*2.5. Statistical Analysis*

A descriptive analysis of the sample was made, where the continuous variables were expressed by means and standard deviations (SD), and categorical variables expressed through frequencies and percentages. In addition, a bivariate analysis was carried out using the t-student or chi-square test according to the nature of the variables (continuous or categorical, respectively).

In order to compare post-surgical complications requiring re-intervention, a survival cox regression model was applied. The dependent variable in the model was a dichotomous variable indicating whether reoperation was required (yes/no) and the time until reoperation. The following covariates were included in the model: type of nail (Gamma3/InterTAN), type of fracture (Basicervical/Other), age and sex. Additionally, due to concerns related to the rule of ten outcome events per predictor variable, the effect was also estimated using full propensity score matching with the MatchIt package, which is particularly suitable for modeling rare events.
