Fall prevention with supplemental and active forms of vitamin D: a meta-analysis of randomised controlled trials
BMJ 2009; 339 doi: https://doi.org/10.1136/bmj.b3692 (Published 01 October 2009) Cite this as: BMJ 2009;339:b3692All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
In response to Dr. Jose AP da Silva: Fall prevention with Vitamin D.
Clarifications needed.
1J Christopher Gallagher, 2Clifford Rosen
1Creighton University School Medicine, Omaha, USA
2Maine Medical Center, Portland, ME, USA
Recently, in response to a letter to the editors of this journal (1)
requesting clarification on comments offered in an Institute of Medicine
(IOM) report (2) regarding a meta-analysis of vitamin D and fall
prevention (3), the journal published a reply from the authors of the meta
-analysis (4). As members of the IOM committee, we offer the following
response to the authors' reply.
The IOM report criticized reporting inconsistencies and
misrepresentations of results in the meta-analysis as conducted Bischoff-
Ferrari and colleagues. The IOM report's critique raises substantive
concerns and suggests that there should be a re-examination of the results
of the meta-analysis and its interpretation by the authors.
First, in their reply to this journal, Bischoff-Ferrari et al.
acknowledge that they included a trial conducted by Broe et al. (5) in
violation of their stated eligibility criteria. However, they maintain
that it was appropriate to do so because it had a high quality fall
assessment. Even if we give the benefit of the doubt to the authors that
inclusion of this trial is reasonable, it cannot justify the inaccurate
reporting of their methods. Inclusion of the Graafmans et.al study (6)
also violated the inclusion criteria since it was designed as a fracture
study and falls was only added as an end point in a cohort study 2 years
after randomization.
Second, the authors assessed the dose-response relationship between
vitamin D dose and risk of at least one fall. Dose-response relationships
can be explored in many ways, and therefore a range of sensitivity
analyses should be presented. The authors opted to compare subgroups of
studies above or below a cutoff of vitamin D dose and serum 25
hydroxyvitamin D picked by "visual inspection", i.e., in an entirely data-
driven way. The authors found statistically significant differences in two
subgroups (in a meta-regression framework) and interpreted it as evidence
for an inverse relationship. However, analyses using nearby cutoffs are
not statistically significant. Further, re-analysis of the authors' data
using the vitamin D dose as a continuous variable showed no statistically
significant relationship. Therefore, it is fair to state that the
relationship of dose and effect on falls is not clear.
It should be noted that among the negative studies of vitamin D and
falls, the study by Broe et al (5) with only 25 or 26 subjects in each
group on vitamin D doses of 200,400,600 IU (6) was not powered for falls
analysis. In the Graafmans' study with 172 subjects on 400 IU (6)
assessment of falls was carried out for 28 weeks after 2 years on vitamin
D when any benefit from vitamin D could have occurred earlier. In four of
the studies the trial lasted 2, 3 and 5 months. Thus, inclusion of this
trial (6) was quite unlike any of the others where fall collection started
immediately after randomization.
Third, Figure 3 in the published meta-analysis is misleading; it was
not simply misinterpreted by the IOM. The figure uses the layout of a
typical meta-regression plot and shows a "trend line" that resembles a
meta-regression summary line.
Furthermore, trying to establish a threshold between falls and serum 25
hydroxyvitamin D is too simplistic particularly when the serum 25
hydroxyvitamin D assays were performed in studies over a period of 12
years (1996-2008) using different assay methods that are not comparable as
demonstrated on inter assay quality control testing (7)
In a recent double-blinded trial by the same team of authors they found
that vitamin D3 treatment, 2000 versus 800 IU/d, did not reduce falls
(28%; 95% CI, ?4% to 68%) (8).
Taken together, the internal inconsistencies of the paper, the lack
of attention given to the visual display of the quantitative information
in the paper, and the selectivity of the dose-response analyses justified
a re-analysis of the authors' data and led to a more conservative
interpretation of their findings by the IOM.
In summary, the Meta Regression analysis on falls does not support a
role for Vitamin D or show a threshold effect in reducing falls. Any
conclusions regarding a dose response for vitamin D supplementation and
the risk of falls remain tenuous.What is needed now are properly designed
double blind studies of vitamin D on falls and no more meta analyses.
