The Whys of Cannabis Self-Medication

Access to medical and recreational cannabis has expanded rapidly, far outpacing rigorous scientific evidence supporting its efficacy for most clinical indications. As legalization has broadened availability and also reduced perceived risk, clinicians increasingly encounter patients viewing cannabis for symptom relief rather than as a substance of misuse. This shift raises a key question: Why has cannabis become so widely used as medicine despite limited evidence for sustained benefit and mounting evidence of harm?

Top Reasons for Cannabis Use

Source: Dr Mark Gold

Individuals often use cannabis to cope with anxiety, stress, sleep disturbance, chronic pain, depressed mood, or fear of failure. These patterns are clinically concerning because symptom-driven cannabis use is rarely sustainable and is associated with tolerance, escalation, and increased risk of cannabis use disorder (CUD).

Stress and Cannabis

Recent research provides persuasive evidence that stress biology plays a central role in motivating cannabis use. A longitudinal study by Park and colleagues at the Department of Integrative Physiology and Neuroscience at Washington State University examined biological and behavioral traits in rats before cannabis exposure, following them through several weeks of voluntary cannabis vapor self-administration. This work addressed a longstanding limitation in the field: Most prior animal studies relied on forced injections of synthetic cannabinoids or isolated constituents, failing to capture the complexity inherent in human cannabis use.

Researchers found that elevated basal corticosterone levels—reflecting heightened hypothalamic–pituitary–adrenal (HPA) axis activity—emerged as a strong predictor of subsequent cannabis seeking. Cortisol shifts whole-brain mechanisms, amplifying emotions and coding survival-relevant emotional memories. Researchers have found lower baseline morning levels of anandamide, a lipid mediator acting as an endogenous ligand of CB1 receptors, which are the targets for Δ9-tetrahydrocannabinol, the psychoactive ingredient in Cannabis sativa. These findings strongly support a vulnerability model in which stress-related neurobiology precedes and predicts cannabis use rather than emerging as a consequence of it. Cannabis may be seen as a self-medication for stress, but stress responses are both adaptive and help us to remember emotional experiences in a crisis.

Repeated Cannabis Use and Dependence

The National Institute on Drug Abuse (NIDA) estimates that approximately 9 percent of individuals using cannabis develop dependence, with substantially higher rates among daily users, those using high-potency products, adolescents, and individuals who self-medicate.

Clinical data consistently show that individuals using cannabis for self-medication engage in heavier use, escalate more rapidly, and are at greater risk of developing CUD than those using cannabis primarily for social purposes. Self-medication implies learning that cannabis alleviates distressing internal states. Over time, this learning promotes habitual use and dependence.

Although cannabis may provide short-term relief, repeated exposure increases tolerance and withdrawal-related symptom rebound, thereby perpetuating symptoms motivating use.

Genetic Liability and Psychiatric Comorbidity

Genetic factors contribute substantially to liking cannabis and frequent use, and account for 30–80 percent of the variance in CUD risk. Twin studies consistently estimate heritability to be substantial, and genome-wide studies led by Gelernter and Yale University colleagues demonstrated shared genetic liability with cannabis use and depression, risk-taking, and externalizing behaviors.

A review by David Gorelick in the New England Journal of Medicine highlights the extent of psychiatric and substance-related comorbidity in CUD. Approximately two-thirds of individuals with CUD have at least one additional substance use disorder, most commonly to alcohol or tobacco. Co-occurring tobacco smoking is found in the majority of people wth CUD and makes quitting quite difficult. Nearly half also have a non-substance psychiatric disorder, most often major depressive disorder, posttraumatic stress disorder, or generalized anxiety disorder. These comorbidities are associated with greater CUD severity and poorer treatment outcomes.

Among individuals with schizophrenia, about one-quarter meet criteria for CUD, and cannabis use is associated with earlier onset, greater symptom severity, increased disability, and higher rates of suicide attempts in bipolar disorder. Cannabis use is also consistently associated with an approximately threefold increased risk of psychosis. Frequent use and high-potency products contribute substantially to psychotic disorder incidence.

Cannabis Essential Reads

What Happens in the Brain When Cannabis Is Used Every Day

Early Cannabis Use Linked to Memory Slowdown in Teens

Reward Deficiency, Cognitive Flexibility, and Habit Formation

Reduced baseline reward sensitivity—manifesting as anhedonia or low motivation—may increase susceptibility to cannabis self-medication. Cannabis may transiently normalize reward processing through indirect dopaminergic and endocannabinoid modulation, reinforcing continued use. Genetic variation, early life stress, and psychiatric illness all contribute to this phenotype.

Population-Level Context and Risk Factors

Early onset of cannabis use—particularly before age 15—is a robust predictor of CUD and other substance use disorders during young adulthood. This early initiation often occurs concurrently with externalizing problems but also predicts internalizing issues, suggesting it may reflect a general risk factor for mental disorders.

Across medical cannabis registries and consumer surveys, the most commonly-cited reasons for use include chronic pain, anxiety or depressed moods, sleep disturbance, and stress, with recreational enjoyment reported less consistently among frequent users. Meta-analytic data show four of the five most common use indications reflect attempts to relieve internal distress, aligning closely with a self-medication framework.

Risks for CUD increase sharply with frequency and duration of use, rising from approximately 3.5 percent among annual users to more than 30 percent of daily users. Early initiation, high-potency products, adverse childhood experiences, psychiatric comorbidity, polysubstance use, and genetic liability all increase risks.

Conclusion

Across animal models, human genetics, and population-level data, a coherent model of cannabis self-medication emerges. Individuals with heightened stress responsivity, reduced reward sensitivity, and impaired cognitive flexibility are most likely to experience cannabis as highly reinforcing and symptom-relieving—and most likely to develop CUD. Diagnosis and treatment of the underlying or co-existing problems are preferable to self-medication. Genetic liability shapes these vulnerabilities, while repeated cannabis exposure induces neuroadaptations that perpetuate symptoms, reliance, and CUD.