Current issues of ACP Journal Club are published in Annals of Internal Medicine


Vitamin D supplementation did not reduce fractures in the elderly

ACP J Club. 1996 July-Aug;125:16. doi:10.7326/ACPJC-1996-125-1-016

Source Citation

Lips P, Graafmans WC, Ooms ME, Bezemer PD, Bouter LM. Vitamin D supplementation and fracture incidence in elderly persons. A randomized, placebo-controlled clinical trial. Ann Intern Med. 1996 Feb 15;124: 400-6. [PubMed ID: 8554248]



To assess the effectiveness of vitamin D supplementation on the incidence of hip and other peripheral fractures in elderly persons.


Randomized, double-blind, placebo-controlled trial with a 3- to 4-year follow-up.


The Netherlands.


2578 persons who were ≥ 70 years of age (mean age 80 y, 74% women) and reasonably healthy. Persons were excluded who had a history of hip fracture or total hip arthroplasty, known hypercalcemia, sarcoidosis, or urolithiasis. Spontaneous use of vitamin D supplements or multiple vitamins was discouraged. Follow-up was > 99%.


Participants were allocated to vitamin D3, 400 IU once per day (n = 1291), or to placebo (n = 1287). Participants were encouraged to consume at least 3 servings of dairy products per day. Compliance was assessed by checking tablet containers at each 6-month refill time and by questionnaire every year.

Main outcome measures

Hip and peripheral fractures analyzed by survival analysis. Hazard rate ratios (HRs) were calculated, unadjusted and adjusted for covariates (age; sex; residence; and outdoor, sunshine, and walking scores).

Main results

Analysis was done by intention to treat and by active treatment, the latter to include patients as long as they were using the trial medication. Hip fractures occurred in 58 patients (4.5%) who received vitamin D supplementation and in 48 patients (4%) who received placebo (P = 0.39) (Table). Other fractures occurred in 77 patients (6%) who received vitamin D supplementation and in 74 patients (6%) who received placebo (P = 0.86) (Table). According to the intention-to-treat analysis, the unadjusted HR was 1.18 (95% CI 0.81 to 1.71) for hip fractures and 1.03 (CI 0.75 to 1.40) for other fractures. Subgroup analysis by residence (apartments and homes for the elderly) or age ≥ 80 years, excluding persons who regularly used vitamins and adjusting for covariates, did not appreciably change the results. The active treatment analysis also showed no difference between the vitamin D and placebo groups. The unadjusted HR for hip fractures was 1.10 (CI 0.71 to 1.70) and for other fractures was 1.02 (CI 0.73 to 1.40). Adjusting for covariates did not change these results.


Vitamin D supplementation did not decrease the incidence of hip or other peripheral fractures in elderly persons.

Sources of funding: Praeventiefonds and Solvay-Duphar, Inc. (vitamin D and placebo tablets).

For article reprint: Dr. P. Lips, Department of Endocrinology, Academisch Ziekenhuis Vrije Universiteit, PO Box 7057, 1007 MB Amsterdam, the Netherlands. FAX 31-20-444-0502.

Table. Vitamin D supplementation vs placebo for preventing fractures in the elderly*

Outcomes at 3 to 4 y Vitamin D Placebo RRI (95% CI) NNH
Hip fractures 4.5% 4% 20% (-17 to 75) Not significant
Other fractures 6.0% 5.7% 3.7% (-24 to 41) Not significant

*Abbreviations defined in Glossary; RRI, NNH, and CI calculated from data in article.

Updated Commentary

The results of this large, randomized clinical trial of vitamin D in the prevention of osteoporotic fractures come on the tail of a similar 3-year study from France that found a substantial protective effect of vitamin D and calcium on the risk for hip fracture (1, 2). The results of this current study must be placed in the proper context so that they do not lead to the frequently heard incantations of health care providers that the number of conflicting studies make it impossible to know what to do. Lips and colleagues clearly point out that several important distinctions exist between their study that was done in the Netherlands and the one done in France.

First, the vitamin D doses differed (800 IU/d in the French study vs 400 IU/d in the Dutch study). Recent comparisons between vitamin D doses in a trial done in the United States (3) show that 800 IU/d is superior to 400 IU/d for maintaining bone mineral density of the femoral neck. Second, the French study included a calcium supplement, whereas the Dutch study only advised in writing that participants should consume 3 servings of calcium per day.

The third difference between the populations was that the French study included nursing home residents, which may account for lower 25-hydroxy vitamin D levels and greater suppression of serum parathyroid hormone levels with treatment in that study population. Nursing home residents, in general, have lower vitamin D levels because they tend to remain indoors with inadequate vitamin D supplementation.

Since 1996, 2 additional studies have examined the efficacy of vitamin D in preventing vertebral and non-vertebral fractures. Baeksgaard and colleagues (4) compared 560 IU of cholecalciferol with placebo (both groups also received 500 mg calcium) for 5 years in postmenopausal women. The baseline calcium intake for each group was 876 mg/day. The relative risk for fracture in the active vitamin D group was 0.33 (95% CI 0.001 to 8.05). In 1997, Dawson-Hughes (5) randomized men and women (mean age 71.5 y) to either 700 IU of cholecalciferol or placebo (500 mg calcium to both groups) for 3 years. The baseline calcium intake per group was 746 mg/ day. The relative risk of non-vertebral fractures was 0.45 (95% CI 0.22 to 0.91). A meta-analysis on this topic by Gillespie and colleagues (6) was limited because it did not address the effect of vitamin D on bone density. Further, Gillespie and colleagues took a relatively conservative approach to pooling the data and made little use of regression methods to explore the appropriateness of combining data across different forms of vitamin D and variations in study design. As such, no pooled results exist for standard formulations of vitamin D used in the above trials. If the studies of Chapuy (1, 2), Lips, and Dawson-Hughes (5) were pooled, the relative risk of non-vertebral fracures with cholecalciferol treatment would be 0.78 (0.55 to 1.09). The 2 studies in which calcium was co-administered with the vitamin D showed the greatest fracture reduction. From these studies, it can be concluded that to prevent non-vertebral fractures, vitamin D should be given at sufficiently high doses with adequate calcium supplementation to persons who have inadequate vitamin D intake.

Douglas P. Kiel, MD, MPH
Harvard Medical School Division on AgingBoston, Massachusetts, USA


1. Chapuy MC, Arlot ME, Duboeuf F, et al. Vitamin D3 and calcium to prevent hip fractures in elderly women. N Engl J Med. 1992;327:1637-42.

2. Chapuy MC, Arlot ME, Delmas PD, Meunier PJ. Effect of calcium and cholecalciferol treatment for three years on hip fractures in elderly women. BMJ. 1994; 308:1081-2.

3. Dawson-Hughes B, Harris SS, Krall EA, et al. Rates of bone loss in postmenopausal women randomly assigned to one of two dosages of vitamin D. Am J Clin Nutr. 1995;61:1140-5.

4. Baeksgaard L, Anderson KP, Hyldstrup L. Calcium and vitamin D supplementation increase spinal BMD in healthy postmenopausal women. Osteoporos Int. 1998;8:255-60.

5. Dawson-Hughes B, Harris SS, Krall EA, Dallal GE. Effect of calcium and vitamin D supplementation on bone density in men and women 65 years of age or oder. N Engl J Med. 1997;337:670-6.

6. Gillespie WJ, Avenell A, Henry DA, O'Connell DL, Robertson J. Vitamin D and vitamin D analogues for preventing fractures associated with involutional and post-menopausal osteoporosis. Cochrane Database Syst Rev. 2001(2):CD000227.