Serum osteocalcin is a sensitive marker of suppressive effects of exogenous glucocorticoids on bone turnover. It has been suggested, however, that the degree of suppression detected by different assays may vary. Whether discrepancies between various assays influence conclusions from group studies of exogenous glucocorticoids has not been evaluated. The aim of the present study was to compare the CAP fluoroimmunoassay (FEIA), OSTK-PR and ELSA-OSTEO assays for assessment of prednisolone-induced effects on serum osteocalcin. Twelve men and eight premenopausal women aged 19-45 (mean 31) years were studied. All subjects were healthy. The design was a randomised double-blind, placebo-controlled parallel- group study with 2 days run-in, 3 days treatment and 4 days run-out. During run-in and run-out no medication was given. During the treatment period the subjects took either 20 mg prednisolone twice daily or placebo. Blood was collected on the last day of each period. Intra- and intergroup comparisons showed prednisolone treatment to be associated with a statistically significant suppression of osteocalcin which was detected by all assays (ANOVA;P<0.0001). In the individual subjects the response to prednisolone was the same for each assay. The CAP FEIA, OSTK-PR and ELSA-OSTEO assays seem equally sensitive for evaluation of osteocalcin in group studies of oral glucocorticoids.