To clarify the incremental prognostic value of SOFA-2 compared to the original SOFA score, focusing on data accessibility and analysis methods.
Approach:
Data Availability Comparison: Examined the differences in component data availability between SOFA and SOFA-2, highlighting significant disparities in respiratory and neurologic components.
Inverse Probability Weighting Analysis: Critiqued the covariates used for inverse probability weighting and the clarity of the estimand intended by the weighted comparison.
Exact-Matching Analyses: Discussed the limitations of reciprocal exact-matching analyses in demonstrating incremental prognostic value from using both scores.
Key Findings:
SOFA-2 shows improved data availability for calculating scores compared to the original SOFA.
The unequal availability of data may influence the observed reclassification between the two scoring systems.
The rationale and methodology for inverse probability weighting require clearer specification.
Exact-matching analyses indicate associations but do not confirm incremental prognostic value.
Limitations:
Differential data availability may skew results.
Incomplete description of covariates used in inverse probability weighting.
Exact-matching analyses do not directly test for incremental prognostic value.
