Introduction: The SOFA score is an extensively validated tool for gauging disease severity with myriad publications in diverse fields including respiratory critical care. It is utilized in clinical trials and research throughout the world to standardize the assessment of disease severity. Healthcare has changed significantly since the initial 1996 publication validating SOFA with a proliferation of automated charting and electronically assessed vital signs.
Methods: We investigated how a simplified method of reporting the SOFA score compared to a traditional methodology. In the simplified method, the SOFA score was scored using the worst of the S:F or P:F ratios closest to noon and midnight. In the traditional method, the SOFA score used the worst of all available values. We collected data on 100,628 adult patients consecutively admitted to ICUs in the Intermountain Health Canyons region (14 hospitals) between 2018-2022. Our primary outcome was the time it took for an experienced data abstractor to generate a SOFA value.
Results: Our proposed standardized method took 37 seconds vs 65 seconds per patient day for the comprehensive method (p<0.001). The prognostic value of the tabulated scores was similar between methods (table below).
Conclusion: A simplified, standardized way of tabulating SOFA respiratory score can used to save time and accurately represent disease severity when abstracting data.
AUC 30-Day mortality | AUC Vent free days@28 | AUC ICU free days@28 | |
Traditional worst S/F or P/F of the day | 0.687 (0.682-0.692) | 0.529 (0.5031-0.553) | 0.544 (0.523-0.564) |
Worst of S/F or P/F closest to noon/midnight | 0.668 (0.662-0.673) | 0.551 (0.525-0.576) | 0.543 (0.522-0.564) |
p-value (Delong) | <0.001 | 0.010 | 0.956 |