Patients hospitalized with severe infections and hypothermia, a cohort study of mortality and prognostic factors

Methods Prospective follow-up of all patients admitted to a medical emergency department (ED) from 1 August 2009-31 August 2011. Patients were included the first time they presented with severe infection in the study period. Hypothermia was defined as a rectal temperature below 36.0°C. Patients without a measured rectal temperature at arrival were excluded. Severe infections were defined as a discharge diagnosis indicating infection as well as presence of organ failure. To assess whether hypothermia was an independent prognostic factor, we computed multivariable logistic regression analysis, adjusted for different potential confounders.