Fifteen articles describing 18 studies were included (n=522,507): 4 RCTs (n=34,991), 6 prospective (n=119,707) and 2 retrospective (n=125,636) cohort studies, and 6 case-control studies (n=242,173).
Statin users versus non-users (18 studies): there was no statistically significant difference between the groups when the RCTs were pooled, and no significant heterogeneity (OR 1.03, 95% CI: 0.91, 1.16; 4 studies). When cohort studies, case-control studies and all studies were pooled, fracture rates were significantly lower in statin users: OR 0.77 (95% CI: 0.59, 1.00; 8 studies), OR 0.62 (95% CI: 0.45, 0.85; 6 studies) and OR 0.77 (95% CI: 0.66, 0.90; 18 studies), respectively. There was statistically significant heterogeneity for these three analyses (p<0.01).
Statin users versus other lipid-lowering drug users (3 studies): there was no statistically significant difference between the groups (OR 0.96, 95% CI: 0.67, 1.37).
Although Egger’s test was not statistically significant for most analyses, Begg’s funnel plot suggested possible publication bias. Further investigation suggested that small studies were more likely to be published if they were positive.
The results of subgroup and meta-regression analyses were also reported in the review.