U.S. overdose deaths fell to about 70,000 in 2025, yet 84 percent of those needing substance use care went without it. Operators face both numbers.
Key Takeaways
- The number that moved: Federal provisional data put 2025 overdose deaths near 70,000, a third fewer than the 2022 peak and the third straight annual drop.
- The funding rationale: Much of the case for public addiction spending rested on rising mortality, and that argument carries less force as deaths fall.
- The need that remained: Roughly one in six Americans who required addiction care in 2025 received it, according to SAMHSA’s national survey.
- The operator’s shift: Growth now depends less on crisis volume than on reaching untreated patients, keeping them in care and documenting outcomes payers can price.
In May, the National Center for Health Statistics released a number that few people in addiction treatment would have predicted three years earlier. Provisional data, the agency said, indicated an estimated 69,973 drug overdose deaths in the United States during 2025, a decrease of almost 14 percent from 2024 and the third consecutive annual decline. Opioid-involved deaths fell from an estimated 55,296 to 44,564. Five states (Rhode Island, New York, North Carolina, Alabama and Vermont) recorded declines of 25 percent or more. The final count for 2022, published in an NCHS data brief, had been 107,941.
The addiction treatment sector spent the past decade describing itself through that curve. Mortality was the statistic in appropriations testimony, in settlement negotiations and in investor presentations, and its upward slope carried an argument that more capacity was needed immediately. The slope has turned. The disease has not, at least not at a comparable pace. That gap is now a business condition that operators, payers and investors have to price.
How Far U.S. Overdose Deaths Have Fallen
The decline has been visible for some time. In February 2025, the CDC reported roughly 87,000 deaths in the 12 months ending September 2024, down from about 114,000 the year before, and called it the lowest total for any 12-month period since June 2020. “It is unprecedented to see predicted overdose deaths drop by more than 27,000 over a single year,” Allison Arwady, speaking for the CDC’s National Center for Injury Prevention and Control, said in that release. The 2025 estimate extends that trend by another large step. Deaths involving cocaine and psychostimulants such as methamphetamine declined as well, according to NCHS, a detail that matters in a system where polysubstance use is common and where stimulants still lack an FDA-approved medication, as Acuity detailed in its reporting on the stimulant crisis.
The figures carry caveats that belong in any plan built on them. The CDC’s provisional overdose dashboard, updated monthly, warns that declining counts “could be due to incomplete data, true decreases in the number of deaths, or a combination of the two,” and that true trends cannot be confirmed until final data are available. The decline is also uneven. NCHS reported that New Mexico, Arizona and Colorado each saw increases of 10 percent or more in 2025.
That unevenness has direct consequences for multi-state operators. A company with clinics in Rhode Island and New Mexico now faces two different local markets inside one portfolio, with deaths down by a quarter or more in one and rising by at least a tenth in the other. National averages can hide exactly the regional divergence that determines where new capacity, outreach budgets or clinical staff should go, and the monthly jurisdiction-level counts on the CDC dashboard are more useful for that planning than the annual headline.
SAMHSA Survey Shows the Substance Use Treatment Gap Persists
Mortality captures the gravest outcome of addiction rather than its reach. For reach, the standard reference is SAMHSA’s National Survey on Drug Use and Health, which the agency published for 2025 in late July, in what it called the earliest public release in the survey’s history. The survey estimates that 47.2 million people aged 12 or older needed substance use treatment in the past year. Of those, 16.0 percent, or 7.6 million people, received it. The remaining 84.0 percent, about 39.6 million people, did not.
SAMHSA’s summary of the release put the number meeting criteria for a substance use disorder at nearly 45 million, 15.3 percent of the population 12 and older, and the survey counts 4.0 million people with an opioid use disorder. These are self-reported estimates of the civilian, noninstitutionalized population, so people in jails, prisons and many residential settings fall outside them. Even within those limits, the survey describes a treatment market in which the unserved population is roughly five times the served one.
What the Overdose Decline Looks Like Inside Treatment Operations
Public company filings offer a partial view of operating conditions, with two cautions: one company is not a market, and none of these figures isolate the effect of mortality trends. Acadia Healthcare, which runs a network of opioid treatment clinics it calls comprehensive treatment centers, reported second-quarter 2026 revenue of $141.2 million for that segment, essentially flat against $141.5 million a year earlier. In its first-quarter results, the company attributed a 6.5 percent decline in specialty treatment facility revenue to facilities in Pennsylvania and to closures after the first quarter of 2025. Its residential treatment revenue, by contrast, rose 11.6 percent in the second quarter.
The wider facility base is large and varied. SAMHSA’s 2024 survey of treatment facilities counted 17,829 eligible facilities providing substance use treatment, and their economics differ sharply by setting. An opioid treatment program paid through a bundled rate depends on keeping patients enrolled week after week, while a residential program depends on admissions and length of stay. Office-based buprenorphine practices, including those adopting the injectable formulations Acuity examined this month, depend on prescriber capacity and payer mix. A falling death count touches each model differently, and none of them in a straight line.
The same survey hints at where the field is already positioned for the next phase. Of the eligible facilities it counted, 9,302 provided both substance use and mental health treatment, a little more than half of the substance use total. Those combined programs are better placed to reach patients whose first contact with care is for anxiety, depression or a psychiatric crisis rather than for addiction itself, a group whose relative weight rises if fewer patients arrive by way of an overdose.
How Falling Mortality Changes the Addiction Treatment Funding Argument
The funding side is where a lower death count may register first. Public spending on addiction treatment has long been argued in terms of lives lost, and that framing loses some of its force as the figure declines. Medicaid remains the largest payer, and its coverage is shifting under the 2025 reconciliation law’s work requirements and their behavioral health exemptions, which Acuity covered in June. Opioid settlement payments run on fixed schedules into the late 2030s and do not adjust to mortality trends, though how quickly those dollars reach providers is a separate matter.
As the emergency framing recedes, the terms of the payer conversation may shift with it. A program justified by overdose prevention can point to naloxone distribution and warm handoffs from emergency departments. A program asked to justify itself on outcomes has to show retention, lower acute utilization and recovery measures that hold up in a contract. The reimbursement gap Acuity described in May, in which evidence-based services are paid inconsistently, becomes more consequential when urgency no longer carries the budget argument on its own.
Retention and Untreated Patients as the Next Growth Measures
If mortality stops being the dominant metric, the untreated population becomes the obvious replacement. Reaching some share of the 39.6 million who went without care requires channels different from a crisis-driven intake system: primary care integration, peer outreach, co-occurring mental health care and payment models that fund the work between visits. Several of those channels already have reimbursement structures. Certified Community Behavioral Health Clinics, paid a clinic-specific cost-based rate, are expanding to ten more states between July 2026 and July 2027. Peer recovery support is covered by Medicaid in 48 states, although, as Acuity found, H0038 rates and billing rules vary widely.
Integration with mental health care is the other lever, and the billing silos that separate the two remain a practical barrier for organizations that treat both conditions. None of these channels replaces crisis capacity, which still matters in the states where deaths are rising. Together they describe a market whose growth depends less on the size of an emergency and more on the reach and durability of care.
NCHS will keep publishing monthly provisional counts, and the 2025 figure will be revised as records close. SAMHSA’s next survey will show whether the 84 percent figure moves at all. For an industry that once measured its urgency by the first number, the second is the one that describes the size of the market still waiting to be served.
