Table 3

Cox regression for analysing in-hospital mortality by five patient groups according to the different start of sepsis treatment protocol in both hospitals

Total patient cohort
(n=750)
n (%)
In-hospital mortality,
n (% per group)
Intravenous antibiotic treatment,
n (% per group)
HR (95% CI)P value
1. SIRS+, qSOFA−, not treated according to protocol319 (42.5)18 (5.6)179 (56.1)1Reference
2. SIRS+, qSOFA−, treated according to protocol338 (45.1)16 (4.7)187 (55.3)0.8 (0.4 to 1.6)0.502
3. SIRS+, qSOFA+, treated according to protocol80 (10.7)21 (26.3)67 (83.8)3.3 (1.7 to 6.2)<0.001
4. SIRS−, qSOFA+, not treated according to protocol2 (0.3)1 (50)2 (100)18.8 (2.5 to 143.8)0.005
5. SIRS−, qSOFA+, treated according to protocol6 (0.7)3 (50)6 (100)5.0 (1.5 to 17.4)0.011
Missing, n (%)5 (0.7)
  • Values are expressed in n (number) and percentages, which denote the proportion within a group of patients. Because of missing vital signs, qSOFA could not be calculated for five patients (figure 2). These patients were categorised as ‘missing’ and excluded from this analysis.

  • qSOFA, quick Sequential Organ Failure Assessment; SIRS, systemic inflammatory response syndrome.