The authors investigated the respiratory effects of diazepam in 24 healthy volunteers using a modified Read rebreathing circuit. Resting end-tidal CO2 (PETCO2) and the slopes of the ventilatory (VE/PETCO2) and occlusion pressure (P0.1/PETCO2) response to CO2 were measured just prior to and 5, 20, 40, and 60 min after diazepam, 0.1 mg/kg iv. The slope of VE/PETCO2 for all 24 subjects analyzed as a single group was never significantly depressed. The slope of P0.1/PETCO2 for all 24 subjects analyzed as a single group was significantly depressed only at 5 min after diazepam. The resting PETCO2, however, had small but statistically significant increases throughout the 1 h of study. Group or cluster analysis of the slope of P0.1/PETCO2 clearly divided subjects into one group of five subjects, whose P0.1/PETCO2 slope was significantly and consistently augmented for 1 h and a second group of 19 subjects whose P0.1/PETCO2 slope was always less than control for the entire hour. Diazepam may, through effects on pulmonary mechanics and/or the central nervous system, sometimes enhance respiratory responses to CO2 rebreathing. Failure to select for such group effects when studying drug effects by CO2 rebreathing may obscure the severity and duration of respiratory depression that occurs in the majority of individuals. Resting PETCO2 indicated consistent depression of resting minute ventilation by diazepam and may be a more appropriate or sensitive measure of mild or subtle drug-induced respiratory effects.