Cannabis use disorder is the second most common substance use disorder and rising, even as high-potency products and federal rescheduling normalize use faster than treatment can respond.
Key Takeaways
- The disorder is common and growing: More than 20 million Americans had cannabis use disorder in 2024, making it the second most common substance use disorder after alcohol. Close to a third of regular users develop it.
- Today’s cannabis is not the drug older research studied: THC potency has risen severalfold, and concentrates can exceed 90 percent, sharpening the risks of dependence and psychosis. The evidence base and the public perception both lag the product.
- Rescheduling normalizes use, not treatment: The partial federal move to Schedule III eases the industry’s path while doing nothing to build treatment capacity. Normalization tends to raise use, and use is what drives the disorder.
- The treatment side is thin: There is no FDA-approved medication for cannabis use disorder, behavioral therapy is underused, and few present for care. The mismatch between commercial momentum and clinical response is widening.
Few drugs have been reinvented in the public mind as thoroughly as cannabis. Within living memory, it was the gateway drug of anti-drug filmstrips and mandatory-minimum sentences, the substance a generation was taught would lead inexorably to heroin and ruin.
That framing collapsed with startling speed. Cannabis is now legal for recreational use in roughly half the states and permitted medically in most of the rest, folded into wellness routines and late-night comedy and the ordinary texture of adult life, having completed the same migration from vice to lifestyle that alcohol made generations ago. The cultural verdict has effectively been rendered, and it is that cannabis is basically fine.
Into that settled consensus, the federal government in April 2026 moved marijuana a notch down the schedule of controlled substances, from the most restrictive tier toward the middle, and the coverage treated it, understandably, as a business story: what it means for the cannabis industry, for tax treatment and banking access, for the investors and operators of a maturing legal market.
There is another story underneath that one, and it is the story the behavioral health field has an obligation to tell, because it is the one the industry coverage skips. Cannabis use disorder is real, it is common, it is growing, and the product driving it bears little resemblance to what the culture made its peace with. As the legal and commercial machinery around cannabis becomes more sophisticated and the social permission more complete, the treatment side remains an afterthought, underdeveloped and underfunded, and the gap between how easy cannabis is to sell and how hard the disorder is to treat is widening rather than closing.
Cannabis Use Disorder Is More Common Than the Culture Admits
Its scale is larger than the casual consensus assumes. According to federal survey data, more than 20.6 million Americans aged twelve and older had cannabis use disorder in 2024, which makes it the second most common substance use disorder in the country, behind only alcohol. The condition is defined, like other substance use disorders, by a pattern of use that continues despite mounting harm, and it is not a marginal risk of cannabis use but a fairly common one: close to a third of people who use cannabis regularly develop the disorder, with the risk climbing sharply for those who begin in adolescence and those who use high-potency products.
This perception gap is the problem. Because cannabis is widely understood as benign, both the public and much of the medical system treat cannabis use disorder as a contradiction in terms, a worry invented by people who never accepted legalization. That framing does not survive contact with the data. The disorder produces genuine impairment and drives a meaningful share of addiction-related hospitalizations, its withdrawal syndrome is real enough to carry its own diagnostic criteria, and, as the next section takes up, the high-potency products now dominating the market carry a documented association with psychosis that the old filmstrips never imagined. In the gap between the cultural assessment of cannabis as harmless and the clinical reality of twenty million people with a diagnosable disorder, a great deal of untreated suffering sits unaddressed.
Why High-Potency Products Changed the Risk
Part of what makes the current moment distinct is that the product itself has changed almost beyond recognition. The cannabis that shaped the culture’s easygoing attitude, and much of the older research that attitude leaned on, contained THC concentrations around 4 percent. Today’s flower routinely runs several times that, and the concentrates, the vapes, dabs, and extracts that make up a growing share of the market, can exceed 90 percent THC. This is not a difference of degree so much as a different drug, delivering a quantity and speed of THC that the plant in its older form could not approach, and the human evidence base has barely begun to catch up to what that does.