References
1 da Silva JA. Clarifications needed, please. BMJ. 2011; 342:d2602
2 Institute of Medicine. Dietary reference ranges for calcium and vitamin
D. 2010.http://www.iom.edu/Reports/2010/Dietary-Reference-Intakes-for-
Calcium-and-Vitamin-D
3 Bischoff-Ferrari HA, Dawson-Hughes B, Staehelin HB, Orav JE, Stuck
AE, Theiler R, et al. Fall prevention with supplemental and active forms
of vitamin D: a meta-analysis of randomised controlled trials. BMJ 2009;
339:b369
4 Bischoff-Ferrari H, Willet WC, Orav JE, Kiel DP, Dawson-Hughes B.
Fall prevention with vitamin D. BMJ 2011; 342:d2608
5 Broe KE, Chen TC, Weinberg J, Bischoff-Ferrari HA, Holick MF, Kiel
DP. A higher dose of vitamin D reduces the risk of falls in nursing home
residents: a randomized, multiple-dose study. J Am Geriatr Soc2007; 55:234
-9.
6 Graafmans WC, Ooms ME, Hofstee HM, Bezemer PD, Bouter LM, Lips P.
Falls in the elderly: a prospective study of risk factors and risk
profiles. Am J Epidemiol1996; 143:1129
7 Carter GD, Berry JL, Gunter E, Jones G, Jones JC, Makin HL, Sufi S,
Wheeler MJ. Proficiency testing of 25-hydroxyvitamin D (25-OHD) assays.
Steroid Biochem Mol Biol. 2010 Jul; 121(1-2): 176-9.
8 Bischoff-Ferrari HA, Dawson-Hughes B, Platz A, Orav EJ, St?helin
HB, Willett WC, Can U, Egli A, Mueller NJ, Looser S, Bretscher B, Minder
E, Vergopoulos A, Theiler R. Effect of High-Dosage Cholecalciferol and
Extended Physiotherapy on Complications After Hip Fracture. Arch Intern
Med. 2010 May 10; 170(9): 813-20.
Biochem Mol Biol. 2010 Jul; 121(1-2): 176-9.
Competing interests: No competing interests
In response to Dr. Jose AP da Silva: Fall prevention with Vitamin D.
Clarifications needed.
1,2Bischoff-Ferrari HA, 3Willett WC, 4Orav JE, 5Kiel DP, 6Dawson-
Hughes B
1Center on Aging and Mobility, University of Zurich, Switzerland;
2Dept. of Rheumatology, University Hospital Zurich, Zurich, Switzerland,
3Department of Nutrition, Harvard School of Public Health, Boston, USA;
4Department of Biostatistics, Harvard School of Public Health, Boston,
USA; 5Hebrew Senior Life Institute for Aging Research, Harvard Medical
School, Boston, USA; 6USDA Human Nutrition Research Center on Aging, Tufts
University, Boston, USA
We agree with Prof. da Silva, and wish to clarify the three issues
raised by the recent IOM report1 concerning our 2009 meta-analysis on
vitamin D and fall prevention2 (A). In addition, we comment on the overall
recommendation of the IOM (B) on vitamin D and fall prevention1.
(A) Rebuttal on the issues raised by the IOM - regarding our 2009
meta-analysis on vitamin D supplementation and fall prevention:
1. It was stated that our inclusion/exclusion criteria for the
selection of trials were problematic for two studies. Of 8 studies
included in the primary analysis, the IOM questioned the inclusion of Broe
et al.3, which did not pre-specify falls as a primary or secondary
outcome. While this did violate our inclusion criteria, the Broe trial
took advantage of a high-quality fall assessment at the trial site
throughout the course of the trial. In all other regards: blinding,
randomization and fall ascertainment methodology; the Broe trial qualified
for the primary analysis.
The IOM also questioned the omission of Law et al4. The Law study was
excluded appropriately because it was not blinded as required. When the
Law trial was included in a sensitivity analysis along with 6 other
additional trials that did not meet our inclusion criteria, the benefit of
vitamin D on fall prevention remained significant.
