Classification of comorbidity in trauma: agreement and reliability of the pre-injury ASA-PS Scale
© Ringdal et al; licensee BioMed Central Ltd. 2012
Published: 22 March 2012
Pre-injury comorbidity influence outcome in severely injured patients. Pre-injury comorbidity graded according to the American Society of Anesthesiologists Physical Status (ASA-PS) classification system (Table 1) is an independent predictor of survival in trauma patients, and is suggested as a comorbidity score in the Utstein Trauma Template. Little is known about the levels of agreement and reliability of pre-injury ASA-PS scores. The objective of the study was to examine if pre-injury ASA-PS was a reliable scale for grading comorbidity of trauma patients among a representative group of trauma registry coders.
18 Norwegian trauma registry coders were invited to participate in an agreement and reliability study, of which 50 real but anonymised patient medical records were distributed. Agreement was analysed using proportions, and reliability was analysed using quadratic weighted kappa (κw) with 95% confidence intervals (CI) as the primary outcome measure, and unweighted kappa (κ) analysis that included ‘ Unknown’ values as secondary outcome measure.
15 of the invitees responded to the invitation, and ten participated. Agreement between the participants and the gold standard was 73.6%. After recoding the ‘ Unknown’ values into pre-injury ASA-PS 1, agreement was 80.4%. We found moderate (κw=0.76, 95%CI 0.64-0.87) to substantial (κw=0.95, 95%CI 0.89-0.99) rater-to-gold standard reliability using κw (Table 2 and Figure 1), and fair (κ=0.46, 95%CI 0.29-0.64) to substantial (κ=0.83, 95%CI 0.68-0.94) reliability using κ (Table 3 and Figure 2). Inter-rater reliability showed a mean κw=0.82 (95%CI: 0.80-0.84), and a mean κ=0.57 (95%CI: 0.54-0.60).
Rater-to-gold standard agreement and reliability varied from moderate to substantial for the primary outcome measure, and from fair to substantial for the secondary outcome measure. The findings of the study indicate that the pre-injury ASA-PS scale may be a reliable score for classifying comorbidity in trauma patients. The Utstein Template should specify how to handle unknown values.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.