Reviews/Osteoporosis Fracture Prevention/Calcium and vitamin D fracture prevention
Calcium And Vitamin D · ostfrac_13 · v2
The question is whether taking calcium and vitamin D together lowers the chance of breaking a bone in older people. Across the primary trials, the direction of effect leans in favour of the supplements, but when each trial is analysed as everyone was originally assigned to treatment, the results do not reach statistical reliability: the largest trial [5] and the population trial [2] were not significant, and the confirmatory Decalyos II result [1] was only a trend. A clear reduction appears only in a secondary analysis that removed people who stopped taking their pills [5] and in a subgroup of long-term users [6], both from the same single trial. One further trial [3] showed fewer falls but did not measure fractures at all. Because a consistent favourable direction is paired with an almost uniform lack of firm significance in the main analyses, the panel judged the claim plausible but not demonstrated.
Four eligible primary RCTs test combined calcium plus vitamin D against fracture in older adults, and the point estimates consistently favour treatment: [5] reports a hip-fracture hazard ratio of 0.88 (95% CI 0.72 to 1.08), [2] an adjusted hazard ratio of 0.83 (95% CI 0.61-1.12) for any fracture, and [1] a hip-fracture relative risk of 1.69 (95% CI 0.96, 3.0) for placebo versus active treatment. Statistical significance is reached only in a compliance-censored analysis of [5], where the hip-fracture hazard ratio fell to 0.71 (95% CI 0.52 to 0.97), and in the subgroup of [6], where five or more years of supplementation gave a hazard ratio of 0.62 (95% CI 0.38-1.00). [3] adds only indirect support, reporting a relative risk of falling of 0.48 (95% CI 0.26-0.90) with no fracture endpoint. The two significant fracture results come from the same WHI trial [5] and [6]. The contested point: The evidence divides on whether a reproducible favourable direction amounts to a demonstrated effect, given that the flagship intention-to-treat analyses [2] and [5] are null and [1] is a non-significant trend. The only significant fracture reductions come from a compliance-censored analysis [5] and a subgroup [6] of a single trial, so independence and robustness are the contested points.
All 18 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from Osteoporosis Fracture Prevention, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
The largest and most rigorous trial (WHI [5], N=36,282) was null in ITT for hip fracture (HR 0.88, CI 0.72-1.08). Significance appears only after censoring non-adherers (0.71, 0.52-0.97) — a per-protocol analysis prone to healthy-adherer bias, not a valid demonstration of effect.
Decalyos II [1] reports RR 1.69 with CI 0.96-3.0 crossing 1 — a non-significant trend. Reading it as evidence that the combination 'reduces' hip fracture overstates a result the trial did not establish.
Where it remains uncertain. The largest trial [5] and the population trial [2] were null on intention-to-treat and Decalyos II [1] a trend (RR 1.69, 95% CI 0.96, 3.0); the two significant results come from the same WHI trial [5] and [6].
Aleatory
Fracture is a relatively rare event, so even the large trials produce intervals that cross 1, e.g. [5] hip fracture 0.72-1.08 and [1] 0.96-3.0, leaving the true effect imprecise.
Epistemic
Eligible primary data are postmenopausal and institutionalised women [1][2][5][6]; there is no male or community-dwelling male evidence to cover 'older adults' as worded.
Inferential
The significant hip-fracture result in [5] (0.71, 0.52-0.97) depends on censoring non-adherers, an analysis prone to healthy-adherer bias rather than intention-to-treat inference.
The load-bearing papers from a corpus of 18, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Confirmatory RCT in institutionalised elderly women shows a hip-fracture trend favouring combined calcium+vitamin D, but the CI crosses 1 (not significant).
“The relative risk (RR) of HF in the placebo group compared with the active treatment group was 1.69 (95% CI = 0.96, 3.0)”
Panel auditInternal validity: high · scope of inference: Very elderly (mean 85) institutionalized women with vitamin D insufficiency and secondary hyperparathyroidism.Key limitations: CI crosses 1.0 · institutionalized/frail only · BMD subgroup of 114 only
Population-based RCT shows favourable direction for any fracture but explicitly does not reach statistical significance.
“the risk of any fracture decreased in the vitamin D and calcium group by 17% [adjusted hazard ratio (aHR) = 0.83; 95% confidence interval (CI) 0.61-1.12]”
Panel auditInternal validity: medium · scope of inference: General population of postmenopausal women aged 65-71 in northern Finland (high latitude).Key limitations: no placebo control · no result statistically significant · relatively young/healthy cohort
Indirect mechanistic support: combined supplementation reduces falls, a fracture precursor, though no fracture endpoint is reported.
“Relative risk of falling in the groups supplemented with vitamin D was 0.48 (95% CI 0.26-0.90) compared with controls.”
Panel auditInternal validity: medium · scope of inference: Elderly women in orthogeriatric rehabilitation after a hip fracture (secondary prevention).Key limitations: fracture not an endpoint · small sample · surrogate outcomes only
3 of 18 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
Consistently favourable effect direction across primary RCTs is offset by near-uniform absence of ITT significance and reliance on single-trial subgroup/per-protocol analyses, placing the claim in the Uncertain band.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
Adequately powered ITT fracture trial. Every eligible primary trial is null or a trend on intention-to-treat [2][5][1]; significance rests on secondary analyses. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A large randomised trial of combined calcium plus vitamin D powered for hip and total fracture, analysed by intention-to-treat as the primary endpoint.
Open question: Does combined calcium and vitamin D reduce fracture risk on intention-to-treat analysis, not only after censoring non-adherers? Why it is still open: The significant hip-fracture estimate in [5] (0.71, 0.52-0.97) appears only after compliance censoring, while the ITT estimate was 0.88 (0.72-1.08).
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19