Variation in the care of septic shock: The impact of patient and hospital characteristics

Purpose: The aim of this study was to examine treatments of septic shock in a sample of US hospitals and to assess whether patient and hospital characteristics are associated with use of sepsis therapies. Materials and Methods: We studied 192 hospitals that treated 50 or more adults with septic shock between 2004 and 2006. We examined hospital-level variation in commonly used therapies including mechanical ventilation, activated protein C (APC), hydrocortisone, central venous pressure (CVP) monitoring, albumin/colloid, and pulmonary artery catheters. We calculated interquartile range to assess the hospital-level variation in treatment. We developed hierarchical mixed-effects logistic regression models to examine the association between patient and hospital characteristics and selected treatments. Results: A total of 22 702 patients met the inclusion criteria. When compared with patients younger than 45 years, patients 75 years or older were as likely to receive mechanical ventilation but less likely to receive APC (odds ratio [OR], 0.35 [95% confidence interval, 0.27-0.45]), hydrocortisone (OR, 0.65 [0.56-0.75]), or CVP monitoring (OR, 0.73 [0.63-0.84]). Compared with whites, black patients were more likely to be mechanically ventilated (OR, 1.15 [1.05-1.25]) but less likely to receive hydrocortisone (OR, 0.86 [0.78-0.95]) or APC (0.70 [0.58-0.86]). Conclusion: Treatment of septic shock varies across hospitals. In contrast to mechanical ventilation, treatments with weaker supporting evidence showed greater variation, especially among black and older patients.