Dual Diagnosis Statistics 2026: How Many Americans Have Co-Occurring Disorders, the Treatment Gap, and New England Data

A dual diagnosis, having a mental health condition and a substance use disorder at the same time, is one of the most common situations in behavioral health and one of the most commonly misreported. Much of what circulates online still cites survey data from 2018 or earlier, even though the federal survey behind those numbers changed its data collection methods in 2021 and substantially revised its treatment questions after that.

This page gathers the current numbers in one place: how many Americans have co-occurring disorders, which conditions and substances pair most often, what happens to veterans and young people, how few people receive integrated care, and how New Hampshire and Massachusetts compare with the rest of the country. Every statistic carries its source and survey year, so you can check each one yourself.

Dual diagnosis key numbers at a glance

• 21.2 million U.S. adults had both any mental illness (AMI) and a substance use disorder (SUD) in the past year (2024 National Survey on Drug Use and Health, released July 2025).
• About 1 in 3 of the 61.5 million adults with any mental illness also had a substance use disorder (2024 NSDUH).
• Nearly half of the 14.6 million adults with serious mental illness (SMI), 6.9 million people, also had a substance use disorder (2024 NSDUH).
• Only 14.5 percent of adults with co-occurring AMI and SUD received treatment for both conditions in 2024; 41.2 percent received neither (2024 NSDUH).
• 13.3 percent of young adults aged 18 to 25 had co-occurring AMI and SUD in 2024, the highest rate of any age group (2024 NSDUH).
• 792,000 adolescents aged 12 to 17 had both a major depressive episode and a substance use disorder in 2024 (2024 NSDUH).
• 21.3 percent of veterans who served in a combat zone had a past-year mental illness, and 12.3 percent had an alcohol use disorder (2022-2024 NSDUH pooled data, published November 2025).
• 8.7 percent of New Hampshire adults and 8.9 percent of Massachusetts adults had co-occurring SUD and mental illness, versus 8.0 percent nationally (2023-2024 NSDUH state estimates).
• Provisional CDC data predict 69,147 U.S. drug overdose deaths for the 12 months ending January 2026, a 13.2 percent one-year decline (CDC National Vital Statistics System, June 2026 release).

What counts as a dual diagnosis, and how it is measured

Clinicians use the terms dual diagnosis, co-occurring disorders, and comorbidity interchangeably: a person meets criteria for at least one mental health condition and at least one substance use disorder within the same period. SAMHSA's co-occurring disorders overview lists three reasons the conditions travel together: some substances can produce symptoms of a mental health problem, some people use alcohol or drugs to self-medicate a mental health condition, and the two kinds of disorder share underlying causes such as genetic vulnerability and early exposure to stress or trauma. Its treatment guidance is blunt: someone with a mental health problem and a substance use disorder must treat both issues.

The national numbers on this page come mostly from the National Survey on Drug Use and Health (NSDUH), which distinguishes three measures:

• Any mental illness (AMI): a diagnosable mental, behavioral, or emotional disorder in the past year, of any severity, excluding developmental and substance use disorders.
• Serious mental illness (SMI): the subset of AMI that substantially interferes with or limits one or more major life activities.
• Substance use disorder (SUD): meeting DSM-5 criteria for an alcohol use disorder or a drug use disorder in the past year.

One measurement note worth knowing before quoting any of these figures: in the 2024 NSDUH annual report (2025), a person counts as having co-occurring disorders when they met criteria for both conditions in the same 12-month window, which is not the same as showing that both were active in the same week or month. For a plain-language walkthrough of how these conditions interact in treatment, see our guide to integrating mental health care in addiction treatment.

How many American adults have co-occurring disorders?

