The clinical problem behind the study is rooted in the transformation of the illicit drug supply. In British Columbia and across much of North America, unregulated fentanyl has largely displaced heroin, and most people entering hospital with opioid use disorder are now fentanyl-dependent. Traditional induction onto opioid agonist treatment, which typically relies on methadone or buprenorphine started at low doses, was designed around the pharmacology of shorter-acting opioids. Because fentanyl is extraordinarily potent, highly lipophilic, and accumulates in fat stores from which it slowly re-releases, patients dependent on it can experience severe, protracted withdrawal even when standard buprenorphine initiation protocols are followed. Precipitated withdrawal, in which the partial agonist buprenorphine abruptly displaces fentanyl from opioid receptors and triggers sudden, intense withdrawal symptoms, is a well-recognized failure mode that can cement a patient’s distrust of hospital-based care.

SIFI approaches the problem from the opposite direction. Rather than starting with a medication designed to displace fentanyl, clinicians use fentanyl itself, delivered intravenously in controlled, repeated doses, to suppress withdrawal symptoms until the patient is physiologically stable enough to transition onto a longer-acting opioid agonist such as slow-release oral morphine or methadone. The protocol is guided by structured monitoring instruments, including the Clinical Opioid Withdrawal Scale, which quantifies withdrawal severity, and the Pasero Opioid-Induced Sedation Scale, which grades sedation depth. Nurses assess respiratory rate and sedation before each dose, and administration is inhibited when clinical signs indicate accumulating opioid effect. In principle, this titration logic mirrors how any hospital manages pain with intravenous opioids, but here the goal is not analgesia alone; it is a deliberate, supervised pharmacological bridge from the chaotic unregulated supply to regulated, prescribed therapy.

The research team, drawn from Vancouver General Hospital’s Complex Pain and Addiction Service and collaborating institutions including the University of British Columbia and the BC Centre for Excellence in HIV/AIDS, conducted a retrospective chart review of patients prescribed the SIFI protocol at their institution. The investigators registered the study on ClinicalTrials.gov before data collection and examined a battery of outcomes: adverse events during induction, self-directed discharges during or after the protocol, whether the primary reason for hospital admission was ultimately addressed, transition onto opioid agonist treatment, emergency department visits or readmissions within thirty days of discharge, and the dosing characteristics of the induction itself. The cohort comprised seventeen patient encounters, involving sixteen individuals, one of whom underwent SIFI during two separate admissions.

The safety findings are the study’s most reassuring result. Across all encounters, SIFI produced transient sedation in three patients, none of whom required resuscitation, and no serious adverse events were recorded. This matters because fentanyl’s potency and its tendency to accumulate in tissue raise legitimate concerns about respiratory depression when the drug is administered repeatedly. The structured sedation scoring embedded in the protocol appears to have caught accumulating opioid effect before it reached dangerous territory. The authors caution, appropriately, that the cohort is small and the design retrospective, so the safety signal must be confirmed in larger, prospective studies. Nonetheless, in a population where untreated withdrawal itself carries serious risks, including dehydration, cardiac strain, and profound distress that drives patients back to the toxic unregulated supply, even preliminary evidence of tolerability is consequential.

The efficacy signals, while more mixed, are striking in places. Compared with admissions in the three months before SIFI was used, patients who received the induction stayed dramatically longer in hospital before leaving against medical advice. Mean time in hospital before self-directed discharge rose from 0.2 days to 9 days for the entire cohort, and among patients who did not leave against advice, the average reached 49.2 days. In practical terms, patients who previously walked out within hours of admission remained long enough for clinical teams to address the infections, wounds, or other conditions that brought them in. Most patients had their primary reason for admission addressed, six patients were established on opioid agonist treatment at discharge, and the cohort showed lower readmission rates in the month following admission compared with their historical pattern.