Clearest among the concerns is psychosis. High-potency cannabis has been associated with a substantially elevated risk of psychotic disorders, particularly among adolescents and young adults and those with a family history, and clinicians describe encountering new cases of psychosis in the wake of heavy high-potency use and existing patients destabilizing after it. The SAMHSA advisory on cannabis use disorder issued this year catalogs the range of harms, from the acute anxiety and psychotic episodes that high concentrations can trigger to the longer-term risks, and it lands in a market moving the opposite direction, toward ever more potent products sold with ever less friction. The treatment system is being asked to respond to a drug that is changing faster than the research documenting it, and to a market whose commercial logic points relentlessly toward higher potency, because concentrated products are cheaper to ship, easier to consume discreetly, and more profitable per unit of plant.
How Rescheduling Widens the Gap
Into this picture arrives the rescheduling, and its logic runs the wrong way for the treatment side. The action that took effect on April 28, 2026 moved two narrow categories, FDA-approved marijuana drug products and marijuana handled under a state medical license, from Schedule I to Schedule III, leaving recreational cannabis and everything outside those categories still in Schedule I. A separate and broader question, whether marijuana as a whole should be reclassified, went to an expedited administrative hearing that concluded on July 15, 2026, with a decision not expected until late in the year at the earliest. Even the partial change marked a historic federal softening. Rescheduling is not legalization, a distinction most of the coverage blurred, but its practical and symbolic effect is to normalize, to signal that the federal government now regards cannabis as meaningfully less dangerous than it once claimed, and normalization has predictable consequences for use.
What undermines the treatment side is the asymmetry of what the change touches. Rescheduling eases the path for the cannabis industry, on taxes, on research, on the general legitimacy of the enterprise, and does nothing whatever to build the capacity to treat the disorder that rising use produces. There is no provision in a scheduling decision for treatment infrastructure, no funding for the behavioral health programs that would absorb the additional cannabis use disorder that normalization tends to generate, no mechanism connecting the commercial momentum to the clinical response. The commercial and cultural machinery around cannabis grows more capable every year. The treatment side does not, and each step that makes cannabis easier to sell without making the disorder easier to treat widens the distance between them.
What Cannabis Use Disorder Means for Treatment Providers
For the treatment field, cannabis use disorder is an oddly neglected opportunity, a population in the tens of millions that the system barely engages. The engagement numbers are strikingly low relative to the prevalence, both because few people with the disorder see it as a problem warranting treatment and because the treatment side has never organized around it the way it has around opioids or alcohol.
The adolescent picture sharpens the point: young people who begin using early carry roughly double the lifetime risk of developing the disorder, cannabis vaping among high schoolers has climbed, and treatment for adolescent cannabis use disorder is frequently delayed or never sought, which means the population most susceptible to the long-term harms is also the least likely to reach care. A field that waits for these patients to present voluntarily will wait a long time, because the defining feature of the disorder, in cannabis as in every other substance, is that it does not feel from the inside like a problem worth treating until it has already cost a great deal. That neglect is itself the opening: a provider that takes cannabis use disorder seriously, screens for it, and builds a credible clinical offering is addressing a need that twenty million people have and that almost no one is competing to meet.
Clinical tools exist, if imperfectly. There is no FDA-approved medication for cannabis use disorder, which is a real limitation and an active area of research, but the behavioral treatments that work for other substance use disorders have shown real effectiveness for cannabis too, from motivational enhancement to contingency management, and they are approaches the field already knows how to deliver, including through the contingency-management models now expanding across addiction treatment.
Harder is the framing. A provider entering this space has to treat as a serious clinical condition something the surrounding culture, and increasingly the federal government, is busy declaring benign, and has to reach a population that mostly does not think it needs help. Neither task is easy, and neither is optional if the treatment side is to keep pace with a market that grows more capable every year. The prevalence is documented, the tools are known, and the demand will continue to build regardless of whether the field organizes to meet it.