The 2024 NSDUH annual national report, released in July 2025, is the most current national estimate. Among adults aged 18 or older in 2024:

• 33.0 percent, or 86.6 million adults, had either AMI or an SUD in the past year.
• 61.5 million adults (23.4 percent) had any mental illness.
• 46.3 million adults had a substance use disorder. Counting everyone aged 12 or older, 48.4 million people (16.8 percent) had an SUD, including 27.9 million with an alcohol use disorder and 28.2 million with a drug use disorder.
• 21.2 million adults had both AMI and an SUD. That is about one third of all adults with mental illness, and slightly less than half of all adults with a substance use disorder.

By age, co-occurring AMI and SUD affected 8.1 percent of all adults in 2024: 13.3 percent of young adults aged 18 to 25, 11.3 percent of adults aged 26 to 49, and 3.8 percent of adults 50 or older (2024 NSDUH). Put plainly, that is how common dual diagnosis is: about 1 in 12 American adults had both conditions in a single year.

The overlap runs in both directions, which matters for screening. An adult with a mental illness has roughly a 1 in 3 chance of also meeting criteria for a substance use disorder, and an adult with a substance use disorder has close to a 1 in 2 chance of also having a mental illness (2024 NSDUH). This is why programs that treat depression, anxiety, and addiction together screen every patient for both.

Why you cannot compare today's numbers with the old 9.2 million figure

Many widely shared statistics pages still report that 9.2 million American adults have co-occurring disorders. That number is real, but it is from the 2018 NSDUH (published 2019), when 9.2 million adults, 3.7 percent, had both AMI and an SUD.

The jump from 9.2 million to 21.2 million is not a measured doubling of the condition. SAMHSA redesigned the survey in 2021, moving from in-person-only interviews to multimode collection, and states plainly in the 2024 report that estimates from 2021 onward "are not comparable with estimates from 2020 or prior years." Definitions also shifted, including the move to DSM-5 substance use disorder criteria, which widened who counts as having an SUD. Real increases and methodological changes are mixed together, so the honest statement is: the current, correctly measured figure is 21.2 million (2024 NSDUH), and the 9.2 million figure describes a different survey era and should be dated as 2018 whenever it is quoted.

Which comes first: mental illness or substance use?

Researchers have studied the directionality question for decades, and the evidence does not support a single answer. The National Institute on Drug Abuse (2024) describes three pathways that often operate at once:

1. Shared risk factors. Genetics, adverse social environments, trauma, and chronic stress raise the risk of both conditions. One analysis NIDA cites estimates that over 30 percent of adults with a substance use disorder experienced childhood trauma such as abuse or neglect.
2. Mental illness leading to substance use. People may use alcohol or drugs to blunt symptoms of depression, anxiety, or post-traumatic stress. Relief is temporary, and the underlying condition typically worsens. For PTSD specifically, the VA National Center for PTSD (2025) reports that research suggests PTSD usually develops first for most people who have both PTSD and a substance problem.
3. Substance use contributing to mental illness. Substances change some of the same brain circuits disrupted in mental disorders. NIDA notes evidence that cocaine use can worsen bipolar disorder and that cannabis use is linked to earlier onset of psychosis in people with genetic risk factors.

The National Institute of Mental Health maintains an overview of these interactions on its substance use and mental health topic page. The clinical consequence is the same regardless of sequence: NIDA states that research shows integrated treatment, which addresses both conditions together, leads to better health outcomes.

"In practice it rarely matters which condition came first, because by the time someone reaches care the two are feeding each other. Treat the substance use and leave the depression or the trauma untouched, and the untreated half tends to pull the person back. Integrated care at the outpatient level means one team, one plan, and both conditions on the table in every session."

Dr. Richard Marasa, Medical Director, Clear Steps Recovery

Common dual diagnosis pairings, with current data

No ranking page in this topic reliably sources its pairing claims, so here is what the primary literature actually shows for the five pairings clinicians see most.

PTSD and alcohol

In the United States, over 4 in 10 adults (45 percent) who have PTSD also have problems with drug or alcohol use, according to the VA National Center for PTSD (2025). In the same research, veterans who had PTSD at some point in their lives were 2 times more likely to have problems with alcohol use and 3 times more likely to have problems with drug use than veterans without PTSD.

