To the Editor: Many studies have demonstrated that passive exposure to marijuana smoke can lead to detectable concentrations of tetrahydrocannabinol (THC) in oral fluid (OF) (1 –3). It is reasonable to assume that subjects breathing air heavily laden with THC will absorb some of the drug. An important related issue is how long concentrations will remain detectable. The magnitude of the exposure depends greatly on many variables, including the duration of exposure, the potency of the marijuana being smoked, and the concentration of smoke in the enclosed space (i.e., room size, number of smokers and number of passive inhalers). Peak concentrations in OF have generally been observed at the end of exposure followed by a rapid decline typically within one hour (1, 2). Recently, Moore et al. (3) found that although THC was present in the OF sampled from the passively exposed subjects, no THC carboxy acid metabolite (THC-COOH) was detected at a cutoff of 2 pg/mL. The authors concluded that THC-COOH is likely to be a valid marker for active marijuana use, and recommended that “in order to avoid false positive oral fluid results assigned to marijuana use, by analyzing for only THC, the metabolite THC-COOH should also be monitored.” In our view, two key issues for workplace drug testing programs are raised by this research. First is the question of the validity of THC-COOH as a unique marker of cannabis use. We believe that for a marker to be useable, it must be scientifically validated that the marker occurs only in the “true” condition (characterized by such variables as dose response, time course of appearance and disappearance, and potential interferences (false positives) and does not occur in the “false” condition (no active use). It is well known that THC-COOH reaches peak concentrations in plasma at 0.5 to 4 hours and in urine at 8 to 14 hours, thus indicating a delay or lag time necessary for THC to be metabolized to THC-COOH in the liver, released into the blood and excreted in body fluids. Therefore, THC-COOH may have been present in the passively exposed subjects’ OF at times not sampled in previous studies. Second, for drug testing program policy makers, a major question is how likely it is that an individual would get sufficient passive exposure to test positive at some later time. At present, it remains unclear how long detectable levels of THC persist after the exposure ends. The answer to this question needs to be explored scientifically by evaluating dose-response,