OBJECTIVE: To compare serum total, serum free and salivary cortisol in critically ill children. DESIGN: Prospective observational cohort study. SETTING: Tertiary pediatric critical care unit at Ronald McDonald Children's Hospital at Loyola University Medical Center. PATIENTS: We enrolled 59 patients (4 weeks to 18 years of age) between January 2012 and May 2013. Thirty-four patients were included in the salivary to serum free cortisol correlational analysis. INTERVENTIONS: Blood and saliva samples were obtained simultaneously within 24 hours of admission between the hours of 6 AM and 12 PM. Salivary cortisol was tested by liquid chromatography/tandem mass spectrometry, serum free cortisol by liquid chromatography/tandem mass spectrometry followed by equilibrium dialysis, and serum total cortisol by liquid chromatography/tandem mass spectrometry. MEASUREMENTS AND MAIN RESULTS: Salivary and serum free cortisol values from 34 patients had a correlation coefficient (r) of 0.87 (95% CI, 0.75-0.93; p < 0.0001). The total serum and salivary cortisol values had a correlation coefficient (r) of 0.67 (95% CI, 0.42-0.81; p < 0.0001). The total serum and serum free cortisol values had a correlation coefficient (r) of 0.83 (95% CI, 0.69-0.91; p < 0.0001). CONCLUSIONS: Serum free and salivary cortisol values correlate in critically ill children. Salivary cortisol can be used as a surrogate for serum free cortisol in critically ill pediatric patients. Salivary cortisol is a cost-effective and less invasive measure of bioavailable cortisol and offers an alternate and accurate method for assessing critical illness-related corticosteroid insufficiency in children.