|Patient level pooled analysis of 68,500 patients from seven major vitamin D fracture trials in US and Europe
|The vitamin D Individual Patient Analysis of Randomized Trials (DIPART) group
This review concluded that doses of 10 to 20µg of vitamin D were not effective in preventing fractures, but, given together with calcium, they reduced hip fractures, total fractures, and probably vertebral fractures, irrespective of age, sex, or previous fracture status. The authors' conclusions appear to be appropriate for the evidence presented.
To identify those participant characteristics that influence the anti-fracture efficacy of vitamin D or vitamin D plus calcium with respect to any fracture, hip fracture, and clinical vertebral fracture, and to assess the influence of dosage and co-administration of calcium.
MEDLINE, EMBASE and the Cochrane Central Register of Controlled Trials (CENTRAL) were searched, without language restrictions, from January 1966 to July 2008; search terms were reported.
Studies were included provided they were individual or cluster randomised controlled trials, with at least one intervention arm in which vitamin D was given and one arm in which it was not, and included at least 1,000 patients. The primary outcome was any fracture, with hip fracture and clinical vertebral fracture as secondary outcomes.
For most trials, vitamin D dosage was oral D3 ranging from 10 to 20μg per day; several trials added 1,000mg of calcium. The comparators were generally placebo, but also included leaflets, environmental interventions, and 1,000mg of calcium. In the included trials, the mean age was 69.9 years (range 47 to 107) and 14.7% of participants were male. The percentage of patients with a previous hip fracture ranged from 0.5 to 26.2, with a vertebral fracture ranged from 0.2 to 1.5, and with other fracture ranged from 7.4 to 82.8. Trial setting varied, and included general practice, care home, and community; most trials were undertaken in the UK, with others from the USA, Norway, and Denmark.
It was unclear how many reviewers were involved in selecting studies for inclusion.
Assessment of study quality
The authors stated that contradictory data was queried with the participating centres who conducted the trials.
Each corresponding author submitted the anonymised individual patient data in the form used in their original trial.
Methods of synthesis
Trial data were combined using individual patient data meta-analysis. Pooled hazard ratios (HRs) and 95% confidence intervals (CIs) were derived. Logistic regression analysis identified the significant interaction terms and fixed-effect Cox’s proportional hazards models incorporated age, sex, fracture history, and hormone therapy and bisphosphonate use. Analyses were restricted to 36 months as only one trial was of longer duration. All analyses were done on an intention-to-treat basis. Several analyses were performed to test the sensitivity of the results to the exclusion of one, or more, individual trial's data, and to compare vitamin D doses (10μg versus 20μg).
Results of the review
There were 11 eligible trials and seven of these were included in the review (n=68,517 patients, range 1,144 to 36,282); six were individually randomised (one quasi-randomised by birth date) and one was cluster randomised. Four trials were unavailable as their authors were unwilling, or unable, to provide patient level data.
Trials using vitamin D with calcium showed reduced overall risk of fracture (HR 0.92, 95% CI 0.86 to 0.99), reduced risk of hip fracture (HR 0.83, 0.69 to 0.99) across all trials and in those trials that used 10µg of vitamin D with calcium (HR 0.74, 95% CI 0.60 to 0.91).
For vitamin D alone in daily doses of 10µg or 20µg, no significant effects were found.
No interaction was found between fracture history and treatment response, nor any interaction with age, sex, or hormone replacement therapy.
A sensitivity analysis, incorporating published aggregate data from the unavailable trials, produced broadly similar results in favour of the intervention.
This individual patient data analysis indicated that vitamin D alone in doses of 10 to 20µg was not effective in preventing fractures, but given together with calcium it reduced hip fractures, total fractures, and probably vertebral fractures, irrespective of age, sex, or previous fracture status.
The review question and inclusion criteria were clear. Three databases were searched without language restrictions, which reduces the chances of bias and omission of relevant studies. It does not appear that attempts were made to locate unpublished and ongoing trials and some trials may have been missed. Inclusion was restricted to trials of at least 1,000 patients (36 smaller trials were excluded). The authors stated that discordant data were checked with participating centres, but no further details were provided. The data were analysed using appropriate statistical techniques. The rationales for the subgroup and sensitivity analyses were clear. Although the data were available from only seven of 11 eligible trials, those data obtained represented 84% of the eligible patients and sensitivity analyses incorporating aggregate published data from the unavailable trials produced similar results, indicating that the findings are likely to be robust.
The authors' conclusions were appropriate for the evidence presented.
Implications of the review for practice and research
Practice: The authors recommended vitamin D at doses of at least 10μg per day, combined with 1,000mg of calcium, and in high-risk patients these should be supplemented by bisphosphonates or other anti-osteoporotic drugs.
Research: The authors stated that further studies were required to assess whether the route of administration, long time interval between doses, nature of the vitamin D preparation, or type of patient cohort (older age and greater fracture risk) was responsible for the failure of standalone vitamin D in the prevention of fractures. Additional trials were also required to assess the outcomes associated with daily vitamin D at higher doses without calcium.
National Institutes of Health, US Department of Health and Human Services, National Heart, Lung and Blood Institute, contract numbers NO1WH22110, 24152, 32100-2, 32105-6, 32108-9, 32111-13, 32115, 32118-32119, 32122, 42107-26, 42129-32, and 44221.
The vitamin D Individual Patient Analysis of Randomized Trials (DIPART) group. Patient level pooled analysis of 68,500 patients from seven major vitamin D fracture trials in US and Europe. BMJ 2010; 340:b5463
Other publications of related interest
Correction to odds ratio for hip fractures in trials of vitamin D plus calcium cited in: BMJ 2011; 343:d5245.
Subject indexing assigned by NLM
Aged; Aged, 80 and over; Bone Density Conservation Agents /therapeutic use; Calcium /therapeutic use; Dietary Supplements; Drug Therapy, Combination; Europe; Female; Fractures, Bone /prevention & Hip Fractures /prevention & Humans; Male; Randomized Controlled Trials as Topic; Secondary Prevention; Spinal Fractures /prevention & United States; Vitamin D /therapeutic use; control; control; control
Date abstract record published
This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.