Questioning vitamin D status of elderly fallers and nonfallers: a meta-analysis to address a ‘forgotten step’

Authors

  • C. Annweiler,

    Corresponding author
    1. Department of Neuroscience, Division of Geriatric Medicine and Memory Clinic, Angers University Hospital and UPRES EA 4638, University of Angers, UNAM, Angers, France
    2. Robarts Research Institute, The University of Western Ontario, London, ON, Canada
    • Correspondence: Cedric Annweiler MD, PhD, Department of Neuroscience, Division of Geriatric Medicine, Angers University Hospital, 49933 Angers Cedex 9, France. (fax: ++33-2-41-35-48 94; e-mail: CeAnnweiler@chu-angers.fr).

    Search for more papers by this author
  • O. Beauchet

    1. Department of Neuroscience, Division of Geriatric Medicine and Memory Clinic, Angers University Hospital and UPRES EA 4638, University of Angers, UNAM, Angers, France
    2. Biomathics, Angers, France
    Search for more papers by this author

Abstract

Background

Previous meta-analyses to determine the efficacy of vitamin D supplementation to prevent falls in the elderly have shown mixed results. Inconsistencies might depend on the dose of supplements, suggesting that serum 25-hydroxyvitamin D (25OHD) concentration could influence the risk of falling. Our objective was to systematically review and quantitatively analyse the relationship between serum 25OHD concentration and the occurrence of falls.

Methods

A Medline search was conducted in December 2013, with no date limit, using the Medical Subject Heading terms ‘Vitamin D’ OR ‘Ergocalciferols’ OR ‘Vitamin D deficiency’ combined with ‘Accidental Falls’ OR ‘Gait disorders, neurologic’ OR ‘Gait apraxia’ OR ‘Gait’ OR ‘Recurrent Falls’ OR ‘Falling’. Fixed and random-effects meta-analyses were performed to determine the following: (i) the effect size of the difference in 25OHD concentration between fallers and nonfallers and (ii) the risk of falling according to serum 25OHD concentration.

Results

Of the 659 retrieved studies, 18 observational studies – including ten cross-sectional and eight cohort studies – met the selection criteria. All were of good quality. The number of participants ranged from 80 to 2957 (44–100% women); 11.0% to 69.3% were fallers. Serum 25OHD concentrations were 0.33 × SD lower in fallers compared to nonfallers [pooled effect size 0.33; 95% confidence interval (CI) 0.18–0.47]. The risk of falls was inversely associated with serum 25OHD concentration [summary odds ratio (OR) 0.97; 95% CI 0.96–0.99]. The association between falls and hypovitaminosis D varied according to the definition used; the summary OR for falls was 1.23 (95% CI 0.94–1.60) for 25OHD <10 ng mL−1, 1.44 (95% CI 1.17–1.76) for 25OHD <20 ng mL−1 and 0.95 (95% CI 0.81–1.11) for 25OHD <30 ng mL−1.

Conclusions

Fallers have lower 25OHD concentrations, notably more often <20 ng mL−1, than nonfallers. These findings help to determine the profile of target populations that would most benefit from vitamin D supplements to prevent falls.

Ancillary