Exposure to violence, victimization differences and maxillofacial injuries in a Brazilian state capital: a data mining approach

Abstract

To identify victimization differences and evaluate maxillofacial injuries among victims of interpersonal violence in a Brazilian state capital using a data-mining approach. A cross-sectional study analyzed 1946 reports of physical violence registered in an Institute of Forensic Medicine and Dentistry. Sociodemographic characteristics of the victims, mechanisms of violence, time of occurrence and patterns of maxillofacial injuries were evaluated. Data analyses included descriptive statistics, Pearson’s chi-square test and decision tree analysis using the Chi-square Automatic Interaction Detector (CHAID) algorithm. Most victims were male (54.6%) and the majority of events occurred during nighttime (53.7%). Maxillofacial injuries affecting more than one region accounted for the largest proportion of cases (33.1%). The final model of the CHAID decision tree demonstrated multilevel interaction among factors through stepwise pathways to explain the victimization differences between men and women. The profile of victimization could be explained by the sociodemographic characteristics of victims, such as employment status (p < 0.001), marital status (p = 0.008) and age group (p < 0.001), as well as by the type of maxillofacial trauma (p = 0.007). Gender is an important factor related to violence exposure and maxillofacial injuries. Young adult men were more likely to exhibit severe maxillofacial injuries compared to women. It is necessary to challenge interpersonal violence through awareness, prevention and education programs. Efforts to minimize the impact of violence on the population health and well-being require specific guidelines in Brazil.