Depression, mood disorders, and opioids

A national study in the Journal of the American Board of Family Medicine (Davis et al., 2017) found that 18.7 percent of adults with mood and anxiety disorders used prescription opioids, versus 5.0 percent of adults without those conditions, and that adults with mental health disorders received 51.4 percent of all opioid prescriptions dispensed in the United States, roughly 60 million of 115 million prescriptions per year.

Anxiety and benzodiazepines

An analysis of national survey data in Psychiatric Services (Maust et al., 2019) found that 30.6 million U.S. adults (12.6 percent) used benzodiazepines in a year, 5.3 million of whom (2.2 percent) misused them. Misuse accounted for 17.2 percent of all benzodiazepine use, was highest among adults aged 18 to 25, and was strongly associated with misuse of prescription opioids or stimulants. Because these medications are often prescribed for the very anxiety that drives the misuse, benzodiazepine addiction treatment generally needs to address the underlying anxiety disorder at the same time.

Schizophrenia, cannabis, and nicotine

A meta-analysis of 123 studies covering 165,811 people in Drug and Alcohol Dependence (Hunt et al., 2018) found that 41.7 percent of people with schizophrenia spectrum disorders had a substance use disorder: 27.5 percent involving illicit drugs, 26.2 percent cannabis, and 24.3 percent alcohol. Nicotine adds another layer: in the 2024 NSDUH, 41.4 percent of adults with serious mental illness used tobacco or vaped nicotine in the past month, double the 20.5 percent rate among adults with no mental illness. We look at this pairing in depth in our article on schizophrenia and substance use.

Bipolar disorder and alcohol

A companion meta-analysis in the Journal of Affective Disorders (Hunt et al., 2016) found that among treatment-seeking patients with bipolar disorder, comorbid alcohol use was the most common substance problem at 42 percent, followed by cannabis at 20 percent and other illicit drugs at 17 percent. Alcohol appears at or near the top of nearly every pairing in this list, which is one reason alcohol addiction treatment programs routinely screen for mood disorders and trauma at intake.

"At intake we assume nothing and screen everyone for both conditions, because the presenting problem is often not the whole story. The drinking that brings someone in may be sitting on top of untreated post-traumatic stress, and a benzodiazepine problem often began as an attempt to quiet an anxiety disorder that was never fully treated. If you only ask about the substance, you miss half the picture."

Dr. Richard Marasa, Medical Director, Clear Steps Recovery

Serious mental illness and addiction: the SMI numbers

Serious mental illness, the narrower category covering conditions that substantially limit major life activities, shows the strongest overlap with substance use of any group in the national data (2024 NSDUH):

• 14.6 million adults (5.6 percent) had SMI in 2024. Nearly half of them, 6.9 million people, also had a substance use disorder.
• 54.0 million adults (20.6 percent) had either SMI or an SUD.
• Adults with SMI were far more likely to use substances than adults with no mental illness: 57.4 percent used illicit drugs in the past year (versus 21.2 percent), 50.5 percent used marijuana (versus 18.8 percent), and 9.5 percent misused opioids (versus 2.0 percent).
• Among adults with co-occurring SMI and an SUD, 70.1 percent received either mental health or substance use treatment in 2024, and 29.9 percent, about 2.1 million people, received neither.

Dual diagnosis in adolescents and young adults

Young adults aged 18 to 25 carry the heaviest co-occurring burden of any age group in the 2024 NSDUH:

• 45.8 percent, 16.0 million young adults, had either AMI or an SUD in the past year.
• 25.9 percent (9.1 million) had a substance use disorder, and 13.3 percent had co-occurring AMI and an SUD.
• Roughly 3 in 10 (30.5 percent) had either serious mental illness or an SUD.

