R software 3

The data were analyzed using R Software 3.6.3. Quantitative variables are presented as means and standard deviations, and qualitative variables as frequencies and percentages. We also performed a bivariate analysis to identify possible factors associated with Injury mechanism and status and admission as well as to assess differences between data from years 2007 and 2017. For this purpose the Kruskal–Wallis test, Chi-square test, and Spearman correlation coefficient were calculated accordingly.
In order to explore more deeply the factors associated with place at death, we use Classification and Regression Trees⁴⁸ as implemented in Rpart R package⁴⁹ as an exploratory tool to uncover complex interactions between selected covariates and the place at death. This method has been used before successfully by one of the authors to explore associations on outcomes of interest in the context of observational studies (see for example^{50 –52 (link)}).
Finally, georeferencing maps were done using R Software 3.6.3 and Google Maps. Coordinates of homicide and trauma hospitals were obtained by Google Maps and introduced in R Software 3.6.3. With the data, a heat map on homicide cases, rates per locality, and the rate of arrival to the emergency room with vital signs per locality were graphed.

Data analyses were performed using RevMan software 5.3 (Cochrane Collaboration, Oxford, UK) and the “meta” package in R software 3.6.0 (R Foundation for Statistical Computing, Vienna, Austria) [13 ]. Odds ratios (ORs) were used to compare dichotomous variables. The natural logarithm of HRs and the corresponding standard error were used for the meta-analysis of survival data [14 (link)]. We tested heterogeneity by Cochran’s Q test and Higgins I² statistic. A P value by Q test of less than 0.10 and I² > 50% indicated existence of statistically significant heterogeneity. A random-effects model was used for outcomes when heterogeneity existed; otherwise, a fixed-effects model was used. Sensitivity analysis was conducted by sequentially omitting each individual study in order to evaluate the stability of the synthetic results. We also assessed publication bias using contour-enhanced funnel plot [15 ] with a P value lower than 0.05 for Egger’s test indicating significant statistical publication bias [16 ]. All results were reported with 95% confidence intervals (CIs), and a two-sided P value of < 0.05 was considered statistically significant.

Statistical analysis was performed using R Software 3.6.0 (R Foundation for Statistical Computing, Vienna, Austria). The reliability of rating screw positioning according to Gertzbein and Robins was compared between the observers, and for each observer, between both assessments. Likewise, the image quality scale was tested using Intraclass Correlation Coefficients (ICC) for categorical data. Observer agreement was then compared between NoMAR and MAR. The values of intraclass correlation coefficients were rated according to Landis and Koch [18 (link)]: correlation was considered excellent if r > 0.80, good if r = 0.61 to 0.80, fair if r = 0.41 to 0.60, and poor if r ≤ 0.40.