Introduction

Opioids represent among the most powerful analgesics available and are widely used for acute and chronic pain.1 Although highly effective for pain relief, opioids have serious side effects and substantial risks, including misuse, dependence, and overdose.2 Overdose deaths involving prescription opioids increased substantially beginning in the 1990s but have declined in recent years. Since 2013, overdose deaths involving synthetic opioids, particularly illicitly manufactured fentanyl and fentanyl analogues, have risen sharply.3 In 2023, approximately 105,000 people in the U.S. died from drug overdose,3 and nearly 73,000 of these deaths involved synthetic opioids, primarily fentanyl,4 accounting for approximately 92% of all opioid-involved overdose deaths. These trends reflect the evolving nature of the opioid crisis, with recent increases driven by the growing availability of high-potency synthetic opioids.

Fentanyl plays a critical role as a potent synthetic opioid used for managing severe pain, particularly in patients with cancer and in post-surgical settings,4,5 but its high potency also increases the risk of overdose, making its use tightly regulated. Fentanyl is 100 times more powerful than morphine and heroin,6 and its analogues can be even more potent. Even small doses can lead to fatal overdose, and both non-fatal and fatal overdoses involving fentanyl have risen substantially over the past decade.7 Legal pharmaceutical fentanyl is prescribed or administered under medical supervision to treat severe pain.8 In contrast, illicitly manufactured fentanyl is distributed through illicit drug markets,9,10 and has emerged as a major driver of recent fentanyl-related overdoses11 and overdose deaths.12 Individuals with opioid use disorder frequently use illicit fentanyl, markedly increasing their risk of overdose and death.13

Despite extensive evidence on the adverse outcomes associated with illicit fentanyl use, it is difficult to track individual-level illicit fentanyl exposure because such use often occurs outside healthcare systems14 and is not directly captured in clinical data, limiting accurate estimation of its true scale and impact. Urine drug testing (UDT) serves as a valuable diagnostic tool across medical disciplines, including addiction medicine,15 and plays a critical role in detecting nonprescribed or illicit drug use.16,17 Previous studies have used UDT data to track drug trends, including changes in nonprescribed fentanyl co-detection with cocaine or methamphetamine,18,19 and to estimate overdose deaths using aggregated UDT data.17 However, very few of these studies have efficiently differentiated between medical-source and illicit-source fentanyl exposure, leaving a critical gap in understanding how the source of fentanyl exposure influences opioid-related harmful outcomes.

Leveraging patient-level electronic health records (EHRs), we inferred potential illicit-source fentanyl (non-medical) exposure by linking UDT results with documented fentanyl-related medication records using a sequential time-window screening method to differentiate medical- and illicit-source fentanyl exposure. We characterized longitudinal and regional patterns in opioid-related harmful outcomes stratified by the source of fentanyl exposure and estimated the risk of the outcomes associated with illicit-source fentanyl exposure.