Cox regression multivariate analysis was subsequently conducted using forward likelihood ratio on SPSS IBMM, New York, NY, USA).
Multivariate analysis was subsequently performed using the Cox proportional hazards model for all significant variables examined in the univariate analysis.
However, as the risk profile was different between these two populations, a multivariate analysis was subsequently performed for verification.
A multivariate analysis was subsequently conducted including factors found to differ between cases and controls using a significance level of p ≤ 0.05 of all results in bivariate analyses.
Multivariate analysis was subsequently carried out using stepwise Cox proportional hazards modelling for OS; best model was constructed using automated methods.
Subsequently, multivariate analysis of logistic regression was conducted to study association of mortality and variables such as AIS, NISS, ISS, sex, age, height and aetiology (intentional vs. non-intentional).
In each of the selected major subgroups, survival analysis including Kaplan-Meier, univariate and multivariate analysis, is performed on subsequently more stratified groups based on gene expression subtypes (basal-like, HER2-enriched, luminal A, luminal B, normal-like) [24], [29], ER-status, LN-status, histological grade, and treatment status (untreated or tamoxifen treated tumors).
Subsequently, multivariate analysis was performed, taking the most important variables into account.
Subsequently, multivariate analysis will allow the determination of the important confounders affecting the diagnostic performance by the method of backward elimination.
