Fig. 3

Test of effect moderation of timing of treatment in prespecified subgroups for time from onset of sepsis to start of antimicrobial therapy (panel a) and conduction of surgical source control (panel b) on 28-day mortality. OR: odds ratio per hour delay of treatment. Only cases with times between 0 and 48 h were considered. Effect moderation was tested by introducing individual interaction terms between respective effect moderator and the effect of timing of antimicrobial therapy or timing of surgical source control, respectively, in the logistic regression models. Regression models were calculated by logistic hierarchical linear models with a random intercept. No. of patients gives the number of cases with complete data compared to the total number of patients within the respective subgroup. Models adjusted for the following covariates: age and gender, origin of infection, location of the patient at the onset of sepsis, focus of infection, microbiological confirmation of infection, study phase (trial vs. surveillance phase), and group the hospital was randomized to (intervention vs. control). ICU: intensive care unit; IMC: intermediate care unit