Related Experiment Video
Updated: Jul 28, 2026
06:55
Inverse Probability of Treatment Weighting (Propensity Score) using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
Improving the predictive ability of the ICD-based Injury Severity Score
1Department Preventive and Social Medicine, Injury Prevention Research Unit, University of Otago, Dunedin, New Zealand. gabrielle.davie@ipru.otago.ac
Summary
Integrating mortality data and comorbidity information significantly enhances the predictive accuracy of the International Classification of Disease (ICD)-based Injury Severity Score (ICISS). This improved ICISS is crucial for accurate injury research and public health surveillance.
Area of Science:
- Public Health
- Trauma Research
- Health Informatics
Background:
- Accurate measurement of injury severity is vital for effective research, surveillance, and policy development in injury prevention and control.
- The International Classification of Disease (ICD)-based Injury Severity Score (ICISS) is a tool used to assess injury severity, but its predictive ability may be limited by the data sources used.
Purpose of the Study:
- To evaluate whether incorporating hospitalization and mortality data, along with comorbidity information, improves the predictive performance of the ICD-based ICISS.
- To compare the predictive accuracy of the standard ICISS with modified versions that include additional data sources.
Main Methods:
- Logistic regression modeling was employed to assess the predictive ability of the ICISS using hospital discharge data alone and nine modified ICISS versions.
- The study analyzed inpatient and fatality data from New Zealand between 2000 and 2003, focusing on S00-T89 ICD-10-AM diagnoses.
- Model performance was evaluated based on discrimination (concordance), calibration, and goodness-of-fit.
Main Results:
- A total of 186,835 cases, including 9,968 deaths, were analyzed.
- The modified ICISS incorporating both mortality data and Charlson comorbidity conditions demonstrated the highest concordance and calibration.
- Models using only hospital discharge data underestimated mortality risk, while those using both hospital and mortality data overestimated it.
Conclusions:
- The predictive capability of the ICISS can be substantially improved by integrating both mortality data and comorbidity information into its calculation.
- Enhanced ICISS provides more accurate injury severity measurement, benefiting research, surveillance, and injury policy implementation.