Fig. 3

The nonlinear correlation analysis for SAR with a fragility fracture probability through the RCS model. After adjusting for confounding factors, the data were fitted by a restricted cubic spline Cox proportional hazards regression model, and the model was constructed with 4 knots at the 5th, 35th, 65th, and 95th percentiles of the SAR (reference was the 5th percentile). The solid lines indicate HRs, and the shadow shapes indicate 95% CIs. The RCS model demonstrated a J-shaped dose-dependent association between the SAR and a fragility fracture endpoint event (P for overall < 0.001, P for nonlinear = 0.866). RSC analysis showed that the risk of fragility fractures progressively increased with increased SAR and vice versa