2. It was stated that the dose-response relationship in our Figure 3
was inappropriately presented. Our intent was simply to summarize the
findings at varying levels of vitamin D and achieved 25(OH)D in the
treatment groups. The figure was used to visually identify a threshold at
which vitamin D, and 25(OH)D, appeared efficacious (i.e. the RR dropped
below 1), these threshold levels were then used in formal meta-regressions
to explain the significant heterogeneity. Unfortunately, our intentions
were not sufficiently clear and the Figure 3 was misinterpreted as
representing the actual meta-regression. We therefore present Figure 3
without the trend line through the RRs which may have led to the
misinterpretation. For our actual published meta-regression, the RRs were
analyzed on a log scale, looking for a difference between low (< 700 IU
vitamin D per day) and high dose (700 to 1000 IU vitamin D per day) as
identified in Figure 3, and not a linear trend, and individual studies
were included separately and not aggregated in dosage level.
3. Finally, it was stated that one trial (Broe et al.) which had 5
arms (4 different doses of vitamin D versus a common placebo) was
incorrectly treated in our meta-regression as 4 independent meta-analysis
entries. We agree that there are stochastic dependencies (correlations)
between the corresponding risk ratios which refer to the same placebo arm.
Therefore, we converted the published summary results into a pooled
patient-level database and re-ran the analyses as a random effects
logistic regression, allowing for between-patient correlation within all
arms of the Broe study. In this re-analysis, when treatment is the only
predictor (regardless of dose level), there is a significant reduction in
the odds of falling based on our primary analysis: OR=0.73 [.62, .87];
p=.0004. When the model is expanded to capture the impact of both high
dose and low dose treatment, high dose vitamin D (700 to 1000 IU vitamin
D per day) reduces the odds of falling (OR=0.66 [.53, .82] p=.0002),
while low dose vitamin D does not (OR=1.14 [.69, 1.87]; p=.61).
However, different than the reported result, the interaction term does not
reach significance (p = 0.06), but this does not invalidate the overall
significant result or the significantly lower risk with doses of 700 IU or
larger.
Revised Figure 3 without trend line through RRs and showing all
trials individually:
Legend Figure 3: Markers filled indicate trials with oral vitamin D3
(cholecalciferol) and markers unfilled indicate trials with oral vitamin
D2 (ergocalciferol)
(B) Comment on the overall recommendation of the IOM regarding
vitamin D and fall
prevention
The IOM did a thorough review on the effect of vitamin D on fall
prevention. Their synopsis is that the evidence of vitamin D on fall
prevention is inconsistent, which is in contrast to the 2011 assessment of
the Agency for Healthcare Research and Quality (AHRQ) for the U.S.
Preventive Services Task Force5 , the 2010 American Geriatric
Society/British Geriatric Society Clinical Practice Guideline6, and to the
2010 assessment by the IOF7, all 3 of which identified vitamin D as an
effective intervention to prevent falling in older adults. Here are the
primary results that were published in the IOM report (omitting the
analyses restricted to the two studies of injectable vitamin D and
sensitivity analyses):
In summary, both the overall analysis and the majority of the subset
analyses support the use of vitamin D in the prevention of falling. A few
of the analyses that were non-significant (1A + B and 2) showed effect
sizes comparable to the overall pooled findings and the lack of
significance is likely due to reduced sample size and power. The set of
analyses which showed no benefit (4, 4A+B) were based on 4 studies, which
either used low dose vitamin D8, had less than 50% adherence9, had a low-
quality fall assessment10 or used one large bolus dose of vitamin D among
seniors in unstable health11. Thus, we would argue that the main
inconsistency raised by the IOM is based on 4 studies that cannot be
considered reliable indicators of true treatment efficacy, and that these
do not invalidate the overall findings.
References:
1. Institute of Medicine. Dietary Reference Ranges for Calcium and
Vitamin D. http://wwwiomedu/Reports/2010/Dietary-Reference-Intakes-for-
Calcium-and-Vitamin-D/Report-Briefaspx 2010.
2. Bischoff-Ferrari HA, Dawson-Hughes B, Staehelin HB, et al. Fall
prevention with supplemental and active forms of vitamin D: a meta-
analysis of randomised controlled trials. Bmj 2009;339:b3692.
3. Broe KE, Chen TC, Weinberg J, Bischoff-Ferrari HA, Holick MF, Kiel
DP. A higher dose of vitamin d reduces the risk of falls in nursing home
residents: a randomized, multiple-dose study. J Am Geriatr Soc 2007;55:234
-9.
