Reference intervals: A novel approach to detect drug abuse in a pain patient population

Background: Pain physicians have few objective ways of determining which of their patients are drug abusers. Traditionally, these include psychological tests, physical examination, patient history, and urine drug testing. The traditional urine drug testing information provided to pain physicians mainly identifies patient compliance or drug diversion with qualitative information, that is, the patient is positive or negative for the presence of the drug in excreted urine. Although this information is useful for establishing compliance and identifying diversion, it is incomplete because it does not identify drug abuse. Objective: The authors endeavored to determine whether quantitative urine drug testing and mathematical estimators of the upper limits of excretion could be used to identify possible drug abusers. Study Design: Analysis of quantitative urine drug tests and application of mathematical models for reference interval estimation of common analytes to determine whether they could be used to define upper 97.5 percentile limits of excretion in the pain patient population. Methods: The authors analyzed 8,971 consecutive urines from patients on chronic opioid therapy using nonparametric, parametric, robust, and transformed estimators to derive the upper 97.5 percentile concentration values of 31 drugs and their metabolites. Results: The authors showed that the mathematical models used to define reference intervals could be applied to urinary drug excretion. As an example, an upper limit of excretion of approximately 100,000 ng/mL was established for morphine. Limitations of the study included lack of information on medication history, time of last dose before urine collection, age, sex, and complete medical history. Better estimates of the upper limits of excretion can be obtained by physicians applying their knowledge of dosage and collection times. Conclusions: Application of a reference interval model allows identification of a patient population that excretes extremely high amounts of drug or its metabolite when compared with the rest of the population. Explanations for this high excretion include high dosage medication by prescription and drug abuse, determination of which can be done by the treating physician. The authors suggest that this patent-applied-for analytical model can become a potential tool to alert physicians to patients who may be abusing drugs.