Abstract
In forensic toxicology, it has been debated if hair testing allows an estimation of the intensity of cocaine use—an assumption that may have risen because self‐reports in a forensic setting are of uncertain validity per se. We therefore investigated the relationship between self‐reported cocaine use and cocaine hair concentrations (including its main metabolites benzoylecgonine and norcocaine) in chronic cocaine users voluntary participating in psychiatric study settings. Additionally, we tested whether hair testing can distinguish between individuals with and without a diagnosis of cocaine dependency. Cocaine users (N = 195) from three independent experimental studies reported their average powder cocaine consumption in g/week over the last 3–4 months in an interview and provided a 3‐ to 4‐cm hair sample assayed with liquid chromatography tandem‐mass spectrometry. Moreover, study participants were assessed with the Structured Clinical Interview (SCID‐IV) for psychiatric diagnoses. Using linear regression models, we found a robust correlation between cocainetotal (sum of cocaine and metabolites) hair concentration and self‐reported cocaine use in g/week (rcocainetotal = 0.47, p < 0.001), indicating that 1000 pg/mg cocainetotal corresponded to a use of 0.80 g/week (confidence interval [95%]: 0.56–1.07 g/week). In logistic regression models, cocainetotal hair concentration predicted cocaine dependency with a sensitivity of 0.79 and a specificity of 0.65 (threshold 0.5), suggesting its acceptable capacity to distinguish dependent from non‐dependent cocaine users. The findings may have significant implications for forensic and clinical practices, encouraging the use of hair analysis as a potential tool for monitoring cocaine use and dependence.
Grison, Simon