4. Law M, Withers H, Morris J, Anderson F. Vitamin D supplementation
and the prevention of fractures and falls: results of a randomised trial
in elderly people in residential accommodation. Age Ageing 2006;35:482-6.
5. Michael YL, Whitlock EP, Lin JS, Fu R, O'Connor EA, Gold R.
Primary care-relevant interventions to prevent falling in older adults: a
systematic evidence review for the u.s. Preventive services task force.
Ann Intern Med;153:815-25.
6. American Geriatric Society/British Geriatric Society Guidelines on
Fall Prevention in older Persons 2010.
http://wwwamericangeriatricsorg/files/documents/health_care_pros/FallsSu...
7. Dawson-Hughes B, Mithal A, Bonjour JP, et al. IOF position
statement: vitamin D recommendations for older adults. Osteoporos Int.
8. Graafmans WC, Ooms ME, Hofstee HM, Bezemer PD, Bouter LM, Lips P.
Falls in the elderly: a prospective study of risk factors and risk
profiles. Am J Epidemiol 1996;143:1129-36.
9. Grant AM, Avenell A, Campbell MK, et al. Oral vitamin D3 and
calcium for secondary prevention of low-trauma fractures in elderly people
(Randomised Evaluation of Calcium Or vitamin D, RECORD): a randomised
placebo-controlled trial. Lancet 2005;365:1621-8.
10. Trivedi DP, Doll R, Khaw KT. Effect of four monthly oral vitamin
D3 (cholecalciferol) supplementation on fractures and mortality in men and
women living in the community: randomised double blind controlled trial.
BMJ 2003;326:469.
11. Latham NK, Anderson CS, Lee A, Bennett DA, Moseley A, Cameron ID.
A randomized, controlled trial of quadriceps resistance exercise and
vitamin D in frail older people: the Frailty Interventions Trial in
Elderly Subjects (FITNESS). J Am Geriatr Soc 2003;51:291-9.
Competing interests: No competing interests
The recent Institute of Medicine report concludes that there is no
evidence for benefits of vitamin D beyond its effects on bone health and
denies, specifically, that there is a demonstrated effect upon the
frequency of falls. This rested heavily on the committee's re-
interpretation of the meta-analysis of vitamin D supplementation by
Bischoff-Ferrari et al. published in BMJ in 2009, whose methodology is
criticized in the report.
The committee claimed that a regression line through the data points
was inappropriately fit and also claimed that the authors included one
multiple dose trial incorrectly. Although this reader believes that the
IOM may have misinterpreted the original report, it would be useful to all
interested in this field if the authors could be invited to clarify their
analysis and respond to the criticisms of the IOM.
Sincerely,
Competing interests: No competing interests
Vitamin D acts a cellular receptor in the muscle and nervous system
to reduce the muscle weakness in the body. Vitamin D appears to have an
important function as a regulator of cell differentiation and cell growth.
Elderly patients with recurrent history of falls benefit from the
ergocalciferol supplementation in addition to calcium, leading to a 19 %
reduction in relative fall risk (1). PTH also remains as a significant
independent predicter of falls. According to latest research, calcium and
vitamin D supplementation not only improves the bone marrow density but
also helps in the improvement of muscle strength. This in turn helps in
the reduction of further falls.
Reference:
1. Prince R L et al. Effects of ergocalciferol added to calcium on the
risk of falls in the elderly high risk woman. Archives of Internal
Medicine. Jan 2008;vol./is.168/1 (103 -8).
Competing interests:
None declared
Competing interests: No competing interests
Further to recent discussion of the benefits and risks of vitamin D
and calcium[1,2], we wish to draw attention to another unintentional harm.
Vitamin D and calcium supplements are primarily used to treat
osteoporosis and vitamin D deficiency. Such conditions may arise as a
consequence of cultural dress[3] or dietary abstention from animal
products.[4] Therefore, it is unfortunate to realise that almost all oral
vitamin D or calcium supplements prescribed in the UK contain animal
product. Gelatin is present in most tablet, granular, effervescent and
syrup supplements. High-dose injections of ergocalciferol and alfacalcidol
are gelatin free but the latter may contain ethanol.[5]
Doctors may not be aware of important excipients; other vitamin D
supplements may contain peanut oil.[6] Gelatin is also found in of many
other drugs, such as the shell of propranolol tablets. The British
National Formulary (BNF) does not list gelatin as an excipient. Currently
one would have to consult the summary of product characteristics (SPC) to
obtain this information.
