Hidden patterns among the fatally injured pedestrians in an Iranian population: application of categorical principal component analysis (CATPCA)

Background Identifying hidden patterns and relationships among the features of the Fatal Pedestrian Road Traffic Injuries (FPRTI) can be effective in reducing pedestrian fatalities. This study is thus aimed to detect the patterns among the fatally injured pedestrians due to FPRTI in East Azerbaijan province, Iran. Methods This descriptive-analytic research was carried out based on the data of all 1782 FPRTI that occurred in East Azerbaijan, Iran from 2010 to 2019 collected by the forensic organization. Categorical Principal Component Analysis (CATPCA) was performed to recognize hidden patterns in the data by extracting principal components from the set of 13 features of FPRTI. The importance of each component was assessed by using the variance accounted for (VAF) index. Results The optimum number of components to fit the CATPCA model was six which explained 71.09% of the total variation. The first and most important component with VAF = 22.04% contained the demographic and socioeconomic characteristics of the killed pedestrians. The second-ranked component with VAF = 12.96% was related to the injury type. The third component with VAF = 10.56% was the severity of the injury. The fourth component with VAF = 9.07% was somehow related to the knowledge and observance of the traffic rules. The fifth component with VAF = 8.63% was about the quality of medical relief and finally, the sixth component with VAF = 7.82% dealt with environmental conditions. Conclusion CATPCA revealed hidden patterns among the fatally injured pedestrians in the form of six components. The revealed patterns showed that some interactions between correlated features led to a higher mortality rate.

Automated summary

CATPCA analysis of data on 1782 fatal pedestrian road traffic injuries in East Azerbaijan, Iran revealed six main components: demographic characteristics, injury type, injury severity, knowledge and observance of traffic rules, quality of medical relief, and environmental conditions. These patterns suggested that interactions between correlated features contributed to a higher mortality rate.