The results are not uniformly favorable. The proportion of admissions ending in self-directed discharge actually increased after SIFI, from 51 percent of encounters in the pre-induction period to 65 percent afterward. The authors interpret this pattern cautiously, noting that the comparison periods and denominators differ and that patients selected for SIFI may represent a particularly high-acuity group whose risk of leaving against advice is extreme. Even so, the finding underscores a central tension in addiction hospital medicine: an intervention can extend stays and improve clinical completion rates for many patients while the overall rate of early departure remains stubbornly high. Withdrawal management during hospitalization, however well executed, competes with the pull of the unregulated supply, the discomfort of institutional settings, and the cumulative distrust many patients carry from prior experiences with the health system.

Why does an intravenous fentanyl induction resonate so strongly with patients who have resisted other forms of treatment? The answer likely lies in the pharmacology of tolerance. A person using unregulated fentanyl daily may require doses of buprenorphine or methadone far above conventional starting amounts to feel any effect, and standard low-dose initiations can leave them in days of untreated withdrawal. By titrating intravenous fentanyl to symptom control, SIFI effectively replicates the doses patients have been self-administering, then transitions them to a regulated, longer-acting alternative once stability is achieved. Slow-release oral morphine, which is used in some Canadian programs for patients who do not respond to first-line agonists, features in the protocol’s transition pathway. The strategy acknowledges a harm-reduction principle that has gained traction in Canadian medicine: for some patients with severe fentanyl use disorder, the pragmatic goal is a regulated opioid supply rather than immediate abstinence.

Broader context sharpens the significance of these findings. British Columbia declared a public health emergency over illicit drug toxicity in 2016, and deaths from unregulated fentanyl have remained catastrophically high since. Hospital admissions represent one of the few guaranteed contact points between this population and the health system, yet they are too often wasted opportunities punctuated by abrupt, withdrawal-driven departures. Interventions that lengthen hospital stays, address the presenting medical problem, and link patients to ongoing treatment could bend mortality curves that have resisted conventional approaches. The study was funded through donor support to the VGH and UBC Hospital Foundation, received ethics approval from the University of British Columbia Clinical Research Ethics Board, and was conducted in accordance with the Declaration of Helsinki, with informed consent obtained from participants.

The path forward will require the studies this retrospective review cannot replace: prospective trials with larger cohorts, comparator arms testing SIFI against standard buprenorphine or methadone inductions, and careful attention to dosing parameters that the chart review only partially captures. Questions about optimal titration ceilings, the ideal timing and target of transition to long-acting agonists, and generalizability beyond a single quaternary hospital with an established addiction consult service all remain open. Still, the Vancouver team’s data sketch a plausible answer to one of addiction medicine’s most frustrating failure modes. For patients in the grip of fentanyl dependence, arriving at a hospital has too often meant arriving at a doorway they quickly exit. A protocol that suppresses withdrawal with the drug their bodies know, monitors sedation with validated scales, and builds a bridge to regulated treatment turned hours-long stays into stays averaging nine days, addressed the medical problems that brought patients through the door, and left a meaningful fraction on therapy at discharge. In a crisis defined by toxic supply and revolving-door admissions, that combination of safety and engagement is a result worth rigorous pursuit.

Subject of Research: Symptom-inhibited fentanyl induction as a hospital-based opioid agonist therapy initiation protocol for patients with fentanyl-dependent opioid use disorder

Article Title: Symptom-inhibited fentanyl induction (SIFI) for hospitalized patients with opioid use disorder: a retrospective chart review

Article References: Li, V. W., Wong, J. S. H., Kim, J. J., Machado, J., Hawamdeh, A. A., Schneiderman, H., Nalliah, S., Harris, M., Montaner, J. S. G., Ignaszewski, M. J., & Azar, P. (2026). Symptom-inhibited fentanyl induction (SIFI) for hospitalized patients with opioid use disorder: a retrospective chart review. Addiction Science & Clinical Practice. https://doi.org/10.1186/s13722-026-00723-w

Image Credits: AI Generated

DOI: 10.1186/s13722-026-00723-w

Keywords: opioid use disorder, fentanyl, SIFI, opioid agonist treatment, hospitalized patients, withdrawal management, self-directed discharge, buprenorphine, slow-release oral morphine, Vancouver General Hospital, retrospective chart review, harm reduction