Substance Use Disorder Deaths in America Soared Sixfold Before Finally Turning Downward

A sweeping quarter-century analysis of United States mortality data has revealed the true scale of the nation’s substance use crisis, documenting more than 7.3 million deaths linked to substance use disorders between 1999 and 2024. The study, published in BMC Psychiatry, draws on the Centers for Disease Control and Prevention’s WONDER database, a publicly accessible repository of every recorded death certificate in the country. Researchers led by Salauddin Ali Khan and colleagues identified deaths among adults aged 15 and older in which substance use disorders appeared as either the underlying cause of death or a contributing factor, using the International Classification of Diseases codes F10 through F19, which cover alcohol, opioids, cocaine, sedatives, and other psychoactive substances. The findings chart a devastating arc of American mortality: a slow climb through the 2000s, an acceleration through the 2010s, and a catastrophic spike during the pandemic years before the first sustained decline in decades.

The headline numbers are stark. In 1999, the age-adjusted mortality rate for substance use disorders stood at 23.4 deaths per 100,000 population. By 2021, that figure had peaked at 147.9 per 100,000, a more than sixfold increase in just over two decades. The years since have brought a measure of relief, with the rate falling to 125.4 per 100,000 in 2024. Age adjustment is a critical statistical technique here: because populations age over time, raw death counts can be misleading. By standardizing to a reference age distribution, epidemiologists can compare mortality across years and demographic groups on equal footing, ensuring that the trends reflect genuine changes in risk rather than shifts in the age structure of the population itself.

To dissect these trends with precision, the research team employed joinpoint regression, a method designed to detect inflection points in mortality time series. Rather than fitting a single straight line to twenty-six years of data, joinpoint analysis identifies the years at which the trajectory changes direction and calculates the annual percent change for each segment. This approach allows investigators to pinpoint exactly when the crisis worsened, when interventions may have taken effect, and whether the recent decline represents a durable turning point or a temporary reprieve. The technique has become a mainstay of cancer and overdose surveillance precisely because it converts noisy annual figures into interpretable phases of acceleration and deceleration.

The demographic breakdown exposes profound inequities. Men died at more than twice the rate of women, with an age-adjusted mortality rate of 138.5 per 100,000 compared with 67.4 for women, a gap that reflects both higher rates of substance use and greater exposure to high-risk patterns such as fentanyl-contaminated opioid supplies. Among racial and ethnic groups, American Indian or Alaska Native populations bore the heaviest burden at 114.4 per 100,000, followed by White individuals at 104.3 and Black or African American individuals at 94.4. Hispanic populations recorded the lowest rate at 45.3 per 100,000, a figure that researchers often attribute in part to the so-called Hispanic paradox, though the authors caution that such aggregate comparisons can mask heterogeneity within groups and differences in death certificate classification practices.

Age patterns add another layer of complexity to the crisis. Adults aged 85 and older showed the highest mortality burden of any group, at 588.2 per 100,000, a figure that underscores how substance use disorders, particularly alcohol-related disease, accumulate lethal consequences over decades of exposure. While much public attention has focused on young and middle-aged adults felled by opioid overdoses, the data reveal that the cumulative toll of chronic substance use among the oldest Americans is staggering. This finding challenges the perception of substance use disorder as exclusively a disease of the young and highlights the need for screening, treatment, and harm-reduction services that extend into geriatric care settings, where substance problems are frequently overlooked or misattributed to aging itself.

Geography proved equally consequential. The Midwest recorded the highest regional mortality rate at 123.8 per 100,000, followed by the South at 98.2, the Northeast at 97.5, and the West at 79.2. At the state level, the spread was even more dramatic, ranging from 41.4 per 100,000 in Alabama to 194.4 per 100,000 in Vermont, a nearly fivefold difference that reflects divergent drug supply patterns, prescribing cultures, treatment access, and policy environments. The prominence of New England states at the top of the distribution aligns with the region’s early and intense exposure to synthetic opioids, while the lower rates in parts of the Deep South may partly reflect differences in death investigation and coding practices as well as genuinely different epidemiology.

Perhaps the most consistent geographic signal in the entire analysis was the urban-rural divide. Nonmetropolitan areas experienced an age-adjusted mortality rate of 121.1 per 100,000, substantially higher than the 91.2 recorded in metropolitan counties. This gap has widened over the course of the study period and speaks to a convergence of structural disadvantages: rural communities face shortages of addiction medicine specialists, longer travel distances to treatment facilities, weaker harm-reduction infrastructure, and economic distress that fuels substance use. The rural overdose crisis, which accelerated after fentanyl penetrated heroin and methamphetamine markets in the 2010s, has left many small communities with mortality rates that rival or exceed those of the worst-affected urban neighborhoods during the crack era, yet with far fewer resources to respond.

The recent decline since the 2021 peak offers genuine grounds for cautious optimism, and the timing is informative. The drop coincides with the expansion of naloxone distribution, the growth of medication-assisted treatment with buprenorphine and methadone, increased availability of fentanyl test strips following federal policy changes, and the stabilization of the post-pandemic environment. Yet the 2024 rate of 125.4 per 100,000 remains more than five times the level recorded in 1999, meaning that even after three years of improvement, Americans are dying from substance use disorders at a pace that would have been unthinkable a generation ago. The authors emphasize that the burden remains substantially elevated and that the gains are fragile, unevenly distributed, and vulnerable to reversal if the drug supply shifts again or if funding for treatment and prevention contracts.

What emerges from this analysis is a portrait of a public health failure measured in millions of lives, but also a roadmap for targeted intervention. The persistence of stark disparities across sex, race and ethnicity, age, region, and urbanicity means that a one-size-fits-all national strategy will inevitably leave the hardest-hit communities behind. The study’s authors call for prevention, harm-reduction, and treatment strategies tailored to the populations and places where mortality concentrates: culturally grounded services for American Indian and Alaska Native communities, expanded access in rural counties, geriatric-aware addiction care, and sustained investment in the Midwest and New England. As the United States enters what may be a new phase of the epidemic, the data make clear that the recent downward trend is an opportunity, not a victory, and that closing the gaps documented here will determine whether the next quarter-century tells a different story.

Subject of Research: Long-term mortality trends and demographic and geographic disparities of substance use disorders among U.S. adults, 1999 to 2024

Article Title: Mortality trends of substance use disorders among adults in the United States, 1999–2024: a CDC WONDER analysis

Article References: Khan, S. A., Kumar, R., Arain, M. T., Qadir, A., Das, K., Dawood, M. H., & Yaqoob, A. (2026). Mortality trends of substance use disorders among adults in the United States, 1999–2024: a CDC WONDER analysis. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08649-y

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08649-y

Keywords: substance use disorders, mortality trends, CDC WONDER, health disparities, drug overdose, rural health, opioid crisis, age-adjusted mortality, joinpoint regression, United States, public health, BMC Psychiatry