Comparing international classification of diseases – Based injury severity measures for predicting mortality in a Greek adult trauma population

Abstract text (incl. references and figure legends)

Introduction: The International Classification of Diseases (ICD) is increasingly being used for performance monitoring in trauma patients. However, there is an ongoing debate on the preferrable approach between empirically derived metrics calculated directly from the ICD codes and mapped metrics after being converted to Abbreviated Injury Scale (AIS). The purpose of this study was to compare the performance of ICD-based Injury Severity Scores (ICISS) and AIS mapped scores generated by the ICD Program for Injury Categorisation in R software (ICDPIC–R) to predict mortality in a Greek trauma population.

Material & Methods: This retrospective single-center cohort study enrolled adult trauma patients (≥16 years) hospitalized between January 2015 and December 2019. For each patient were calculated two different ICISS values: multiplicative injury (ICISS) and single worst injury (SWI). Based on two mapping algorithms (Trauma Quality Improvement Program and National Inpatient Survey) provided by ICDPIC-R, three AIS-based severity scores were derived: Maximum AIS (MaxAIS), Injury Severity Score (ISS) and New Injury Severity Score (NISS). Models" prediction was performed using discrimination and calibration statistics.

Results: A total of 35192 patients were included in the study. Median age was 61 ± 21 years and mortality rate was 2,1%. For the ICISS measures, the area under the curve ranges between 0,857 (95% CI 0.845–0.868) and 0,855 (95% CI 0.843–0.866), while AIS-based measures range between 0,866 (95% CI 0.853–0.878) and 0,837 (95% CI 0.820–0.854). Although Hosmer-Lemeshow chi-square statistic was slightly lower for AIS-based measures, visual inspection of calibration curves demonstrated comparable results to those from ICISS measures.

Conclusions: No meaningful differences were found in performance between ICISS and AIS - mapped measures and thus both approaches could be applied in this population.

Disclosure: Do you have a significant financial interest, consultancy or other relationship with products, manufacturer(s) of products or providers of services related to this abstract? (If not, please enter "No" in the text field.)

NO