Cannabis and Mental Health: It’s Not Just About the Plant, It’s About What’s In It, How You Use It, and Who You Are

A Harvard-affiliated expert review argues that lumping all cannabis use together obscures critical distinctions between THC harms, CBD promise, and individual risk factors, but the narrative format and selective citation strategy mean its conclusions should be weighed carefully against the broader evidence base.

Why This Matters

Cannabis legalization continues to accelerate across jurisdictions, and clinicians are fielding more questions about cannabis and mental health than ever before. Yet the research literature routinely treats “cannabis” as a monolithic exposure, collapsing enormous variation in cannabinoid content, route of administration, frequency of use, and patient vulnerability into a single variable. This matters because patients making decisions about cannabis, whether for symptom management or recreation, deserve guidance informed by these distinctions. A framework that separates THC-specific risk from CBD-specific promise, and that accounts for who is most vulnerable, is urgently needed at the clinical interface.

Clinical Summary

Cannabis use and psychiatric outcomes have been studied extensively, but the resulting literature is complicated by a persistent terminological conflation: the word “cannabis” almost always refers to delta-9-tetrahydrocannabinol (THC), the primary psychoactive constituent, while cannabidiol (CBD) and other cannabinoids with distinct pharmacological profiles are often ignored. In this 2025 Current Opinion article published in CNS Drugs, Sagar and Gruber of McLean Hospital and Harvard Medical School’s MIND Program argue that this conflation distorts both research conclusions and clinical guidance. Drawing on preclinical models, observational studies, clinical trials, and prior reviews, they build a case that THC and CBD exert fundamentally different effects on the endocannabinoid system and, by extension, on psychiatric symptoms, including psychosis, anxiety, mood disorders, and PTSD.

The review’s key contentions include that high-THC, early-onset, and frequent cannabis use is associated with worsened psychosis outcomes, though as a contributing factor rather than a sole cause in a multifactorial model. THC shows dose-dependent, biphasic effects on anxiety, potentially anxiolytic at low doses but anxiogenic at higher ones, while CBD appears not to worsen anxiety and may reduce it at moderate doses. Youth and individuals with genetic or familial predisposition to psychiatric disorders are identified as the highest-risk groups. However, this is a narrative synthesis without systematic search methodology, inclusion criteria, or formal quality appraisal of cited studies. The text available for analysis was also truncated, meaning full coverage of mood disorders, PTSD, and individual vulnerability factors could not be assessed. The authors acknowledge that the evidence base remains heterogeneous and that more rigorous, cannabinoid-specific research is needed before firm clinical recommendations can be made.

Dr. Caplan’s Take

This review names something that has frustrated me for years in clinical practice: the reflexive treatment of “cannabis” as a single pharmacological entity. When a patient tells me they use cannabis, I need to know what they are actually consuming, how much THC and CBD it contains, how often they use it, and what they are using it for. The THC-CBD distinction is biologically real and clinically consequential. But I also recognize that this article is an expert opinion piece, not a systematic review, and the authors have longstanding research interests in medical cannabis that likely shape how they frame the evidence. The CBD therapeutic story, in particular, is still early.

In my practice, I take a detailed cannabis history for every patient, including product type, cannabinoid ratios, frequency, and age of onset. For patients with psychotic spectrum concerns or a strong family history, I counsel clearly about THC-specific risks. For those asking about CBD for anxiety, I explain that the signal is encouraging but not yet definitive, and I discuss potential drug interactions and the lack of standardized dosing. Above all, I resist the temptation to give a single answer to what is never a single question.

Clinical Perspective

This review sits in a growing body of literature calling for disaggregation of cannabis research by cannabinoid content, use pattern, and patient characteristics. Its central argument, that THC and CBD have distinct and sometimes opposing psychiatric effects, aligns with preclinical pharmacology and a subset of clinical trial data. However, clinicians should note that the CBD anxiolytic evidence cited here is drawn from a relatively small and heterogeneous pool of studies, and the claim of low-dose THC anxiolysis has more support in animal models than in human clinical data. The review’s conclusions about psychosis risk from THC-dominant cannabis are better supported by epidemiological evidence, though the causal architecture remains debated. What the review does not provide is a quantified synthesis of effect sizes or a transparent assessment of study quality across the conditions it covers.

