105 Risk prediction of implant loss following implant-based breast reconstruction Table 3. Risk model for implant loss Risk factors OR (95% CI) β coefficient P-value BMI* 1.045 (1.015-1.075) 0.0439 0.003 Current smoker 2.186 (1.584-3.019) 0.7823 <0.001 Previous radiotherapy 1.684 (1.007-2.816) 0.5213 0.047 Prepectoral placement 1.745 (1.205-2.526) 0.5566 0.003 Constant -3.8715 Risk model including four risk factors associated with implant loss within the training cohort. Corresponding ORs (95% CIs), beta regression coefficients and P-values (significant values are denoted in italic) are depicted. BMI indicates body mass index; CI, confidence interval; OR, odds ratio. *Continuous variable, OR calculated per 1 kg/m increment. Figure 2. Validation of the risk prediction model using a calibration plot Comparison between predicted probabilities (x-axis) and the observed probabilities (y-axis, with corresponding 95% CIs) of implant loss in the validation cohort (blue line). The orange line indicates perfect agreement. 7
RkJQdWJsaXNoZXIy MTk4NDMw