Studies assessing the effect of glucocorticoids on beta-endorphin regulation in man have yielded inconsistent results. As measured by a highly specific beta-endorphin assay procedure, plasma immunoreactive beta-endorphin (irB-EP) in six healthy subjects was significantly suppressed, along with cortisol, after a 2 mg oral dose of dexamethasone (DEX) (pre-DEX mean +/- S.E.M. in pg/ml = 15.3 +/- 2.0, post-DEX = 9.1 +/- 0,5 t = 3.46, p less than 0.01) but not after placebo (pre-placebo = 17.8 +/- 2.9, post-placebo = 17.2 +/- 1.5, t = 0.27). A less specific B-EP assay system, which yielded spurious irB-EP values (102 +/- 2.3 pg/ml) in plasma stripped of B-EP silicic acid, did not detect this DEX-induced change in plasma irB-EP levels (pre-DEX = 119 +/- 9.3, post-DEX = 112 +/- 5.0, t = 0.64; pre-placebo = 118 +/- 6.4, post-placebo = 119 +/- 4.3, t = 0.32). The methodological artifact encountered in this relatively non-specific assay system appears to account for the failure of an earlier study to demonstrate DEX suppression of irB-EP in man.