From a practical standpoint, CBD is generally well tolerated but has known interactions with cytochrome P450 enzymes, particularly CYP3A4 and CYP2C19, which can affect levels of common psychiatric medications including citalopram, sertraline, and certain benzodiazepines. Clinicians should screen for these interactions when patients report CBD use. The most immediately actionable takeaway from this review is procedural: every cannabis-related clinical conversation should begin with specifics about what the patient is actually using, not just whether they use “cannabis.”

Study at a Glance

Study Type

Narrative expert opinion review (Current Opinion format)

Population

General cannabis users; subgroups include youth, individuals with psychotic vulnerability, anxiety disorders, mood disorders, and PTSD (preclinical and clinical populations cited)

Intervention

Not applicable (review of cannabis/cannabinoid exposures across studies)

Comparator

Not applicable (no single comparator; contrasts THC vs. CBD, recreational vs. medical use)

Primary Outcomes

Psychiatric symptom trajectories across psychosis, anxiety, mood, and PTSD domains

Sample Size

Not applicable (narrative synthesis of existing literature)

Journal

CNS Drugs

Year

2025 (Vol. 39, pp. 113-125)

DOI or PMID

Not provided in source material

Funding Source

Authors affiliated with McLean Hospital / Harvard Medical School MIND Program

What Kind of Evidence Is This

This is a Current Opinion article, a narrative expert review format that synthesizes existing research through the lens of author expertise rather than through systematic, reproducible methodology. It occupies a position below systematic reviews and meta-analyses in the evidence hierarchy. The most important inference constraint is that without transparent search strategies, inclusion criteria, or formal quality appraisal of cited studies, readers cannot independently verify whether the selected evidence is comprehensive or representative, nor can they rule out selection bias favoring the authors’ conceptual framework.

How This Fits With the Broader Literature

The review’s core argument, that cannabinoid-specific research is necessary, resonates with prior work including the 2017 National Academies of Sciences report on cannabis and health, which similarly noted the difficulty of drawing conclusions from studies that fail to characterize the exposure beyond “cannabis use.” The THC-psychosis association has been strengthened by studies such as the EU-GEI multi-site case-control study (Di Forti et al., Lancet Psychiatry, 2019), which linked daily use of high-potency cannabis to increased odds of psychotic disorder. The CBD evidence for anxiety draws in part from Zuardi and colleagues’ work and the more recent Appiah-Kusi et al. (2020) trial, though effect sizes have been inconsistent across studies. This review extends the conversation by framing these disparate findings through a unified moderator-variable lens, but does not itself resolve the inconsistencies in the underlying evidence.

Common Misreadings

The most likely overinterpretation is reading this review as evidence that CBD is an established treatment for anxiety or psychosis. The review identifies a promising signal, but the clinical trial base for CBD in psychiatric conditions remains small, dosing is not standardized, and the evidence is not yet sufficient to support therapeutic recommendations. A second common misreading would be interpreting the distinction between THC harms and CBD promise as a binary: that THC is categorically harmful and CBD categorically safe. The authors themselves note that THC may have anxiolytic effects at low doses, and CBD’s long-term safety profile, particularly in combination with psychiatric medications, is not fully characterized. This is a framework for asking better questions, not a clinical decision tool.

Bottom Line

This expert review makes a conceptually important argument: cannabis research and clinical guidance must move beyond treating cannabis as a single entity and instead account for cannabinoid content, use patterns, and individual vulnerability. It does not provide original data or systematic evidence synthesis, and its conclusions, particularly regarding CBD’s therapeutic potential, outpace what the current trial base firmly supports. For clinicians, the immediate value lies in upgrading the specificity of cannabis-related conversations with patients, not in changing prescribing behavior.

References

1. Sagar KA, Gruber SA. Cannabis and mental health. CNS Drugs. 2025;39:113-125.
2. 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: National Academies Press; 2017.
3. 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
4. Appiah-Kusi E, Petros N, Wilson R, et al. Effects of short-term cannabidiol treatment on response to social stress in subjects at clinical high risk of developing psychosis. Psychopharmacology (Berl). 2020;237(4):1121-1130. doi:10.1007/s00213-019-05442-6
5. Zuardi AW, Crippa JA, Hallak JE, et al. A critical review of the antipsychotic effects of cannabidiol: 30 years of a translational investigation. Curr Pharm Des. 2012;18(32):5131-5140. doi:10.2174/138161212802884681