Every year, thousands of people arrive at emergency departments across the United States having survived an opioid overdose. For many, it is not their first time. For some, it will not be their last. Researchers and clinicians have long searched for effective opioid overdose peer navigator programmes to intervene in that critical window, while someone is still in the hospital and still alive. A new study published in JAMA Network Open in February 2026 put one such programme to the test. The results were sobering and, for the field, instructive.

The Relay Programme: Opioid Overdose Peer Navigators at the Bedside

The intervention under examination was Relay, a programme run by the New York City Department of Health and Mental Hygiene since 2017. Relay is one of the longest-running opioid overdose peer navigator programmes in the country. When a patient arrives at a participating emergency department after a nonfatal opioid overdose, clinical staff call a centralised hotline. A trained peer wellness advocate then travels to the hospital, typically within the hour.

The peer’s role is not clinical. It is human. They sit with the patient, build rapport, and share their own experience of substance use. They provide overdose risk reduction education, hand out naloxone kits, and give care bags with practical items such as snacks and socks. With the patient’s consent, they follow up by telephone or in person for 90 days after discharge, offering support, treatment referrals, and connections to harm reduction services.

The logic is compelling. People who have lived through addiction and recovery can build trust in ways that doctors and nurses, however skilled, sometimes cannot. They speak from experience rather than from a textbook. In a setting as chaotic and often stigmatising as an emergency department, that kind of connection matters.

What the Opioid Overdose Peer Navigator Trial Found

Researchers enrolled 247 adults from four New York City emergency departments between October 2020 and June 2022. They randomly assigned participants either to receive Relay or to a control group receiving site-directed care (SDC). The SDC group was not left without support. For ethical reasons, researchers gave them naloxone kits, overdose response information, a printed list of local treatment programmes, and details about Relay for self-referral.

After 12 months, the trial found no statistically significant difference between the two groups. The primary outcome measured opioid-related adverse events, covering fatal and nonfatal overdoses and any substance use-related emergency department visit.

Relay participants averaged 3.29 such events over the year. The site-directed care group averaged 4.10. The rate ratio was 1.02. That figure is statistically flat. The p-value reached 0.90, about as far from significance as a result can be.

The peer navigator intervention, at least as delivered during this trial, did not measurably reduce harm.

A Strikingly High Death Toll

The most alarming finding applies to both groups equally. Within 12 months of their emergency department visit, nearly one in ten participants had died. Specifically, 24 of 247 participants (9.7%) were dead by the end of follow-up. Of those deaths, 70.8% linked directly to overdose.

That mortality figure is higher than earlier research recorded. A Massachusetts study covering 2011 to 2015 found roughly one in 20 emergency department overdose patients died within a year. The Relay trial authors suggest the rise reflects the increasingly lethal unregulated drug supply. Fentanyl and other synthetic opioids now contaminate much of the street drug market, raising the stakes sharply for every person who uses.

These numbers are a stark reminder. People presenting to emergency departments after opioid overdose are not merely unwell. They face an acute risk of losing their lives.

Why the Peer Navigator Intervention May Have Fallen Short

The researchers are candid about what likely undermined Relay’s effectiveness in this setting. The study ran through the COVID-19 pandemic. That period disrupted transport for peer wellness advocates, shortened hospital stays, and pushed many contacts from in-person meetings to telephone calls. Telephone-only contact reduced the ability to build meaningful rapport.

The follow-up figures tell a clear story. More than half of participants assigned to Relay (53.6%) received zero post-discharge contacts from a peer wellness advocate. Among those who did receive follow-up, half had three contacts or fewer across the entire 90-day period. Peer navigators made just 11 referrals to medications for opioid use disorder (MOUD) across the whole intervention arm.

Structural barriers made things harder still. A significant proportion of participants were homeless or unstably housed, making telephone follow-up unreliable. Around 71.9% of Relay participants and 68.4% of those in the control group had a history of incarceration. Competing pressures of housing insecurity, poverty, and daily addiction management left little room for sustained engagement with support services.

Patients Valued the Connection

Despite the null primary outcome, the study holds an important counterpoint. Among Relay participants who interacted with a peer wellness advocate during their emergency department visit, satisfaction ran high. On a scale of one to ten, participants rated the helpfulness of talking to a peer wellness advocate at an average of 8.74, satisfaction with care at 8.84, and comfort in the conversation at 9.31.

Those scores do not describe people who found the intervention pointless. They describe people who felt heard and valued speaking with someone who genuinely understood their situation.

The real challenge is not whether an opioid overdose peer navigator can matter. It is whether the structures around them are strong enough to turn that initial connection into sustained, measurable change.

The Mortality Question Stays Open

One finding deserves close attention, even though the trial lacked power to confirm it. At 12 months, all-cause mortality sat at 7.2% in the Relay arm and 12.3% in the site-directed care group. A hazard ratio of 0.55 suggests Relay participants may have been roughly half as likely to die during follow-up. Yet the confidence interval was wide (0.22 to 1.30) and the p-value of 0.18 did not reach statistical significance.

The study had only 38% power to detect a mortality difference, even when the true effect was to halve the risk of death. That is a meaningful limitation. The trial did not have enough participants to say anything definitive about survival outcomes. Larger studies with longer follow-up periods need to examine this question directly.

What Needs to Change

Researchers point to several directions for future work. An opioid overdose peer navigator may prove most effective not as a stand-alone intervention but as a bridge to medications for opioid use disorder. Buprenorphine and methadone have the strongest evidence base for reducing overdose mortality. Peers may be well placed to connect patients to these options, especially when trust is high early in a hospital visit.

Relay has already started responding to the trial evidence. Programme leadership moved to strengthen services during the emergency department visit itself, increase referrals, and distribute fentanyl and xylazine test strips. Outreach methods are also under review to better reach people facing severe structural barriers.

There is also a broader point about what trials capture and what they miss. This study did not measure Relay’s impact on emergency department staff morale, on stigma reduction, or on the wellbeing of peer wellness advocates. Qualitative research run alongside the trial found that staff, patients, and peer wellness advocates all valued Relay in ways that do not appear in adverse event counts. A programme can matter deeply to the people inside it, even when hard outcome data prove difficult to shift.

The Wider Picture

The opioid crisis remains one of the gravest public health emergencies in the United States. Overdose has driven premature death for years. A modest decline in 2023 offered some hope, but the numbers remain devastating. Emergency departments sit at the centre of this crisis. They see patients at their most vulnerable and, crucially, at their most reachable.

A recent Canadian study found that 70% of people who died from an opioid-involved overdose had visited an emergency department in the year before their death. These visits are critical touchpoints. They are opportunities that, if missed, may not come again.

This trial does not argue that those opportunities should be abandoned. It argues they need far more investment. More intensive follow-up, stronger links to treatment, and serious effort to address the structural conditions of poverty, homelessness, and criminalisation are all required.

Opioid overdose peer navigators may yet prove to be one of the most important tools in this fight. What this trial tells us is that goodwill and human connection, while necessary, are not enough on their own. The infrastructure around peer workers must match the scale of the problem they face.

Published in JAMA Network Open, February 2026. Trial registration: ClinicalTrials.gov NCT04317053. Funded by the Centres for Disease Control and Prevention.

Source: jamanetwork