The British Columbia Coroner’s Office reports of overdose deaths in Vancouver and overdose calls to first responders (police and fire department) for Vancouver and for the neighborhood from which participants were recruited (A). Numbers of urine samples with fentanyl detected are shown for participants using nonprescribed opioids and for those taking prescribed opioid agonist therapy (B). The probability of urine samples being positive for fentanyl (C) or opiates (heroin, morphine, or codeine) (D) changed during a 5-month period and differed according to reported use of nonprescribed opioids in the week before urinalysis. Samples were tested for fentanyl/norfentanyl (sensitivity 10 ng/mL, with cross-reactivity for acetylfentanyl, butyrylfentanyl, carfentanil, fluorofentanyl, 4-fluoroisobutyryl fentanyl, furanylfentanyl, 3-methylfentanyl, sufentanil, and thiofentanyl) and for opiates (heroin, morphine, and codeine). Shading represents the 95% confidence interval.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Jones AA, Jang K, Panenka WJ, et al. Rapid Change in Fentanyl Prevalence in a Community-Based, High-Risk Sample. JAMA Psychiatry. 2018;75(3):298–300. doi:10.1001/jamapsychiatry.2017.4432
Planning for the implications of nonprescribed fentanyl use relies on multisource forensic1,2 or clinical samples.3,4 Complementing these descriptions, we report a prospective longitudinal study of change in urine fentanyl prevalence in a high-risk, community-based sample.5
Directly assessed participants were from a health outcomes study of people living in an impoverished neighborhood of Vancouver, Canada.5 For context, overall overdose deaths (Vancouver) and first responder calls (Vancouver and neighborhood-specific) were obtained from the British Columbia Coroner’s Office and Vancouver Police and Fire statistics for the study period (March 1 to July 31, 2017). Participants attended monthly visits and reported prescribed and nonprescribed drug use during the previous week, including fentanyl, buprenorphine, codeine, heroin, hydromorphone, methadone, morphine, and oxycodone. Participants (N = 237) contributed 595 urine samples that were tested for fentanyl/norfentanyl, opiates (morphine, heroin, and codeine), and methadone using detection strips (BTNX Inc). Agreement between reports and detection was assessed by κ statistic. Repeated measures logistic mixed-effects models with random intercept and slope were used to estimate associations between detection and reported opioid use in the prior week. Fixed effects of recent nonprescribed opioid use over time were estimated adjusting for age and sex. The study was approved by the institutional review boards at the University of British Columbia and Simon Fraser University, and participants provided written informed consent.
Between March and April 2017, an upsurge occurred in overdose deaths and first responder calls (Figure, A). Directly assessed participants had a mean (SD) age of 46.4 (12.2) years, were mostly men (184 of 237 [78%]), and were marginally housed or street homeless. Participants had a mean education of 10.2 years. Injection drug use in the past week was reported by 110 of 231 participants (48%). During the 5-month period, nonprescribed opioid use was reported by 91 of 237 individuals (38%), including 57 of 103 individuals (55%) who were prescribed opioid agonist therapy (hydromorphone, methadone, buprenorphine, morphine, or heroin). Fentanyl was detected in 229 of 590 urine samples (39%), including 116 of 222 samples (52%) from participants prescribed opioid agonist therapy (Figure, B). Overall, 83 of 91 participants (91%) reporting nonprescribed opioid use had at least 1 fentanyl-positive sample; 15 of these 83 (18%) reported taking fentanyl (11 of whom reported daily use). Opiates were detected in 196 of 581 urine samples (34%). Agreement between self-report and detection was low for fentanyl (κ = 0.12) and moderate or greater for other opioids (κ range, 0.54-0.84). The probability of fentanyl detection doubled each month (odds ratio, 2.28; P < .001) (Figure, C and Table). With self-reported nonprescribed opioid use, fentanyl detection probability was greater (odds ratio, 34.47; P < .001) and increased at a faster rate over time (odds ratio, 2.34; P = .03). In contrast, opiate detection decreased over time (odds ratio, 0.32; P = .003) (Figure, D and Table). By July 2017, all samples from participants reporting nonprescribed opioid use were fentanyl-positive.
Fentanyl-positive urine samples increased rapidly during a 5-month period while opiate-positive samples declined. In Vancouver, as elsewhere,6 the initial phase of the opioid epidemic was associated with diverted pharmaceuticals. This changed as nonpharmaceutical fentanyl entered the market as a heroin additive. The low concordance between reported fentanyl use and detection is consistent with unawareness of exposure. In the early months, as fentanyl-positive samples rapidly increased in the participants, an increase in overdose calls to first responders occurred in the neighborhood, and fatal overdoses increased city-wide. Some amelioration occurred by July 2017 when fentanyl-positive urine samples were ubiquitous among participants reporting nonprescribed opioid use. Tolerance to the adverse effects of higher potency opioids may be developing among users, as some individuals report actively seeking fentanyl.6 Fentanyl was detected in half of the participants in opioid agonist therapy programs in our study, which raises concern for increasing tolerance.3 Our fentanyl assay demonstrates cross-reactivity with other fentanyl analogues. Rapid and specific tests for fentanyl and related analogues are urgently needed, along with innovative treatments.
Corresponding Author: William G. Honer, MD, Department of Psychiatry, University of British Columbia, 2255 Wesbrook Mall, Vancouver, BC V6T2A1 Canada (email@example.com).
Accepted for Publication: December 6, 2017.
Published Online: January 31, 2018. doi:10.1001/jamapsychiatry.2017.4432
Author Contributions: Dr Honer and Ms Jones had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Jones, Panenka, Barr, MacEwan, Thornton, Honer.
Acquisition, analysis, or interpretation of data: Jones, Jang, Panenka, Barr, Thornton, Honer.
Drafting of the manuscript: Jones, Jang, Barr, MacEwan, Honer.
Critical revision of the manuscript for important intellectual content: Jang, Panenka, Barr, Thornton, Honer.
Statistical analysis: Jones, Jang, Barr, Honer.
Obtained funding: Panenka, Barr, Thornton, Honer.
Administrative, technical, or material support: Panenka, Barr, MacEwan.
Study supervision: Panenka, Barr.
Conflict of Interest Disclosures: Dr Honer reports personal fees from Lundbeck and Otsuka, AlphaSights, and Eli Lilly, all outside the submitted work. No other disclosures are reported.
Funding/Support: This work was supported by the Canadian Institutes of Health Research (CBG-101827, MOP-137103) and the British Columbia Mental Health and Substance Use Services, an Agency of the Provincial Health Services Authority. Dr Honer was supported by the Jack Bell Chair in Schizophrenia.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Create a personal account or sign in to: