Predictive models for the assessment of bodily harm

The Spanish scale to quantify or qualify bodily harm resulting from any unintentional traffic accident prior to 1 January 2016 is established by Royal Legislative Decree (RDL) 8/2004. This scale assigns points to the sequelae, which are converted into Euros using a table that is updated annually. The objective of this study is to develop a predictive model of sequelae points that will enable the estimation of compensation a short time after the accident. This will facilitate the calculation of the money reserve and rapid access to compensation for the injured party. To conduct this study, we developed a database with information from 999 individuals who had suffered car crash injuries which were evaluated according to the scale contained in RDL 8/2004 for medical experts. Predictive models based on logistic regression models were designed on this database. To choose the best model, we calculated Mallow's Cp. The use of hurdle models made it possible to predict the points received by an injured party within a relatively short period of time after the accident. Once these points are known, it is a simple matter to calculate the corresponding compensation. The prediction models developed provide an easy way to predict the compensation to be awarded to the injured party. These models use days of hospitalization, sex, age and the results of international scales based on the Abbreviated Injury Scale. These variables can be used soon after the occurrence of the crash.