Our practice has now changed in light of this. Some of our patients
still choose to take a supplement for health benefits; many, once informed
of the gelatin content, do not. Most patients would prefer to be informed
if their medication contained beef or pork products.[7]
We would welcome clearer excipient labelling in formularies, to aid
prescribing for those who wish to abstain from animal products.
References:
1. H A Bischoff-Ferrari, B Dawson-Hughes, H B Staehelin, J E Orav, A
E Stuck, R Theiler, J B Wong, A Egli, D P Kiel, and J Henschkowski. Fall
prevention with supplemental and active forms of vitamin D: a meta-
analysis of randomised controlled trials. BMJ 2009; 339: b3692
2. Meyer G, Köpke S. Information on harm is missing. BMJ
2009;339:b4395
3. Glerup et al. Commonly recommended daily intake of vitamin D is
not sufficient if sunlight exposure is limited. J Intern Med (2000) vol.
247 (2) pp. 260-8
4. Holick. Vitamin D deficiency. The New England journal of medicine
(2007) vol. 357 (3) pp. 266
5. No authors listed. The Electronic Medicines Compendium. 2009.
http://emc.medicines.org.uk
6. Dixon et al. Did you know this medicine has peanut butter in it,
doctor?. Arch Dis Child (2007) vol. 92 (7) pp. 654
7. S P Sattar. Patient and Physician Attitudes to Using Medications
with Religiously Forbidden Ingredients. The Annals of Pharmacotherapy:
Vol. 38, No. 11, pp. 1830-1835.
Competing interests:
None declared
Competing interests: No competing interests
The effects of vitamin D supplementation on muscle strength, function
and balance and its efficacy on accidental fall prevention have been
subject of a number of systematic reviews over the last years.
We have read the meta-analysis by Bischoff-Ferrari et al. (1) with
astonishment since there is no consideration of potential side effects of
vitamin D and its toxicity. Thus, the meta-analysis is another piece of
evidence indicating that harms of treatment options are regularly
underreported, even when the information is accessible in primary studies
(2). Omission of a treatment’s side effects, however, leads to an
overoptimistic estimate of the expected benefits and does not provide the
necessary information for informed decision making. Balancing the pros and
cons of a treatment is especially important in preventive options since
patients and consumers are confronted with decisions on long-term
treatment targeting events that might (or might not) occur in the remote
future. Coverage of adverse events has long been a key feature of good
clinical reporting which has also been recognised by the recent PRISMA
statement (3).
Vitamin D and its analogues are not harmless. Gastrointestinal symptoms
and renal disease have been reported (4). Bischoff-Ferrari and colleagues
are therefore requested to prepare an amendment on the side effects and
toxicity of vitamin D. Since randomised controlled trials often do not
adequately reflect the underlying risk-benefit profiles of treatment
options (5) inclusion of observational evidence might be necessary in
order to enhance judgement of the applicability of the findings.
As long as proper information on harm is missing, the meta-analysis by
Bischoff-Ferrari (1) does not allow for an unbiased and objective
judgement of the balance between risks and benefit of vitamin D.
Meta-analyses defining the objective of their review as only assessing
efficacy of an intervention rather than assessing efficacy and safety
should be regarded as outdated.
References
(1) Bischoff-Ferrari HA, Dawson-Hughes B, Staehelin HB, Orav JE, Stuck AE,
Theiler R, Wong JB, Egli A, Kiel DP, Henschkowski J. Fall prevention with
supplemental and active forms of vitamin D: a meta-analysis of randomised
controlled trials. BMJ 2009 Oct 1; 339: b3692. doi: 10.1136/bmj.b3692
(2) Papanikolaou PN, Ioannidis JP. Availability of large-scale evidence on
specific harms from systematic reviews of randomized trials. Am J Med
2004; 117: 582-9
(3) Liberati A, Altman DA, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis
JPA, Clarke M, Devereaux PJ, Kleijnen J, Moher D. The PRISMA Statement for
Reporting Systematic Reviews and Meta-Analyses of Studies That Evaluate
Health Care Interventions: Explanation and Elaboration. Ann Intern Med
2009; 151: W-65-94
(4) 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; (2):
CD000227
(5) Gartlehner G. Assessment of adverse effects and applicability- two
areas not (yet) covered adequately in Cochrane reports. Z Evid Fortbild
Qual Gesundhwes 2008; 102: 497-502
Competing interests:
None declared
Competing interests: No competing interests
Bischoff-Ferrari and colleagues report that vitamin D supplementation
(700-1000 IU daily) may reduce falls by 19% to 26% amongst individuals
aged 65 years or older. They emphasise that no fall reduction was observed
for serum vitamin D concentrations below 60 nmol/l. We agree it is
important to identify low vitamin D levels in this population group as
this may result in falls and fractures. High prevalence of hypovitaminosis
D in this age group is recognised (1) and is particularly important in
those patients at high risk of falling such as rheumatology patients (2).
