Skip to main content

Table 3 Analysis of Reclassification Events

From: Physician agreement on the diagnosis of sepsis in the intensive care unit: estimation of concordance and analysis of underlying factors in a multicenter cohort

Reclassification

Number (%) reclassified: attending physician ➔ discharge evaluation by site investigators (Additional file 3: Figure S3–1)

Number (%) reclassified: attending physician ➔ RPD (Additional file 3: Figure S3–2)

Change in apparent in risk profile

Potential consequence of erroneous initial classification

SIRS to sepsis

4 (1.6%)

6 (2.4%)

Low to high

Delayed antibiotic treatment, prolonged hospital stay, and increased morbidity and mortality

Indeterminate to sepsis

12 (4.8%)

22 (8.8%)

Medium to high

Possible delayed antibiotic treatment

SIRS to indeterminate

5 (2.0%)

5 (2.0%)

Low to medium

Possible delayed antibiotic treatment

Sepsis to SIRS

9 (3.6%)

9 (3.6%)

High to low

Excess antibiotic use

Indeterminate to SIRS

18 (7.2%)

16 (6.4%)

Medium to low

Possible excess antibiotic use

Sepsis to indeterminate

15 (6.0%)

12 (4.8%)

High to medium

Possible excess antibiotic use

Total

63 (25.3%)

70 (28.1%)