Fourteen studies of calcium supplements (including 4 RCTs with 3,638 participants, and 3 non-randomised trials with 141 participants), 18 studies of dietary calcium and hip fracture (no RCTs), and 5 studies of dietary calcium and other fracture sites (no RCTs).
A further sixteen observational hip fracture studies were included in this analysis. Eleven were case-control studies with 12,465 participants (4,131 cases and 8,334 controls) and 5 cohort studies with 28,511 participants who experienced 915 fractures.
For calcium supplements, the 4 RCTs reported RRs between 0.3 and 0.7, a reduced fracture risk among women randomised to receive calcium supplements. The 3 non-randomised trials found a lower risk of new vertebral fractures in women given calcium supplements than in untreated women. The observational studies had inconsistent findings: the RRs varied between 0.3 and 2.0, with only one of the studies (with calcium given for 3 or more years) statistically significant.
Pooling the results of all 16 studies of dietary calcium gave an OR for hip fracture of 0.96 (95% CI: 0.93, 0.99) per 300 mg/day increase in dietary calcium which was statistically significant.
The pooled OR in the subgroup of 5 cohort studies was 0.96 (95% CI: 0.91, 1.02), p for heterogeneity = 0.44 which was not statistically significant.
The pooled OR for all 16 studies for 1000 mg of calcium/day was 0.88 (95% CI: 0.80-0.97).