Last verified: April 2026

What CUD Is — and Why It Matters

Cannabis Use Disorder (CUD) is a clinical diagnosis defined in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), published by the American Psychiatric Association. It replaced the earlier DSM-IV distinction between "cannabis abuse" and "cannabis dependence" with a single spectrum diagnosis graded by severity: mild (2–3 criteria), moderate (4–5 criteria), and severe (6 or more of 11 criteria).

The 11 diagnostic criteria include impaired control (using more or longer than intended, unsuccessful efforts to cut down, excessive time spent obtaining or using, craving), social impairment (failure to fulfill major role obligations, continued use despite social problems, giving up activities), risky use (use in physically hazardous situations, continued use despite physical or psychological problems), and pharmacological indicators (tolerance, withdrawal).

CUD is not simply "using a lot of cannabis." It requires clinically significant impairment or distress. Many daily users do not meet criteria; some infrequent users do. The disorder reflects a pattern where cannabis use has become compulsive despite negative consequences — the hallmark of all substance use disorders.

Population Prevalence — The Numbers

The lifetime risk of developing CUD among all people who have ever used cannabis is approximately 9–10% — a figure derived from multiple epidemiological studies, most notably Anthony et al. (1994) and the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). For comparison, the lifetime dependence risk for alcohol is approximately 15%, for cocaine 17%, for heroin 23%, and for tobacco 32%. Cannabis is not the most addictive substance, but calling it "non-addictive" is factually wrong.

The risk is not evenly distributed. Adolescent-onset users (those who begin before age 18) face a substantially higher lifetime risk of approximately 17% — nearly double the overall population rate. Among regular users (defined variously as weekly-to-daily), the prevalence of CUD ranges from 20–30%, depending on the study and criteria used.

The 2024 National Survey on Drug Use and Health (NSDUH) estimated that approximately 7% of past-year cannabis users met criteria for CUD. Given that an estimated 61.9 million Americans used cannabis in the past year, this translates to roughly 4.3 million people with a diagnosable cannabis use disorder — a public health burden that dwarfs many conditions receiving far more attention and funding.

Approximately 9% of those who experiment with marijuana will become dependent. The number goes up to about 1 in 6 among those who start using as adolescents.

National Institute on Drug Abuse (NIDA) — based on Anthony et al. 1994 and subsequent replications

Withdrawal — The Timeline

Cannabis withdrawal was formally recognized as a diagnosis in DSM-5 (2013), though its existence had been documented in controlled studies for over a decade prior. The syndrome is not life-threatening — unlike alcohol or benzodiazepine withdrawal, it carries no seizure risk — but it is clinically significant and contributes to relapse in those attempting to quit.

The withdrawal timeline follows a characteristic pattern:

• Onset: 1–2 days after last use — irritability, anxiety, and sleep disturbance typically appear within 24–48 hours as tissue-stored THC levels decline below the threshold for CB1 activation
• Peak: 2–6 days — symptoms reach maximum intensity, with irritability, insomnia, decreased appetite, restlessness, and depressed mood predominating. Some patients report vivid dreams (likely REM rebound after chronic THC-mediated REM suppression), night sweats, and physical discomfort
• Resolution: 1–3 weeks — most symptoms resolve within two weeks, though sleep disturbance and mild irritability may persist for up to three weeks in heavy long-term users

Approximately 47% of regular cannabis users report withdrawal symptoms upon cessation. The severity correlates with frequency and duration of use, potency of products consumed, and individual genetic variation in cannabinoid metabolism. The prolonged timeline relative to most drugs of abuse reflects THC’s extreme lipophilicity — it accumulates in adipose tissue and is released slowly, creating a gradual rather than abrupt decline in brain cannabinoid levels.

Treatment — What Works

There is currently no FDA-approved pharmacotherapy for cannabis use disorder. Multiple candidates have been investigated in clinical trials — including gabapentin, N-acetylcysteine (NAC), buspirone, and dronabinol (synthetic THC as a substitution agent) — but none has demonstrated sufficient efficacy for regulatory approval. The absence of a pharmacological treatment option distinguishes CUD from alcohol (naltrexone, acamprosate, disulfiram), opioid (methadone, buprenorphine, naltrexone), and tobacco (varenicline, bupropion, nicotine replacement) use disorders.

The evidence-based treatments for CUD are psychotherapeutic:

• Cognitive Behavioral Therapy (CBT) — the most studied approach, focusing on identifying triggers, developing coping strategies, and restructuring cognitions around cannabis use. Effect sizes are moderate but consistent across trials
• Motivational Enhancement Therapy (MET) — a brief (typically 2–4 session) intervention that uses motivational interviewing techniques to resolve ambivalence about change. Particularly effective for patients who are not yet committed to abstinence
• Contingency Management (CM) — provides tangible incentives (vouchers, prizes) contingent on verified abstinence (typically urine drug screens). Produces the largest short-term effect sizes but effects often diminish when incentives are removed

In practice, combination approaches (CBT + MET, or CBT + CM) tend to outperform any single modality. The Marijuana Treatment Project, the largest randomized trial of CUD treatment, found that a 9-session CBT + MET combination produced better outcomes than either 2-session MET alone or a delayed treatment control, though abstinence rates at 15-month follow-up remained modest across all groups.

Risk Factors and Vulnerability

Not everyone who uses cannabis regularly develops CUD, and the factors that increase vulnerability are becoming better understood. Age of onset is the single strongest predictor — the adolescent brain’s ongoing prefrontal cortex development makes it more susceptible to the neuroplastic changes that underlie compulsive use patterns. Frequency and potency matter: daily users of high-THC products are at substantially greater risk than occasional users of lower-potency flower.

Genetic factors account for an estimated 50–70% of the variance in vulnerability to CUD, based on twin studies. Specific gene variants affecting cannabinoid metabolism (CYP2C9), receptor signaling (CNR1), and dopaminergic function (DRD2) have been associated with increased risk, though no single gene is determinative. Comorbid psychiatric conditions — particularly anxiety disorders, depression, ADHD, and PTSD — substantially increase CUD risk, likely through self-medication pathways.

The rising potency of commercial cannabis products is a concern. Average THC concentrations in flower have increased from approximately 4% in 1995 to over 15% in 2021, while concentrates routinely exceed 70–90% THC. Whether this potency escalation is increasing CUD prevalence at the population level is an active area of research, though the mechanistic plausibility is strong.

The Honest Assessment

Cannabis is less addictive than alcohol, tobacco, cocaine, or opioids by every population measure. It is also not "non-addictive," and framing it as such does a disservice to the millions of people who meet diagnostic criteria for CUD. The withdrawal syndrome is real, clinically documented, and for some patients, a significant barrier to cessation.

The gap between cannabis culture’s dismissal of addiction risk and the clinical evidence is wide. Responsible cannabis science acknowledges both realities: that most users do not develop CUD, and that a meaningful minority does — with consequences that include impaired relationships, occupational dysfunction, and sustained inability to reduce use despite a genuine desire to do so.