Background: Untreated disorders of the adrenocortical system, such as Cushing's or Addison's disease, can be fatal, and accurate quantification of a patient's cortisol levels is vital for diagnosis. The objective of this study was to assess the analytical performance of a new fully-automated Elecsys (R) Cortisol II assay (second generation) to measure cortisol levels in serum and saliva. Methods: Four European investigational sites assessed the intermediate precision and reproducibility of the Cortisol II assay (Roche Diagnostics) under routine conditions. Method comparisons of the Cortisol II assay vs. liquid chromatography-tandem mass spectrometry (LC-MS/MS), the gold standard for cortisol measurement, were performed. Cortisol reference ranges from three US sites were determined using samples from self-reported healthy individuals. Results: The coefficients of variation (CVs) for repeatability, intermediate precision, and reproducibility for serum samples were <= 2.6%, <= 5.8%, and <= 9.5%, respectively, and for saliva were <= 4.4% and <= 10.9%, and <= 11.4%, respectively. Agreement between the Cortisol II assay and LC-MS/MS in serum samples was close, with a slope of 1.02 and an intercept of 4.473 nmol/L. Reference range samples were collected from healthy individuals (n = 300) and serum morning cortisol concentrations (5-95th percentile) were 166.1-507 nmol/L and afternoon concentrations were 73.8-291 nmol/L. Morning, afternoon, and midnight saliva concentrations (95th percentile) were 20.3, 6.94, and 7.56 nmol/L, respectively. Conclusions: The Cortisol II assay had good precision over the entire measuring range and had excellent agreement with LC-MS/MS. This test was found suitable for routine diagnostic application and will be valuable for the diagnosis of adrenocortical diseases.