Among adolescents aged 12 to 17, the survey pairs major depressive episode (MDE) with SUD. In 2024, 3.8 million adolescents (15.4 percent) had a past-year MDE and 2.0 million had an SUD. About two fifths of adolescents with an SUD, 792,000 young people, also had a major depressive episode. Of those with both conditions, 27.9 percent received neither type of treatment and only 18.1 percent received both. One genuinely encouraging trend has held through the redesigned survey years: the share of adolescents with co-occurring MDE and SUD declined from 4.1 percent in 2021 to 3.2 percent in 2024.

Veterans and dual diagnosis

SAMHSA's 2022-2024 NSDUH veterans spotlight, published in November 2025, pools three survey years to compare veterans who served in a combat zone with those who did not. Among veterans who served in a combat zone (annual averages, 2022-2024):

• About 1 in 5 (21.3 percent) had any mental illness in the past year, versus 16.9 percent of veterans who did not serve in a combat zone.
• About 1 in 16 (6.4 percent) had serious mental illness, versus 4.6 percent.
• About 1 in 4 (24.8 percent) binge drank in the past month, and about 1 in 8 (12.3 percent) had an alcohol use disorder, both significantly higher than among non-combat-zone veterans.
• About 1 in 20 (4.8 percent) had serious thoughts of suicide in the past year, versus 3.3 percent.
• For overall substance use disorders the gap was smaller and not statistically significant: 17.5 percent of combat-zone veterans had a past-year SUD, versus 15.4 percent of other veterans.

The PTSD link compounds these numbers: as noted above, the VA reports that veterans with lifetime PTSD are 2 times more likely to have alcohol problems and 3 times more likely to have drug problems, and that PTSD usually comes first. Veterans in New Hampshire and Massachusetts can access co-occurring care through community providers, including our VA rehab program that works with VA Community Care referrals.

The fentanyl era: overdose deaths and mental health

The overdose crisis and the co-occurring disorder problem are deeply entangled. In the 2024 NSDUH, adults with any mental illness misused opioids at nearly triple the rate of adults with no mental illness (5.5 percent versus 2.0 percent), and adults with serious mental illness at nearly five times that rate (9.5 percent).

The national mortality curve has finally turned. Provisional data from CDC's National Vital Statistics System predict 69,147 drug overdose deaths for the 12 months ending January 2026, a 13.2 percent decline from the previous year, per the CDC overdose data page (updated June 2026). Those figures are provisional and are updated monthly. A falling national total is genuine progress, but it does not shrink the elevated risk carried by people with untreated mental illness who use opioids or stimulants, which is precisely the population least likely to be receiving complete care.

The integrated treatment gap, reconciled

If you have researched this topic before, you have probably seen three different claims: that about 10 percent, about 12 percent, or exactly 12.7 percent of people with co-occurring disorders receive integrated treatment. These numbers come from different survey years, different definitions, and in one case a different population, and they are almost never labeled. Here is the actual series, each figure tied to its live source:

Where the circulating figures come from:

• "About 1 in 10" is a rounding of the older specialty-facility era, when the received-both share ran between 6.8 and 8.3 percent (2015-2018 NSDUH series, reported in the 2018 annual report).
• "About 12 percent" and "12.7 percent" are real but belong to a subset: in the 2019 NSDUH, 12.7 percent of adults with co-occurring serious mental illness and an SUD received both types of care. It was never the all-co-occurring figure.
• The current number, on the broadest modern definitions: in 2024, 14.5 percent of adults with co-occurring AMI and an SUD received both mental health and substance use treatment, 41.0 percent received only mental health care, 3.2 percent received only substance use treatment, and 41.2 percent, about 8.8 million people, received neither.

Two comparability warnings before quoting a trend from this table. The survey mode changed in 2021, and both sets of treatment questions were substantially revised for the 2022 survey, with further changes in 2024. The older rows counted substance use treatment only when it happened at a specialty facility, while the current questions count treatment in more settings, including medications and telehealth, so the apparent rise from 7 percent to 14.5 percent partly reflects broader questions rather than purely better access. The row-by-row numbers are each accurate for their own year; the series is not a clean trend line.