This is particularly important in patients who are receiving treatment
with bisphosphonates, as these drugs may cause clinically significant
hypocalcaemia in the presence of low vitamin D (3). The new long-acting
powerful intravenous bisphosphonates, such as zoledronic acid, are
increasingly used because of improved compliance and tolerability. The
risk of hypocalcaemia is greater with the more potent preparations and we
have previously reported the case of a patient with unrecognized
hypovitaminosis D who developed rapid life threatening-hypocalcaemia after
zoledronic acid administration (4).
This experience led us to review our practice regarding patients
receiving intravenous Zoledronic acid. Over the last year 43 patients
received intravenous Zolendronic acid: 65% female, mean age 73.7 years,
diagnosis: 58% osteoporosis, 40% Paget’s disease, 2% osteogenesis
imperfecta. Only 32.5% of patients had serum vitamin D levels measured
before treatment. Of those tested, 35.7% had low vitamin D levels (<20
ng/ml). Interestingly, all patients with low vitamin D levels had already
been prescribed oral vitamin D supplementation. Of the patients with
adequate levels, serum vitamin D (64.3%), 44.4% were on oral vitamin D
supplements and 55.6% were not.
Our data suggests that there is no correlation between prescription
of vitamin D supplementation and serum levels of vitamin D. The
explanation for this is likely to be multifactorial, including lack of
patient compliance, and prescription of inadequate preparation or dose of
vitamin D supplements. It is important that clinicians recognise this and
ensure that patients have fully replenished vitamin D stores prior to
intravenous bisphosphonate therapy. It is also important to remember to
consider the dose, preparation, absorption and compliance with vitamin D
replacement (5).
1. Holick MF. High prevalence of vitamin D inadequacy and
implications for health. Mayo Clin Proc 2006;81:353-73.
2. Mouyis M, Ostor AJK, Crisp AJ, Ginawi A, Halsall DJ, Shenker N et
al. Hypovitaminosis D among rheumatology outpatients in clinical practice.
Rheumatology 2008;47:1348-51.
3. Whitson HE, Lobaugh B, Lyles KW. Severe hypocalcaemia following
bisphosphonate treatment in a patient with Paget’s disease of the bone.
Bone 2006;39:954-8.
4. Joshi A, Price E, Collins D, Williamson L. Comment on:
Hypovitaminosis D among rheumatology patients in clinical practice.
Rheumatology 2009;48:203-4.
5. Ryan PJ. Vitamin D therapy in clinical practice. One dose does not
fit all. Int J Clin Pract 2007;61:1894-9.
Competing interests:
None declared
Competing interests: No competing interests
In 2003, Latham et al. performed a systematic review and meta-
analysis of the effect of vitamin D or its metabolites on falls, and
concluded there was no effect [1]. Subsequently, there have been at least
eight further meta-analyses of trials of vitamin D and its
metabolites/analogues and falls, including that recently reported by
Bischoff-Ferrari et al. [2]. Of the nine reviews, four reported a positive
effect of vitamin D or its metabolites/analogues on falls, and five no
effect, or benefits limited to certain subgroups. The differences between
the conclusions of the reviews often depend on selection of studies for
inclusion in the primary and secondary analyses, and the grounds for
exclusion sometimes seem capricious. For example, a negative study [3] was
excluded from the most recent meta-analysis [2] because patients were “in
an unstable health state”, even though this was not specified in the
inclusion criteria and this label could apply to many of the people who
will be treated based on the conclusions of this meta-analysis.
