This study involved the evaluation of the Abbott TDx serum benzodiazepine assay, a fluorescence polarization immunoassay (FPIA), for the detection of lorazepam, adinazolam, and N-desmethyladinazolam in serum. Precision of the assay was determined by using three control serums containing 75, 300, and 700 ng/mL nordiazepam. Between-run precision studies (N = 22) gave mean values of 76, 306, and 690 ng/mL with coefficients of variation of 6.5, 3.3, and 5.7%, respectively. Percent cross-reactivity of serum lorazepam standards (35–500 ng/mL) ranged from 29 to 69%. The cross-reactivity of serum adinazolam ranged from 40 to 47% between 50 and 150 ng/mL and from 38 to 55% for N-desmethyladinazolam between 50 and 250 ng/mL. Serum specimens (48) collected from individuals known to be receiving lorazepam were analyzed. Twenty-two specimens were positive for benzodiazepines. Serum specimens were collected from 0.25 to 24 h after administering a 15-mg oral dose of adinazolam to six volunteers. The FPIA results were compared with combined high-performance liquid chromatographic (HPLC) results for adinazolam and N-desmethyladinazolam. The FPIA method did not detect benzodiazepines at 0.25 h after administration of adinazolam but did detect benzodiazepines from 0.5 to 24 h after administration. The correlation between HPLC (N-desmethyladinazolam) and FPIA results by regression analysis gave the following: y = 0.937x + 4.449, r = 0.98, n = 15. It was concluded that the Abbott FPIA assay for benzodiazepines can detect lorazepam when prescribed in therapeutic doses and when present at greater than 25 ng/mL and can semiquantitatively detect adinazolam or N-desmethyladinazolam or both when present at concentrations greater than 50 ng/mL.