For identifying positive cases, the cutoffs proposed by the SoHT were used. 14 Table 1 reports the class of drugs of the IAs and the corresponding confirmation parameters. All 150 samples were analyzed using IAs as the screening test, and with UPLC-MS/MS as the confirmatory test.
To optimize the screening cutoffs, the result of the IA test was semi-quantitatively expressed in terms of "ng/mg IA units"; the results of the confirmatory analysis were considered positive when the SoHT criteria were respected, considering the main analyte and, when necessary, metabolites. As an example, for a positive cocaine result, the SoHT document assumed that "The presence of benzoylecgonine, norcocaine, cocaethylene, hydroxylcocaines, or hydroxy-benzoylecgonine must be considered 1 (link) with the presence of metabolites (benzoylecgonine, norcocaine, cocaethylene, hydroxyl-cocaines or hydroxy-benzoylecgonine); IA -immunoassay; LC-MS/MS -ultra-performance liquid chromatography coupled to tandem mass spectrometry; MDMA -3,4-methyl-enedioxymethamphetamine; THC -tetrahydrocannabinol.
to confirm use". Then, the result is considered "positive" at confirmation analysis only when cocaine was found above the cutoff, with the presence of a metabolite.
Receiver operating characteristic (ROC) curves were built using GraphPad Prism 9.5.1 software (GraphPad Software, San Diego, USA), which computes them from raw data. To this scope, screening results testing negative by UPLC/MSMS were inserted as "negative", while screening results corresponding to a positive result by UPLC/ MSMS were inserted as "positive". Due to the cross-reactivity of the IA amphetamines test to methamphetamine, for the optimization of the amphetamines IA test, the sample was considered positive on confirmation analysis when the presence of amphetamine or methamphetamine above the confirmation cutoff was assessed. A sample was considered positive for cocaine when the presence of cocaine above the confirmation cutoff and metabolites were assessed, as requested by SoHT guidelines. A sample was considered positive for opiates when a result was positive for morphine, codeine, dihydrocodeine, 6-monoacetylmorphine, heroin, or tramadol above the confirmation cutoffs. A sample was considered positive for MDMA, cannabinoids and methadone when the presence of MDMA, THC and methadone, respectively, were encountered.
The ability of a test to discriminate between positive and negative samples is provided as the area under the ROC curve (area under the curve (AUC)) calculated with a standard error (SE) and 95% confidence interval (95% CI), as well as a p-value. The software automatically tabulates and plots the sensitivity and specificity of the test using each value in the data table as a possible cutoff value. A likelihood ratio is additionally calculated. Screening cutoff values were retrospectively optimized using contingency tables and assessed through ROC analysis. The SE of the area is calculated using the equation from Hanley and McNeil. 18 (link) We determined the optimal cutoff by summing the sensitivity and specificity. Sensitivity, calculated as TP/(TP+FN), and specificity, calculated as TN/(TN+FP), were determined using the numbers of true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN). In cases where the sensitivity at the optimized cutoffs was less than 1, we retrospectively calculated the cutoff with a sensitivity equal to 1, referred to as the "highest sensitivity cutoff" (HS cutoff). Since the samples were processed anonymously for the purposes of this study, which did not involve the collection of any personal data, obtaining ethical approval was deemed unnecessary.
To optimize the screening cutoffs, the result of the IA test was semi-quantitatively expressed in terms of "ng/mg IA units"; the results of the confirmatory analysis were considered positive when the SoHT criteria were respected, considering the main analyte and, when necessary, metabolites. As an example, for a positive cocaine result, the SoHT document assumed that "The presence of benzoylecgonine, norcocaine, cocaethylene, hydroxylcocaines, or hydroxy-benzoylecgonine must be considered 1 (link) with the presence of metabolites (benzoylecgonine, norcocaine, cocaethylene, hydroxyl-cocaines or hydroxy-benzoylecgonine); IA -immunoassay; LC-MS/MS -ultra-performance liquid chromatography coupled to tandem mass spectrometry; MDMA -3,4-methyl-enedioxymethamphetamine; THC -tetrahydrocannabinol.
to confirm use". Then, the result is considered "positive" at confirmation analysis only when cocaine was found above the cutoff, with the presence of a metabolite.
Receiver operating characteristic (ROC) curves were built using GraphPad Prism 9.5.1 software (GraphPad Software, San Diego, USA), which computes them from raw data. To this scope, screening results testing negative by UPLC/MSMS were inserted as "negative", while screening results corresponding to a positive result by UPLC/ MSMS were inserted as "positive". Due to the cross-reactivity of the IA amphetamines test to methamphetamine, for the optimization of the amphetamines IA test, the sample was considered positive on confirmation analysis when the presence of amphetamine or methamphetamine above the confirmation cutoff was assessed. A sample was considered positive for cocaine when the presence of cocaine above the confirmation cutoff and metabolites were assessed, as requested by SoHT guidelines. A sample was considered positive for opiates when a result was positive for morphine, codeine, dihydrocodeine, 6-monoacetylmorphine, heroin, or tramadol above the confirmation cutoffs. A sample was considered positive for MDMA, cannabinoids and methadone when the presence of MDMA, THC and methadone, respectively, were encountered.
The ability of a test to discriminate between positive and negative samples is provided as the area under the ROC curve (area under the curve (AUC)) calculated with a standard error (SE) and 95% confidence interval (95% CI), as well as a p-value. The software automatically tabulates and plots the sensitivity and specificity of the test using each value in the data table as a possible cutoff value. A likelihood ratio is additionally calculated. Screening cutoff values were retrospectively optimized using contingency tables and assessed through ROC analysis. The SE of the area is calculated using the equation from Hanley and McNeil. 18 (link) We determined the optimal cutoff by summing the sensitivity and specificity. Sensitivity, calculated as TP/(TP+FN), and specificity, calculated as TN/(TN+FP), were determined using the numbers of true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN). In cases where the sensitivity at the optimized cutoffs was less than 1, we retrospectively calculated the cutoff with a sensitivity equal to 1, referred to as the "highest sensitivity cutoff" (HS cutoff). Since the samples were processed anonymously for the purposes of this study, which did not involve the collection of any personal data, obtaining ethical approval was deemed unnecessary.