There are relatively few randomised controlled trials (RCT) on the
topic of vitamin D and falls. In the most recent review [2], 10 studies of
3050 participants were included in the primary analysis and 17 studies in
the sensitivity analyses [2]. The ratio of the number of RCTs to meta-
analyses (RCT:Met) [4] for vitamin D and falls is 17:9 or 1.9.
Similarly, there are at least 14 published meta-analyses on vitamin D
and fracture prevention, but only 22 RCTs were included in the most recent
Cochrane review [5]. Again, conclusions differ substantially between
reviews. Most reviews reported no effect of vitamin D overall, but
positive effects on fracture incidence were reported in some subgroups,
including those receiving high dose vitamin D, or calcium and vitamin D co
-supplementation, and individuals in residential care. The RCT:Met for
vitamin D and fractures is 22/14 or 1.6.
We continue to be concerned that, in some areas of medicine, too much
emphasis is being placed on analysis and re-analysis of a limited amount
of trial-level data by meta-analysis, rather than on the design and
conduct of informative RCTs in relevant populations [4]. The conflicting
conclusions generated by this practice have the potential to cause
considerable confusion among health practitioners. We suggest that no
further meta-analyses are conducted on the topics of vitamin D and falls
or fractures until additional adequately powered RCTs are performed,
thereby permitting a meaningful re-evaluation.
References:
1. Latham NK, Anderson CS, Reid IR. Effects of vitamin D
supplementation on strength, physical performance, and falls in older
persons: a systematic review. J Am Geriatr Soc 2003;51:1219-26.
2. Bischoff-Ferrari HA, Dawson-Hughes B, Staehelin HB, Orav JE, Stuck
AE, Theiler R, et al. Fall prevention with supplemental and active forms
of vitamin D: a meta-analysis of randomised controlled trials. BMJ
2009;339:b3692.
3. Latham NK, Anderson CS, Lee A, Bennett DA, Moseley A, Cameron ID.
A randomized, controlled trial of quadriceps resistance exercise and
vitamin D in frail older people: the Frailty Interventions Trial in
Elderly Subjects (FITNESS). J Am Geriatr Soc 2003;51:291-9.
4. Bolland MJ, Grey A, Reid IR. The randomised controlled trial to
meta-analysis ratio: original data versus systematic reviews in the
medical literature. N Z Med J 2007;120:U2804.
5. 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:CD000227.
Competing interests:
None declared
Competing interests: No competing interests
Re:Re:Re:Fall prevention with Vitamin D. Clarifications needed.
We offer the following response to the Drs Rosen and Gallager and the
IOM panel:
We agree with Drs Rosen and Gallagher that no additional meta-
analyses are needed, but argue that the meta-analysis commissioned by the
IOM (never peer reviewed individually), has a misleading conclusion that
the evidence of vitamin D regarding fall prevention is inconsistent and
therefore insufficient. The IOM overall analysis of 12 RCTs (n = 14,101)
showed a significant benefit of vitamin D on fall prevention (OR = 0.89;
95% CI 0.80-0.99), as did the majority of their subset analyses, clearly
supporting the use of vitamin D in the prevention of falling. The set of
analyses which showed no benefit were based on only 4 studies, which
cannot be considered reliable indicators of true treatment efficacy, as
these trials either used low dose vitamin D[1], had less than 50%
adherence[2], had a low-quality fall assessment[3] or used one large bolus
dose of vitamin D among seniors in unstable health[4].
Drs Rosen and Gallagher claim that our peer-reviewed meta-analysis
that selected trials with a double-blinded design and high quality fall
ascertainment may have been flawed regarding the choice of 8 trials and
the method chosen to explain heterogeneity. Regarding the latter, as
pointed out in our response to Dr. DaSilva, both the Broe[5] and the
Graafmans[1] trial met these criteria. We re-analyzed our data to account
for the stochastic dependencies (correlations) between the corresponding
risk ratios in the multiple dosing trial by Broe et al. as suggested by
the IOM and (1) confirmed an overall benefit of vitamin D based on these 8
high quality trials, and (2) confirmed that vitamin D dose is important as
established by the revised Figure 3 which clearly shows a differential
effect of vitamin D in trials that tested a dose of less than 700 IU
vitamin D per day compared with trials that tested a dose of 700 IU per
day or higher. In our re-analysis, when treatment was the only predictor
(regardless of dose level), there was a significant reduction in the odds
of falling based on our primary analysis of the same 8 trials: OR=0.73
[.62, .87]; p=.0004. When the model was expanded to capture the impact of
both high dose and low dose treatment, high dose vitamin D (700 to 1000
IU vitamin D per day) reduced the odds of falling (OR=0.66 [.53, .82]
p=.0002), while low dose vitamin D did not (OR=1.14 [.69, 1.87];
p=.61).
