As overdose deaths tripled, recorded drug-related suicides barely moved. Researchers argue a new measure — self-injury mortality — would give a truer picture of the crisis.

If you or someone you know is struggling, help is available. In the U.S., call or text 988 to reach the 988 Suicide & Crisis Lifeline.

Public health begins with counting. Before a problem can be addressed, it has to be measured — its size, its trends, the people it affects. A group of suicide researchers now warns that when it comes to suicide in the United States, the counting itself has broken down, and the official numbers may be missing a substantial share of the deaths.

Writing in the journal Injury Prevention, researchers led by injury epidemiologist Ian Rockett of University of Rochester Medicine argue that the gap between the country’s soaring drug-poisoning deaths and its relatively flat count of drug-related suicides points to widespread misclassification.

“In public health, the first step is to know the size of a problem,” said Eric Caine, a professor emeritus of Psychiatry and co-founder and past director of the Center for the Study and Prevention of Suicide at University of Rochester Medicine. “Misclassification and undercounting break this chain.”

Numbers that don’t add up

The figures the authors highlight are stark. The national suicide rate in 2024 was 37.5 percent higher than in 2000. Over the same period, the death rate from opioid and other drug poisoning more than tripled.

Yet the recorded rate of suicide by drug poisoning rose only modestly, from 1.3 to 1.8 deaths per 100,000 people.

To Rockett, who has studied the undercounting of injury deaths for more than three decades, that disconnect is a warning sign. He believes that as many as 30 percent of overdose deaths recorded as “accidental” or of “undetermined” intent may actually be suicides.

The problem of intent

Part of the difficulty lies in how deaths are classified. When someone dies of a drug overdose, a medical examiner or coroner must decide whether the death was an accident, a suicide, or of undetermined intent. That judgment depends on evidence about what the person meant to do — a note, a history, statements to others. In many cases, that evidence simply isn’t available.

When intent cannot be established, medical examiners and coroners often default to classifying the death as an accident. The researchers say that tendency is reinforced by a range of pressures: the stigma that still surrounds suicide, limited expertise and investigative resources in many medical examiner and coroner offices, inconsistent training, and the enormous strain placed on those offices by the opioid epidemic, rising suicide rates, and the COVID-19 pandemic.

Caine argues that the very framing of “accidental” overdose may be misleading. “Most drug overdoses are the result of motivated behaviors and are not ‘accidents’ in the strict sense,” he said. “Recurrent illicit drug use, especially with drugs from unknown sources, is really self-injury.”

A different way to count

Rather than trying to divine intent after the fact, Rockett proposes a new measure called self-injury mortality, or SIM. It combines registered suicides with most deaths from drug self-intoxication into a single category. The focus shifts from what a person intended in their final moments to the behavior that preceded their death.

According to Caine, the approach offers “a clearer understanding of the behavior without getting caught up in intentionality questions.”

The authors suggest a practical step to support the measure: adding a SIM checkbox to death certificates, which would help distinguish true accidents from deaths resulting from deliberate acts such as using illicit drugs.

The proposal drew support from two commentators with different professional vantage points — Steven Stack, a sociologist at Wayne State University, and Kurt Nolte, a forensic pathologist at the University of New Mexico School of Medicine. Both agreed that SIM is a well-conceived and more inclusive measure.

Why the count matters

Undercounting is not merely a statistical nuisance. Prevention programs are designed, funded, and evaluated on the basis of these numbers. If a large share of self-inflicted deaths is hidden within the “accidental overdose” category, policymakers may underestimate the scale of the problem, direct resources to the wrong places, or misjudge whether interventions are working.

A more inclusive measure could also help bridge two public health efforts that have often operated separately: suicide prevention and the response to the overdose crisis. Seen through the lens of self-injury mortality, they may be facets of the same underlying problem, one rooted in despair, distress, and behaviors that put people at grave risk.

The proposal is not without complications. Grouping intentional and unintentional deaths together could blur distinctions that matter for some prevention strategies, and changing how deaths are recorded nationally would require coordination among thousands of medical examiners and coroners. But the authors’ central point is hard to dispute: a crisis cannot be confronted effectively if it cannot be seen clearly.

As Caine put it, accurate measurement is the foundation for everything that follows — gauging the problem, designing responses, and knowing whether those responses save lives.

If you or someone you know is in crisis, call or text 988, or chat at 988lifeline.org, to reach the 988 Suicide & Crisis Lifeline.

---

Notes

1. Rockett, I. R. H., et al. “Deterioration of American suicide accounting as rates rise during the 21st century and responses by suicidologists with respective expertise in psychiatry, sociology and forensic pathology.” Injury Prevention, July 14, 2026. DOI: 10.1136/ip-2025-046146.
2. University of Rochester Medical Center. “Experts warn suicides are undercounted in the US.” News release via EurekAlert!, July 27, 2026. https://www.eurekalert.org/news-releases/1137718