"The most sobering thing about these numbers is how many people are working hard in one system while the other half of their condition goes unaddressed. When a family compares programs, the question I would ask is simple: who on the team treats the mental health condition, who treats the substance use, and do they sit in the same room to plan care. If a program cannot answer that concretely, keep looking."

Dr. Richard Marasa, Medical Director, Clear Steps Recovery

New Hampshire and Massachusetts: the New England numbers

State-level co-occurring data exist, but almost no one publishes them. The table below compiles the 2023-2024 NSDUH state prevalence estimates (annual averages, adults 18 or older, published 2025) for New Hampshire and Massachusetts against the national rates:

A few patterns stand out. New Hampshire runs well above the national rate for mental illness (27.7 versus 23.1 percent) and serious mental illness (7.6 versus 5.6 percent). Massachusetts runs higher on the substance side, with elevated alcohol and drug use disorder rates. Both states sit above the national average on every co-occurring measure, and among young adults the gap widens: roughly one in six young adults in each state had co-occurring conditions in 2023-2024.

State health agencies publish the local context behind these rates: the New Hampshire Department of Health and Human Services coordinates public mental health and substance use services through its behavioral health system, and the Massachusetts Department of Public Health tracks the state's overdose burden in its current opioid statistics releases. For residents seeking care, outpatient options in the region include partial hospitalization in New Hampshire and dedicated mental health treatment in Massachusetts.

Criminal justice and the wider social impact

Co-occurring conditions are heavily overrepresented in the justice system, and here too the honest numbers come with dates attached. In the Bureau of Justice Statistics' Survey of Prison Inmates, 2016 (published 2021), 47 percent of state and federal prisoners who had been in the community during the year before admission met DSM-IV criteria for a substance use disorder in the 12 months before entering prison, and 38 percent of all prisoners reported using drugs at the time of their offense. Those data were collected in 2016, the most recent national inmate survey published, and any page quoting prison substance use figures without that date is presenting old data as new.

What integrated outpatient treatment looks like

The evidence base behind "treat both, together" is consistent across agencies: NIDA states that treating co-occurring conditions at the same time, rather than separately, tends to make all the treatments more effective and improve outcomes, and SAMHSA's co-occurring disorders guidance is built on the same principle. In practice, integrated outpatient care means one care team that screens for both conditions at intake, builds a single plan, and delivers therapy for the mental health condition alongside addiction treatment, adjusting both as the person stabilizes.

Structured outpatient levels of care make that possible without stepping away from work or family. An intensive outpatient program typically provides multiple therapy sessions per week with co-occurring support built in, while day-level programs offer more clinical hours for people who need them. Evidence-based therapies used across these levels include cognitive behavioral therapy and dialectical behavior therapy, which was developed specifically for people whose emotional dysregulation and substance use reinforce each other. Clear Steps Recovery provides this kind of integrated outpatient care at its Londonderry, New Hampshire and Needham, Massachusetts locations, and our team can talk through whether outpatient care fits a specific situation.

How these numbers were compiled

Every statistic on this page was verified against a live primary source in July 2026. National prevalence and treatment figures come from the 2024 NSDUH annual national report (survey year 2024, released July 2025); state figures come from the 2023-2024 NSDUH state prevalence estimates; veterans figures come from SAMHSA's 2022-2024 pooled data spotlight; overdose figures come from CDC provisional mortality data and are marked provisional; pairing figures come from the cited peer-reviewed meta-analyses and national studies. One caution governs everything here: NSDUH moved to multimode data collection in 2021, and SAMHSA states that estimates from 2021 onward are not comparable with 2020 or earlier; the treatment questions were further revised in 2022 and 2024, so treatment-receipt figures are only comparable within their own question era. Where older figures appear on this page, they are labeled with their survey year. Statistics from other treatment providers' marketing pages were not used as sources. This page was medically reviewed by Dr. Richard Marasa and last reviewed on July 3, 2026.