Revision of the International Classification of Diseases to include standardized descriptions of multiple injuries and injury severity
Shinji Nakahara & Junichiro Yokota
Volume 89, Number 3, March 2011, 238-240
Table 1. Methods for scoring severity of injuriesa
| Severity scores | Definition | Characteristics | Required resources |
|---|---|---|---|
| Abbreviated injury scale (AIS) | An injury categorization with severity scores assigned to each injury category. Injuries are rated from 1 (minor) to 6 (fatal). | – Not designed for survival prediction. – Determined based on expert consensus. |
– Duplicate coding or computer software (ICDMAP) to obtain AIS severity scores from ICD codes. |
| Injury severity score (ISS) | Indicates overall severity for a patient with multiple injuries. ISS is a sum of the square of the highest AIS severity scores of the three most severely injured body regions (from a choice of six body regions). ISS = AIS12 + AIS22 +AIS32 |
– Does not consider physiological parameters. – Equal weighting given to each body region. – Does not account for multiple injuries in the same body region. |
– AIS severity score |
| Revised trauma score (RTS) | Consists of physiological parameters independent of anatomical injury scores. RTS = 0.9364 × GCS + 0.7326 × SBP + 0.2908 × RRb |
– Physiological parameters are time-sensitive. | – Patient data and statistical software to calculate country-specific coefficients. |
| Trauma and injury severity score (TRISS) | A combination of an anatomical measure (ISS), physiological measure (RTS) and patient ability to withstand injury severity (age) by type of injury (blunt/penetrating). Probability of survival (Ps) is determined using a logistic regression model. Logit (Ps) = β0 + β1 × RTS + β2 × ISS + β3 × ageb | – Widely used in outcome studies because of its good predictive ability. | – Availability of AIS severity score. – Patient data and statistical software to calculate country-specific coefficients. – Computer software to calculate the score because of its mathematical complexity. |
| ICD-based injury severity score (ICISS) | A multiplicative prediction model with an assumption that all injuries contribute to the overall severity. The SRR for each code is empirically derived from the patient data. To obtain ICISS, SRRs of all injuries are multiplied. ICISS = SRRinj1 × SRRinj2 × SRRinj3 × SRRinjn |
– Directly derived from ICD or ICD-CM codes. – Predictive ability is equal to, or better than, that of the TRISS. | – Large patient data set. – Computer software might be required to calculate each patient’s score due to large number of codes |
| Matrix-based method | In a body-region by injury-nature matrix (such as the Barell matrix), the proportions of survival and approximated AIS score are calculated based on data for each cell. These values are used in the same way as ICISS and AIS-based indices. | – Relatively easy to handle due to diminished number of categories compared with other methods. | – Patient data set (not necessarily a large one) and statistical software to calculate country-specific values. – AIS severity score if approximated severity scores are determined. |
GCS, Glasgow Coma Score; ICD, International Classification of Diseases; ICD-CM, International Classification of Diseases-Clinical Modification; RR, respiratory rate; SBP, systolic blood pressure; SRR, survival risk ratio.
a This is not a comprehensive list of injury scores, but rather shows typical and popular indices to indicate their relationships with the ICD codes and required resources.
b Coded values are used for Glasgow Coma Score, systolic blood pressure, respiratory rate and age.