Introduction: Presentations related to illicit drugs are a feature of emergency department practice. Clinicians may form a belief that a patient is intoxicated with illicit drugs based on patient self-report, clinical features on presentation and the local prevalence of illicit drug use. But evidence of the accuracy of this assessment is lacking. The Western Australian Illicit Substance Evaluation (WISE) study enrolled patients believed by their treating clinician to be intoxicated with illicit drugs, and this analysis aims to evaluate the validity of this belief. Methods: A blood sample was taken on patient arrival and details of patient history, examination and interventions were collected by clinical and research staff. Toxicological examination of biological samples used liquid chromatography-mass spectrometry techniques including Quadrupole Time of Flight screening and Triple Quadrupole targeted analyses. Results: Of 632 study presentations, 518 had illicit drugs detected representing a positive predictive value of 0.82 (95% confidence interval 78.7, 84.9). Those with illicit drugs detected were significantly less likely to arrive by police transport (p = 0.010) or to have used alcohol (p < 0.001). They were significantly more likely to report illicit drug use (p < 0.001) and a much smaller proportion were admitted to a psychiatric ward (3.5% vs. 19.3%, p < 0.0001). Heart rate and systolic blood pressure were significantly higher in the illicit drug group (p = 0.004 and p = 0.003). Discussion and Conclusions: In this study, the positive predictive value of clinicians determining if their patient had taken illicit drugs was 0.82. Contemporaneous biochemical analysis in the clinical setting would increase this accuracy and inform patient care.