Cannabis and Mental Health: Why ‘Cannabis’ and ‘d9-THC’ Are Not the Same Thing
A current opinion review argues that conflating cannabis with THC distorts the science of psychiatric risk, and that who uses it, how they use it, and at what dose matters enormously for outcomes across psychosis, anxiety, mood disorders, and PTSD.
Why This Matters
Cannabis legalization is accelerating worldwide, and patients with psychiatric conditions are among the most frequent users and the most frequent askers of clinical questions about it. Yet the research literature, public discourse, and policy frameworks routinely treat “cannabis” as a single exposure, obscuring vast differences in cannabinoid content, dose, route, and user vulnerability. Clinicians need a more precise framework for counseling patients, and this review attempts to provide one at a moment when the gap between patient behavior and clinical guidance is widening rapidly.
Clinical Summary
Cannabis products vary enormously in their ratios of delta-9-tetrahydrocannabinol (d9-THC) to cannabidiol (CBD), and these two cannabinoids have fundamentally different pharmacological profiles. d9-THC is the primary psychoactive constituent associated with dose-dependent anxiogenic effects, transient psychotic symptoms, and, at population level, increased risk of psychotic disorders in vulnerable individuals. CBD, by contrast, acts through distinct receptor pathways and has shown anxiolytic and antipsychotic properties in preclinical models and early-phase clinical trials. A 2025 Current Opinion article in CNS Drugs by Kelly Sagar and Staci Gruber of McLean Hospital and Harvard Medical School argues that the failure to disaggregate these compounds, along with critical use-pattern and individual-level variables, has systematically distorted the cannabis and mental health literature.
The review highlights several key findings from the broader literature. For psychosis, observational data consistently show an association between heavy d9-THC use and psychotic outcomes, but the authors argue that cannabis alone is not sufficient to cause psychotic disorders, framing it instead as one contributing factor within a multifactorial model that includes genetic liability and neurodevelopmental timing. For anxiety, d9-THC appears to have biphasic dose-dependent effects, potentially anxiolytic at low doses but clearly anxiogenic at higher doses, while CBD shows a more favorable profile across preclinical and limited clinical data. Youth and individuals with familial vulnerability to psychosis emerge as the highest-risk subgroups. The authors acknowledge that the observational nature of most cited studies, residual confounding, and inconsistent exposure definitions prevent causal conclusions, and they call for standardized cannabinoid-specific measurement in future research.
Dr. Caplan’s Take
This review names something that has frustrated me for years in clinical practice: the word “cannabis” does almost no useful work in a medical conversation. When a patient tells me they use cannabis, I still know almost nothing about their actual pharmacological exposure. Are they using a high-THC concentrate daily, or a CBD-dominant tincture twice a week? The difference is not trivial. It is the difference between a compound with clear psychotomimetic potential and one with emerging evidence for anxiolytic effects. This review makes that distinction well, even if it cannot quantify the boundaries precisely. The honest answer to the patient asking whether cannabis is “safe” or “helpful” is that the question itself is too vague to answer responsibly.
In practice, I take a detailed cannabinoid history: product type, THC-to-CBD ratio, dose, frequency, route of administration, and timing relative to symptoms. For patients with any personal or family history of psychosis, I counsel clearly against high-THC products and monitor closely. For patients asking about CBD for anxiety, I am transparent that the clinical trial data remain limited and that over-the-counter products are poorly regulated. I do not recommend cannabis as a first-line treatment for any psychiatric condition, but I also do not dismiss the pharmacological specificity that this review rightly emphasizes.
Clinical Perspective
This review sits early in the research arc for precision cannabinoid medicine. It provides a useful conceptual framework, distinguishing cannabinoid type, use pattern, and individual vulnerability, but it does not and cannot provide the quantitative evidence clinicians need to set specific thresholds for safe versus harmful use. The association between heavy d9-THC use and psychotic outcomes in youth is the most robust finding in the broader literature, and this review affirms it. The therapeutic promise of CBD for anxiety and psychosis, while pharmacologically plausible and supported by some controlled data, remains at a stage where it informs clinical watchfulness rather than clinical recommendations. Clinicians should not yet present CBD as an evidence-based treatment for these conditions in routine practice.
From a safety standpoint, d9-THC has well-documented acute cognitive and psychomotor effects, potential for dependence with chronic heavy use, and interactions with sedatives, anticholinergics, and CYP-metabolized medications. CBD, while generally better tolerated, can inhibit CYP3A4 and CYP2C19, raising the possibility of clinically significant drug interactions with common psychiatric medications including clobazam, sertraline, and certain antipsychotics. The single most actionable step clinicians can take now is to routinely ask about cannabinoid-specific exposure, including product type and THC-to-CBD ratio, rather than recording a binary “cannabis use: yes/no” in clinical documentation.
