Background: Serum blood toxicology screens are believed to be important to monitor compliance and to identify levels of illicit substances in patients taking opioids for their chronic pain.Methods: In this study, the authors examine the incidence of tetrahydrocannabinol (THC) in consecutive blood samples of patients given dronabinol. We assessed the incidence of THC in 27 patients who participated in a single-dose, double-blind crossover trial of dronabinol (Marinol® capsules), a synthetic Δ9-THC, as part of a larger study, to determine the reliability of the toxicology screening. Subjects were randomly administered 10 mg or 20 mg of dronabinol or placebo over the course of three 8-hour visits for a combined 228 serum blood samples. Levels of THC were quantified using radioimmunoassay.Results: The majority of the samples (57.4 percent) showed presence of study drug as expected. However, 43 samples (42.6 percent) showed no detectable evidence of THC 4 and 8 hours after administration of dronabinol. Five subjects showed lower serum levels on the higher dose (20 mg) than on the lower dose (10 mg) after 4 hours, and two subjects showed lower levels with the higher dose after 8 hours. One subject had no detectable THC on any dose of dronabinol.Conclusions: These toxicology reports point to higher than anticipated false-negative results with radioimmunoassay blood serum screening. Results could be explained by the lower sensitivity of this screening technique and also in how oral cannabinoids are metabolized. Further investigations are needed on the accuracy of the detection of THC among patients known to have used dronabinol.