Does the Latest Vitamin D Meta-Analysis Cast Doubt on Vitamin D Recommendations?
Headlines today are picking up on latest meta-analysis from Bolland and co-workers, which contains the bold conclusion “there is little justification for prescribing vitamin D supplements to prevent falls”. This is strange, since recent meta-analyses from other groups, such as the U.S. Preventive Services Task Force (USPSTF) and from the rigorous Cochrane Collaboration recommend vitamin D to prevent fractures in institutionalized adults. Has something changed recently in the evidence base to warrant this conclusion?
The answer of course is “no” – the evidence has not abruptly changed. What is different is the way the questions asked and the statistical approach used by Bolland and colleagues. The Cochrane and USPSTF groups both conducted classical meta-analyses of the evidence. This involves pooling the results of clinical trials to determine effects across studies. The group led by Bolland used a “trial sequential meta-analysis”. This technique, sometimes used in clinical trials, utilizes an interim analysis to determine if a particular endpoint will be reached if the study continues to its planned conclusion. The approach is useful for clinical trials as the trial can be stopped if the treatment is not going to be effective, ending clinical trial expenditures. The idea of using this strategy within a meta-analysis (estimating the point at which additional clinical studies would not be likely to change the results within a meta-analyses) is unusual. Unlike the relatively controlled environment of a clinical trial, in which doses, duration of supplementation, endpoint determination and population are similar between participants, there are many unknowns to consider because of the variability in experimental design among studies that can be included in a meta-analysis. The populations studied, the doses used, the duration of supplementation and the endpoints measured all differ widely and may affect the results above and beyond the treatment used. Because researchers build upon learnings from previous trials, they often make significant changes in trial design and outcome measures, making later trials effective whilst earlier trials may have produced null results.
The Bolland meta-analysis found that vitamin D supplementation was unlikely to reduce risk of falls at the 15% threshold. Their sensitivity analysis found that vitamin D did not reduce risk of falls by 10% but the majority of studies found a reduction in falls from vitamin D supplementation. While we can debate the validity of setting a lower a threshold for risk reduction than 10% as a target, over 2.4 million falls occur each year in older US adults (CDC). Even a modest 5% reduction in risk would mean over 100,000 fewer falls per year.
So, why the different conclusions by different research groups? The USPSTF and Cochrane analyses make very specific recommendations regarding the use of vitamin D to prevent fractures whereas the Bolland paper actually looked at falls. Although fractures generally result from falls, not all falls result in fractures.
Vitamin D is thought to have two modes of action: it helps maintain bone mineral density and muscle mass. While an effect on muscle could prevent falls, higher bone mineral density (which can also be the result of improved muscle tone) reduces risk of fractures after a fall. Vitamin D may therefore have a greater impact on reducing fractures than falls. After all, a reduction in falls only indirectly affects risk of fractures. Loss of quality of life and increased health care costs are caused by fractures.
It seems as though the evidence has not changed. The recent publication used an analysis technique that set arbitrary effect thresholds and is perhaps better suited for the relatively homogeneous and controlled circumstances of a clinical study. They reported on risk of falling, which is an endpoint that is only indirectly related to the more clinically useful risk of fractures. The current evidence still supports the use of vitamin D supplements to prevent fractures.
Dr Mark J Bolland PhD, Andrew Grey MD, Greg D Gamble MSc, Prof Ian R Reid MD. Vitamin D supplementation and falls: a trial sequential meta-analysis. The Lancet Diabetes & Endocrinology - 24 April 2014. DOI: 10.1016/S2213-8587(14)70068-3
Avenell A, Gillespie WJ, Gillespie LD, O'Connell D. Vitamin D and vitamin D analogues for preventing fractures associated with involutional and post-menopausal osteoporosis. Cochrane Database Syst Rev. 2009 Apr 15;(2):CD000227. doi: 10.1002/14651858.CD000227.pub3. http://www.ncbi.nlm.nih.gov/pubmed/19370554
Cederholm T1, Cruz-Jentoft AJ, Maggi S. Sarcopenia and fragility fractures. Eur J Phys Rehabil Med. 2013 Feb;49(1):111-7. http://www.ncbi.nlm.nih.gov/pubmed/23575205
Chung M, Lee J, Terasawa T, Lau J, Trikalinos TA. Vitamin D with or without calcium supplementation for prevention of cancer and fractures: an updated meta-analysis for the U.S. Preventive Services Task Force. Ann Intern Med. 2011 Dec 20;155(12):827-38. doi: 10.7326/0003-4819-155-12-201112200-00005. http://www.ncbi.nlm.nih.gov/pubmed/22184690