional amounts of a nutrient to people who already have a large enough intake of that nutrient, or not giving enough to push a person with a deficiency up onto the ascending limb of the response curve, is likely to produce a null response. In this regard, as in several other respects, nutrients are unlike drugs (2). Once an adequate concentration has been achieved, additional intake has no effect. This truism is little more than a restatement of a longstanding skepticism among clinicians about the purpor-ted benefits of many nutrient supplements (3) and is the explicit basis for the recommendations of the Institute of Medicine.

Despite the consensus that more is not better, we have continued to conduct trials (and include them in me-ta-analyses) without regard to ensuring the presence of two key features: baseline status and dose adequa-cy. For example, two large, randomized, controlled trials tested the effect of supplemental calcium on the risks of preeclampsia or fracture in patients whose baseline calcium intakes were already at the recommen-ded levels for adequacy (4,5). Both trials had null outcomes. But both failed to address the underlying hypo-thesis that low calcium intake increases the risk of preeclampsia or fracture because neither trial included a group with low calcium intake. Nevertheless, both trials were included in the systematic review (6) used by the Institute of Medicine in formulating its intake recommendations for calcium (7). Because of their relati-vely large samples, both trials heavily weighted the aggregate effect toward the null hypothesis in the corresponding meta-analysis.

Dose adequacy, the second of the two key considerations, was specifically addressed by the latest meta-analysis (1), which used individual adherence data to modify the assigned dose. They found that fracture risk was reduced only among people who were assigned to receive doses of 800 IU per day or higher – a finding that would be more convincing if it were accompanied by data on the baseline concentration and induced change in the level of 25-hydroxyvitamin D, but very few of the included studies provided this information. Nevertheless, such an intake is consistent with the guidelines for adults that have been issued by the Endo-crine Society (1500 to 2000 IU per day) (8).

The question of how much vitamin D is enough is likely to remain muddled as long as meta-analyses focus on trial methodology rather than on biology. Given the congruence of the findings of this latest meta-analysis (1) with the guidelines from the Endocrine Society, it would appear to be prudent, and probably helpful as well, to ensure an intake at the upper end of the range at which the analysis found a reduction in fracture risk.”

Based on: Heaney R. P. Vitamin D — Baseline Status and Effective Dose. N Engl J Med. 2012; 367:77–78.


  1. Bischoff-Ferrari H. A. et al. A pooled analysis of vitamin D dose requirements for fracture prevention. N Engl J Med. 2012; 367:40–49.
  2. Heaney R. P. Nutrients, endpoints, and the problem of proof. J Nutr. 2008; 138:1591–1595.
  3. Goodwin J. S. and Tangum M. R. Battling quackery: attitudes about micronutrient supplements in American academic medicine. Arch Intern Med. 1998; 158:2187–2191.
  4. Levine R. J. et al. Trial of calcium to prevent preeclampsia. N Engl J Med. 1997; 337:69–76.
  5. Jackson R. D. et al. Calcium and vitamin D supplementation and the risk of fractures. N Engl J Med. 2006; 354:669–683. [Erratum, N Engl J Med 2006;354:1102.]
  6. Chung M. et al. Vitamin D and calcium: a systematic review of health outcomes. Evid Rep Technol Assess (Full Rep). 2009; 183:1–420.
  7. Institute of Medicine. Dietary reference intakes for calcium and vitamin D. Washington, DC: National Academies Press, 2011.
  8. Holick M. F. et al. Evaluation, treatment, and prevention of vitamin D deficiency: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2011; 96:1911–1930.