Analysis of road traffic crashes data: Applying the lorenz curve and associated gini index

ABSTRACT

It has been suggested that the summary index of the Lorenz curve and the Gini index should be used to characterize the exposure-disease association, rather than relative and attributable risks. Nevertheless, the sampling behavior of the Gini index in epidemiological settings, and the relationship between the Lorenz curve and the usual indices of risk, need to be more deeply understood. The aim of this paper was to estimate the odds and to predict the probability of a driver responsible for a Road Traffic Crash [RTC] and road traffic injury on the selected risk factors. And further to assess how these risk factors vary by gender and visualize the exposure of risk factors by applying the Lorenz curve and the Gini index. This is a prospective case-control study which included 217 cases and 217 control drivers. A structured questionnaire was designed to collect demographic information like age, gender, marital status, occupation, residential area, housing conditions, driving experience, type of car, frequency of seatbelt usage, excessive speed and other violations. The Lorenz curve and the associated Gini index are applied for characterizing and testing road traffic injuries stratified by gender. We also showed that the Gini index can itself be interpreted as a probability related to temporal clustering. In conclusion, with the increasing demand for computer technology the proposed method is well-suited for analyzing RTC data. Furthermore, the Lorenz curve and the Gini index can complement the information provided by relative and attributable risks