Study at a Glance
- Study Type
- Narrative review / Current Opinion
- Population
- General population cannabis users; subgroups including youth, individuals with psychosis liability, anxiety, mood disorders, and PTSD (no original data collected)
- Intervention
- Not applicable (review of d9-THC and CBD exposure across cited studies)
- Comparator
- Not applicable
- Primary Outcomes
- Psychiatric outcomes across psychosis, anxiety, mood disorders, and PTSD as reported in cited literature
- Sample Size
- No original sample; largest cited study N=25,747
- Journal
- CNS Drugs
- Year
- 2025
- DOI or PMID
- Not provided in extracted text
- Funding Source
- Not specified in extracted text
What Kind of Evidence Is This
This is a Current Opinion article combining narrative literature review with expert commentary, published in CNS Drugs. It occupies a position below systematic reviews and meta-analyses in the evidence hierarchy. Because it lacks a pre-specified search strategy, formal inclusion and exclusion criteria, or quantitative synthesis, its most important inference constraint is that the cited literature reflects author selection and may not be fully representative of the available evidence base. Conclusions should be treated as expert-informed hypotheses, not established findings.
How This Fits With the Broader Literature
The review’s central thesis aligns with the direction of several high-profile analyses. The 2019 Lancet Psychiatry study by Di Forti and colleagues demonstrated a dose-response relationship between high-potency THC products and first-episode psychosis across multiple European sites, supporting the disaggregation argument this review advances. Similarly, McGuire and colleagues’ 2018 randomized controlled trial of CBD as adjunctive treatment for schizophrenia provided early clinical evidence for CBD’s antipsychotic potential. This review extends these findings by organizing the moderating variables into a single framework, though it does not add new empirical weight.
Where this review departs somewhat from the broader literature is in its framing, which leans toward rehabilitating “cannabis” as a category by emphasizing that harms are concentrated in specific cannabinoid, pattern, and population combinations. This framing is scientifically defensible but should be read alongside more cautious syntheses, such as the 2017 National Academies report, which found substantial evidence of association between cannabis use and psychotic disorders without the same degree of disaggregation.
Common Misreadings
The most likely overinterpretation is to read this review as evidence that cannabis is safe when d9-THC is avoided or when CBD is used instead. The review does not establish that. CBD’s therapeutic evidence remains preliminary, with small trials and unresolved questions about optimal dosing, and the absence of harm from a given cannabinoid profile has not been demonstrated in any long-term, controlled study. A second common misreading would be to conclude that because cannabis “does not cause” psychosis on its own, clinicians can deprioritize cannabis-related psychosis screening. The review’s nuanced framing of cannabis as a contributing factor within multifactorial risk is not the same as declaring it clinically irrelevant to psychotic outcomes.
Bottom Line
This review provides a conceptually valuable framework for thinking about cannabis and mental health with greater pharmacological precision. It does not establish causal relationships, quantify safe thresholds, or validate CBD as a psychiatric treatment. Its primary contribution is argumentative: it makes a well-reasoned case that treating “cannabis” as a monolithic exposure obscures the science. For clinicians, the practical takeaway is to document cannabinoid-specific exposure in every patient and to maintain distinct risk assessments for high-THC and CBD-predominant use, especially in youth and psychosis-vulnerable populations.
References
- Sagar KA, Gruber SA. Cannabis and mental health: current opinion. CNS Drugs. 2025. (Full DOI not available in extracted text.)
- Di Forti M, Quattrone D, Freeman TP, et al. The contribution of cannabis use to variation in the incidence of psychotic disorder across Europe (EU-GEI): a multicentre case-control study. Lancet Psychiatry. 2019;6(5):427-436. doi:10.1016/S2215-0366(19)30048-3
- McGuire P, Robson P, Cubala WJ, et al. Cannabidiol (CBD) as an adjunctive therapy in schizophrenia: a multicenter randomized controlled trial. Am J Psychiatry. 2018;175(3):225-231. doi:10.1176/appi.ajp.2017.17030325
- National Academies of Sciences, Engineering, and Medicine. The Health Effects of Cannabis and Cannabinoids: The Current State of Evidence and Recommendations for Research. Washington, DC: The National Academies Press; 2017. doi:10.17226/24625
Have thoughts on this? Share it:
