Polysubstance Use Defines Substance Disorders Now
Vulnerability and compulsive reward-seeking change forms across substances.
Posted May 11, 2026 | Reviewed by Monica Vilhauer Ph.D.
Addiction medicine is in the midst of a conceptual tectonic shift. Recent work by Director Nora Volkow and her National Institute on Drug Abuse (NIDA) colleagues is crystallizing the direction of the change. What clinicians have historically labeled as separate conditions—alcohol use disorder, opioid use disorder, and cocaine use disorder—are now better understood as different expressions of one common vulnerability.
These NIDA experts reported nationally-representative data from 92,000-plus U.S. adults, revealing that polysubstance use is the dominant pattern among individuals with substance use disorders (SUDs). Nearly three-fourths (73%) of stimulant -involved overdose deaths also included opioids.
Large population-level data make these facts difficult to ignore. Polysubstance involvement is not a niche phenomenon confined to obscure cases. Instead, among those with opioid, stimulant, methamphetamine, or hallucinogen use disorders, most have at least one additional substance use disorder, if not several.
If we include nicotine and cannabis—as we should—the picture shifts dramatically. Tobacco dependence is strongly connected to alcohol and other drug use, and cannabis use disorder substantially increases the likelihood of other substance involvement. When these substances are accounted for, most individuals with substance use disorders are using more than one substance. If most patients no longer have one-substance disorder, the longstanding clinical instinct to identify a “primary” substance becomes untenable.
A second finding is the timing of substance exposure. Early initiation, particularly before age 18, is strongly associated with a greater number and severity of substance use disorders later in life. In contrast, individuals delaying initiation until age 21 or later have substantially fewer disorders and dramatically lower rates of moderate-to-severe addiction. Adolescence is a period of maturation in executive control, reward processing, stress regulation, and decision making . Early exposure to psychoactive substances apparently activates a generalized addiction liability that later emerges across multiple substances.
Genetic studies show substantial overlaps in addiction across alcohol, nicotine, cannabis, opioid, and stimulant use disorders. These are overlapping expressions of a common vulnerability determined by reward sensitivity, impulsivity, stress responsivity, and executive control.
While different substances act on distinct receptors, they ultimately interact with a common reward system: mesolimbic dopamine signaling, incentive salience, stress and anti-reward circuits, and prefrontal regulatory networks. These shared pathways explain a clinical reality: patients don’t use just one drug or remain in a single substance category. They transition, combine, and substitute. Polysubstance use is the expected expression of a system-level disorder.
These new data support a complete reframing of diagnosis and expectations. The designation of a “primary” substance persists because it’s operationally useful. Rather than a DSM with multiple independent substance use disorders, it’s more accurate to conceptualize one SUD with polysubstance expressions.
Current clinical practice is stuck, for now, with substance-specific diagnostic systems, regulatory structures, and even insurance reimbursements organized around particular drugs. Pharmacotherapies are approved for specific disorders.
Consequently, when clinicians identify a “primary” substance, they are usually applying pragmatics rather than identifying a unique disease. The primary drug is often the one with the greatest immediate risk—such as opioids in the context of overdose—and also the one for which an effective medication can be deployed. Or it’s the one producing the most dangerous withdrawal syndrome. These are practical decisions, but conceptually unsustainable and obsolete.
What is particularly striking is not just that polysubstance use is common, but that it has become the default pattern. Four decades ago, people may have used one drug or at least had one substance disorder at a time. Two decades ago, substance use disorders were often conceptualized as relatively discrete. Alcohol dependence, cocaine dependence, and opioid dependence were treated as parallel but separate trajectories. Co-use existed but was frequently framed as comorbidity rather than the core phenotype.
Over the past 10-15 years, that distinction has eroded. Prescription opioids revealed increasing overlaps with benzodiazepines and alcohol. The subsequent transition to heroin and then fentanyl further blurred boundaries . By the mid-2010s, combinations like opioids and stimulants became central to overdose risk. In the fentanyl era, the actual drug supply became polysubstance, with synthetic opioids frequently contaminating stimulants and other drugs, producing involuntary co-exposure.
Earlier eras only appeared more “single-substance” because drug access was narrower, the illicit supply was less contaminated, and clinicians lacked today’s surveillance data. In practice, however, residential treatment programs treated alcoholism , opioid, or cocaine addictions as sufficiently similar to warrant the same mutual-support models, recovery frameworks, and residential interventions—implicitly acknowledging a shared underlying disorder long before the neurobiology was fully understood.
The increasing prevalence of polysubstance use is not just a change in behavior, but a manifestation of a unified addiction vulnerability that has always been present but previously under-recognized.
Over time, drug use patterns evolved toward maximizing euphoria or mitigating negative effects through functional drug combinations. Stimulants are used to counter opioid sedation. Benzodiazepines are used to modulate stimulant-induced anxiety . Substance use has become strategic, determined by pharmacologic effects, availability, and cost.
What are the implications?
First, prevention. If initial exposure activates a generalized addiction vulnerability, then prevention strategies must exceed any single drug. Focusing narrowly on cocaine or fentanyl misses the developmental pathway. Delaying or preventing early use of alcohol, nicotine, and cannabis may have the largest impact on lifetime addiction risk, precisely because these substances are the most common entries into the system.
Second, diagnosis. The field may need to move toward reflecting shared liability rather than discrete categories. Instead of asking which substance is primary, it may be more clinically accurate to describe addiction severity, pattern, and current drug risks. A formulation such as “severe SUD with polysubstance involvement, currently highest risk from opioids” captures the unity of the disorder and need for prioritization.
Third, treatment. A unified model supports person-centered care rather than siloed systems organized around individual substances. It also points toward treatment targeting shared mechanisms—craving, stress dysregulation, impaired executive control—rather than focusing on blocking or substituting individual drugs. Emerging pharmacologic approaches, including those modulating reward and metabolic signaling, suggest it may be possible to treat addiction at the level of core drivers rather than individual expressions of which drugs are used now.
Most importantly, this approach reframes clinical tasks, not as treating a patient’s opioid use or alcohol use in isolation. Instead, the whole person is evaluated and treated—the underlying disorder makes multiple substances compelling, interchangeable, relapsing, and persistent.
Many people with substance use disorders experience concurrent and sequential polysubstance involvement. A patient may simultaneously use—or meet criteria for disorders involving—opioids, stimulants, cannabis, and alcohol, while also shifting primary substances over months or years, depending on availability, their withdrawal state, psychiatric symptoms, costs, or changes in the illicit drug supply. These progressions are manifestations of a shared addiction liability.
The convergence of epidemiologic trends, developmental data, genetic findings, and neurobiological models is forcing psychiatry and addiction medicine to reframe polysubstance disorders as manifestations of a shared neurobiological liability rather than independent diseases.
Today, the challenge is moving beyond substance-specific silos toward recognizing addiction as a unified disorder rarely confining itself to one drug or one psychiatric presentation. Specific substances may change across situations and time, but the underlying addictive vulnerability remains the same.
To find a therapist, visit the Psychology Today Therapy Directory .
Facebook image: Aleksey Korchemkin/Shutterstock
Han B, Volkow ND, Jones CM, Dowell D, Baldwin G, Einstein EB, Subramaniam GA, Olsen Y, Blanco C, Compton WM. Polysubstance use disorders among US adults. Mol Psychiatry. 2026 Apr 29. doi: 10.1038/s41380-026-03618-z. Epub ahead of print. PMID: 42056224.
Tanz LJ, Miller KD, Dinwiddie AT, et al. Drug Overdose Deaths Involving Stimulants ― United States, January 2018–June 2024. MMWR Morb Mortal Wkly Rep 2025;74:491–499. DOI: http://dx.doi.org/10.15585/mmwr.mm7432a1 .
Bailey AJ, McHugh RK. Why do we focus on the exception and not the rule? Examining the prevalence of mono- versus polysubstance use in the general population. Addiction. 2023 Oct;118(10):2026-2029. doi: 10.1111/add.16290. Epub 2023 Jul 12. PMID: 37438998.
Compton WM, Valentino RJ, DuPont RL. Polysubstance use in the U.S. opioid crisis. Mol Psychiatry. 2021 Jan;26(1):41-50. doi: 10.1038/s41380-020-00949-3. Epub 2020 Nov 13. PMID: 33188253; PMCID: PMC7815508.
Gorelick DA. Cannabis-Related Disorders and Toxic Effects. N Engl J Med. 2023 Dec 14;389(24):2267-2275. doi: 10.1056/NEJMra2212152. PMID: 38091532.
Gold MS. Dual disorders: nosology, diagnosis, & treatment confusion--chicken or egg? Introduction. J Addict Dis. 2007;26 Suppl 1:1-3. doi: 10.1300/J069v26S01_01. PMID: 19283969.
Kolla BP, Oesterle T, Gold M, Southwick F, Rummans T. Infectious diseases occurring in the context of substance use disorders: A concise review. J Neurol Sci. 2020 Apr 15;411:116719. doi: 10.1016/j.jns.2020.116719. Epub 2020 Jan 31. PMID: 32070807.
Gold, M. S., & Frost-Pineda, K. (2005). Substance Abuse and Psychiatric Dual Disorders: Focus on Tobacco. Journal of Dual Diagnosis , 1 (1), 15–36. https://doi.org/10.1300/J374v01n01_03
Gold, M. S. (2005). Dual Diagnosis: Discovery of a Critical Role for Environmental Exposure. Journal of Dual Diagnosis , 1 (1), 5–13. https://doi.org/10.1300/J374v01n01_02
Share this post Facebook Bluesky Linkedin Email
There was a problem adding your email address. Please try again.
By submitting your information you agree to the Psychology Today Terms & Conditions and Privacy Policy
Mark S. Gold, M.D., is a pioneering researcher, professor, and chairman of psychiatry at Yale, the University of Florida, and Washington University in St Louis.
This article is part of the Bringwise Psychology Journal — daily insights on human behavior, mental health, and personal growth.