Objectives: Opioid overdose deaths in Massachusetts linked to illicitly-manufactured fentanyl have increased dramatically. In response, an urban safety-net hospital added urine fentanyl testing with reflex confirmation testing to its standard urine toxicology panel. The goals of this study were to describe fentanyl toxicology test results, identify the positive predictive value of presumptive fentanyl immunoassay, and describe co-substance use among those with unexpected fentanyl positive results.
Methods: We included urine toxicology tests from January through June 2016 analyzed at an urban safety-net hospital. We excluded tests from individuals prescribed or administered fentanyl within the preceding 72 hours. Positive fentanyl immunoassay tests underwent reflex chromatography confirmation testing. Samples that confirmed positive for acetyl fentanyl and/or fentanyl and/or norfentanyl were considered true positives.
Results: Of 11,873 urine samples, 10.4% of samples screened fentanyl positive and 8.8% were confirmed fentanyl positive. The positive predictive value of a positive urine fentanyl screen was 85.7%. Of 4398 unique patients, 13.2% had at least 1 test confirmed positive for nonprescription fentanyl. Patients with a confirmed fentanyl positive drug test were more likely to have positive urine drug test for barbiturates, benzodiazepines, cocaine, methadone, and opiates, and less likely to have oxycodone or buprenorphine.
Conclusions: At an urban safety-net hospital, nonprescription fentanyl use was common and was associated with greater use of other substances favoring routine fentanyl testing. Although the positive predictive value of the screening test was high, confirmation testing detected substantial numbers of false positives, especially in older patients. Therefore, fentanyl confirmation testing should be used when results will change treatment approach and patient education.
Copyright © 2020 American Society of Addiction Medicine.