Notably, our result based on 8 high-quality trials included most of
the same studies selected by the IOM in their subgroup analysis of 6 RCTs
(N = 1,833). In both of these analyses, falls and ascertainment were
adequately defined and a significant reduction in falls (OR 0.79 (95% CI
0.65-0.96)) was observed. Even the IOM included the Graafmans trial, as
the trial had a high quality fall assessment throughout the 28 week
observation time and the trial design was double-blinded.
Finally, in the original publication of our peer-reviewed meta-
analysis, we performed a sensitivity analysis including 15 trials of any
study design and fall assessment quality (n = 17,786) and documented a non
-significant 7% fall reduction with vitamin D (RR = 0.93; 95% CI 0.87-
1.01). Notably, variation among the 15 trials was larger than expected (Q-
test: p = 0.009), which could be explained by dose (700 IU + /day; n =
17,281; pooled RR was 0.92 (95% CI, 0.85-1.00)) and further among trials
that tested a higher dose by trial quality (Q-test: p = 0.005).
In summary, we maintain that the evidence of the vitamin D effect on
falls has been misinterpreted by the IOM disregarding the overall benefit
across all trials and the differential efficacy by dose among trials that
tested true treatment efficacy in a double-blinded design with high
quality fall assessment. Thus, given the available evidence today,
vitamin D supplementation for fall prevention should not be delayed as a
recommendation among the senior population. This suggestion is in line
with the Agency for Healthcare Research and Quality (AHRQ) for the U.S.
Preventive Services Task Force[6] , the 2010 American Geriatric
Society/British Geriatric Society Clinical Practice Guideline[7], and to
the 2010 assessment by the IOF[8], all 3 of which identified vitamin D as
an effective intervention to prevent falling in older adults based on
their review of the same evidence.
References:
1. Graafmans WC, Ooms ME, Hofstee HM, Bezemer PD, Bouter LM, Lips P: Falls
in the elderly: a prospective study of risk factors and risk profiles. Am
J Epidemiol 1996; 143(11): 1129-36.
2. Grant AM, Avenell A, Campbell MK, et al.: Oral vitamin D3 and calcium
for secondary prevention of low-trauma fractures in elderly people
(Randomised Evaluation of Calcium Or vitamin D, RECORD): a randomised
placebo-controlled trial. Lancet 2005; 365(9471): 1621-8.
3. Trivedi DP, Doll R, Khaw KT: Effect of four monthly oral vitamin D3
(cholecalciferol) supplementation on fractures and mortality in men and
women living in the community: randomised double blind controlled trial.
BMJ 2003; 326(7387): 469.
4. Latham NK, Anderson CS, Lee A, Bennett DA, Moseley A, Cameron ID: A
randomized, controlled trial of quadriceps resistance exercise and vitamin
D in frail older people: the Frailty Interventions Trial in Elderly
Subjects (FITNESS). J Am Geriatr Soc 2003; 51(3): 291-9.
5. Broe KE, Chen TC, Weinberg J, Bischoff-Ferrari HA, Holick MF, Kiel DP:
A higher dose of vitamin d reduces the risk of falls in nursing home
residents: a randomized, multiple-dose study. J Am Geriatr Soc 2007;
55(2): 234-9.
6. Michael YL, Whitlock EP, Lin JS, Fu R, O'Connor EA, Gold R: Primary
care-relevant interventions to prevent falling in older adults: a
systematic evidence review for the u.s. Preventive services task force.
Ann Intern Med 2011; 153(12): 815-25.
7. American Geriatric Society and British Geriatric Society Guidelines on
Fall Prevention in older Persons.
http://www.americangeriatrics.org/files/documents/health_care_pros/Falls...
2010.
8. Dawson-Hughes B, Mithal A, Bonjour JP, et al.: IOF position statement:
vitamin D recommendations for older adults. Osteoporos Int 2010; 21(7):
1151-4.
Competing interests: No competing interests