Estimating equivalent cutoff thresholds for drugs in blood and oral fluid using prevalence regression: A study of tetrahydrocannabinol and amphetamine

AimTo validate a method for determining equivalent drug cutoff concentrations for tetrahydrocannabinol and amphetamine in blood and oral fluid, which ensures that the drug prevalence in samples of blood and oral fluid taken simultaneously is equal.MethodsA method using regression analysis of drug concentrations for defined percentiles in blood and oral fluid was developed. The accuracy and precision of this technique was investigated. As study populations, 311 cannabis users and 197 amphetamine users from the Rosita-2 Project were used.ResultsA total of 80 paired oral fluid and blood concentrations were needed to determine accurate regression formulae. When using the formulae to calculate drug cutoff concentrations in oral fluid corresponding to 2.0, 4.0, 6.0, 8.0 and 10.0 ng/ml tetrahydrocannabinol in blood and 200, 400, 600, 800 and 1000 ng/ml amphetamine in blood, the accuracy was better than 100 ± 20% compared to actual prevalence in blood with precision better than ±20%.ConclusionPrevalence regression may be a useful tool in estimating equivalent cutoff concentrations in blood and oral fluid.