Cannabis both calms and panics — the biphasic dose-response explains why the same drug produces opposite anxiety effects

Acute anxiety reactions and panic attacks were commonly reported during cannabis intoxication. Across observational studies, two patterns showed up repeatedly: frequent cannabis users had higher rates of anxiety disorders, and people with anxiety disorders reported higher rates of cannabis use. The review could not show that cannabis use leads to persistent anxiety disorders. Explanations ranged from neurobiology to environment and social context, including self-medication, shared vulnerability, and confounding by stress and other substance use. The bottom line in 2009 was correlation in both directions without a clear causal pathway.

A systematic search of Medline, PsycLIT, and EMBASE gathered human studies on cannabis and anxiety. Designs were mostly observational or cross-sectional, with some experimental work focused on acute effects. The review synthesized patterns qualitatively. It did not report a pooled effect size, did not specify the number of included studies, and did not establish temporal ordering between cannabis exposure and the onset of anxiety disorders.

Anxiety is one of the most commonly cited reasons people discuss cannabis, and also one of the most frequently reported adverse reactions during intoxication. This review mapped what was known in 2009, separating acute panic-like reactions from questions about longer-term anxiety disorders, and highlighted how easily co-occurrence can be misread as causation.

The anxiety–cannabis relationship is often presented as simple. It is not. Acute anxiety during intoxication can be real, yet that does not automatically translate into a lasting anxiety disorder. Observational links in both directions are consistent with several scenarios: people with anxiety using cannabis to cope, shared risk factors driving both, or subgroups responding differently depending on dose, potency, age of first use, and context. This review set a baseline for those debates in the pre-legalization era when products, potencies, and measurement standards differed from today.

The review period predates widespread legal markets and high-potency concentrates. Most included studies were observational or cross-sectional, which cannot establish temporal order or causality and are vulnerable to confounding. Specific product types, THC-to-CBD ratios, dose, frequency, and timing of last use were rarely characterized. Anxiety was measured with varying tools, and acute intoxication effects were sometimes conflated with chronic disorders. Publication bias and unmeasured comorbidities, including alcohol and tobacco, were likely.