Non-vertebral fracture incidence

Method

Proportion of participants sustaining at least one incident fracture at a skeletal site outside the spine - hip, wrist, humerus, ankle and others - ascertained radiographically or by adjudicated clinical report, pooled as an odds ratio or relative risk. The sites included, and whether hip fractures were counted, are recorded on the edge.

The method is part of the endpoint identity. Studies measuring the same quantity by a different method are held as separate endpoints.

Notes

Kept separate from vertebral fracture incidence for an empirical reason, not a definitional one: in PMID 17308327 the pooled relative risk here fell below 1 while the vertebral estimate sat above 1. Merging them averages a benefit against a possible harm and reports a null describing neither. Ascertainment also differs, since vertebral fractures are largely morphometric and asymptomatic. Distinct again from CAT:outcome/lumbar-spine-bmd and femoral neck bone mineral density, continuous surrogates in g/cm^2 that move earlier and more reliably than the events they predict.