Screening tests have been used to support a biochemical diagnosis of Cushing's syndrome (CS). Measurements of salivary cortisol offer facilities for studying out-patients. This study assessed salivary cortisol in screening for CS by evaluating hypercortisolism based on circadian rhythm and the overnight 1-mg dexamethasone (DEX) suppression test for out-patients. We evaluated 33 patients with CS. Thirty normal volunteers and 18 obese patients were used as controls. Salivary cortisol (nanograms per dL) levels (mean +/- SEM) were 596 +/- 44, 528 +/- 104, and 1205 +/- 118 (0900 h); 213 +/- 27, 325 +/- 76, and 778 +/- 74 (1700 h); and 95 +/- 8, 133 +/- 26, and 914 +/- 94 (2300 h) in normal controls, obese subjects, and CS patients, respectively. After the overnight 1-mg DEX test, they were 64 +/- 1.1, 107 +/- 25, and 1048 +/- 129, respectively. In the present series, a single out-patient 0900, 1700, and 2300 h measurement and an overnight 1-mg DEX salivary cortisol level above the 90th percentile of the control or obese group values had sensitivities of 65.6%, 81.8%, 100%, and 100% or 78.1%, 57.6%, 93.3%, and 91.4%, respectively. The sensitivity improved (100%) in response to the combination of 2300 h and overnight 1-mg DEX salivary cortisol suppression tests to differentiate between obese and CS subjects. Our data indicate that nighttime sample and overnight 1-mg DEX suppression salivary cortisol tests are sensitive markers for the diagnosis of CS. In addition, the combination of the two tests improves the ability to differentiate between obese and CS patients and may be useful for out-patient screening.