Purpose: Young cannabis users are at increased risk for several mental health issues including dependence, depression and psychosis. THC-COOH (one of the two major metabolites from delta 9-tetrahydrocannabinol) is mostly used as a marker for both cannabis intoxication and abstinence when tested in urine. The present study assesses to which extent urine THC-COOH may help identify problematic cannabis users. Methods: Data are issued from an observational study where, for methodological purposes, specific groups were recruited based on their tobacco and cannabis profile in the past month. Among a total of 269 participants, 57 cannabis-only users and 72 cannabis and tobacco users completed an anonymous self-administered questionnaire describing their cannabis use in the past month, and most precisely in the past 5 days. The CAST (Cannabis Abuse Screening Test), a well validated questionnaire for both clinical and large epidemiological studies, was used to identify problematic cannabis use. Based on 6 items, participants scoring 2 and above were described as problematic cannabis users (PCU; N¼81, mean age 19 years, 64% male) in comparison to non-problematic cannabis users (NPCU; N¼48, mean age 19 years, 63% male). All participants also provided a urine sample that was blindly analyzed for THC-COOH among other substances, using gaschromatography coupled mass-spectrometry. PCU and NPCU were compared on their cannabis use in the past month (at least once a week/less than once a week), number of consumptions in the past 5 days (meanSE), and urine THC-COOH levels (meanSE). As a way to determine if urine THC-COOH is a proper marker to identify PCU, we used the receiver operating characteristic (ROC) curve to assess its sensitivity and specificity using various cut-off values. Results: Smoking cannabis at least once a week was reported more frequently among PCU than NPCU (95% vs. 63%; p<0.001). Using ANOVA, mean cannabis consumption in the past 5 days (11.31.0 vs. 4.30.7; p<0.001) and mean urine THC-COOH level (291.661.4 vs. 82.652.7; p¼0.021) were higher among PCU than for NPCU. With an optimum cut-off of 21 ng/ml, positive urine THC-COOH displayed a sensitivity of 79% and a specificity of 58% for problematic cannabis use. Using values above 120 ng/ml, its specificity increased to more than 95%. Conclusions: Even if urine THC-COOH levels were found to be much higher among problematic cannabis users, such method seems poorly performant in identifying them from other regular cannabis users. Nevertheless, individuals presenting with high value for THC-COOH on urine drug testing should definitely be evaluated regarding cannabis use disorder.