Cannabis Health Risks: Peer-Reviewed Research Consensus

84 studies analyzedLast updated March 7, 2026

Overview

The research base for cannabis health risks includes 84 peer-reviewed studies spanning 2025–2026. Of these, 7 provide strong evidence, including 4 meta-analyses and 0 randomized controlled trials. Key findings with strong support include: meta-analysis of 13 studies found prenatal cannabis exposure significantly increased odds of small-for-gestational-age and low birth weight outcomes, and study of 289,000 surgical patients found cannabis use disorder increased post-surgical cardiovascular risk by 26%, with recreational use risky only for those with high baseline cardiac risk. However, several findings remain debated, and the evidence is not uniform across all areas. Many studies have methodological limitations including small sample sizes, short follow-up periods, and reliance on self-reported data.

What the Research Shows

Findings supported by multiple peer-reviewed studies. Stronger evidence means more consistency across study types.

Meta-analysis of 13 studies found prenatal cannabis exposure significantly increased odds of small-for-gestational-age and low birth weight outcomes

Strong Evidence
1 study|Based on 13 studies with varying definitions of cannabis exposure. Could not account for dose, frequency, route, or timing of use. Many included studies relied on self-reported cannabis use, which lik

Study of 289,000 surgical patients found cannabis use disorder increased post-surgical cardiovascular risk by 26%, with recreational use risky only for those with high baseline cardiac risk

Moderate Evidence
7 studies|Based on limited number of strong-evidence studies.

Where Scientists Disagree

Areas where research shows conflicting results or ongoing scientific debate.

NHANES analysis of 3,690 adults found marijuana/hashish use associated with 4

Moderate Evidence
15 studies|Cross-sectional NHANES design cannot establish causation. Self-reported drug use may be inaccurate. Cannabis smoke exposure patterns (frequency, duration, method) were not fully characterized. Residua

Connecticut crash-toxicology data linkage found alcohol, cannabis, and other drugs significantly predicted driver injury, with lack of safety equipment as the strongest predictor

Moderate Evidence
15 studies|The study covers a single state (Connecticut) and relies on toxicology data that may not reflect impairment at the time of the crash. The counterintuitive finding about driver errors suggests the mode

Large cohort study of 363,952 births found prenatal cannabis use associated with a rare abdominal wall defect (omphalocele) but not with 36 other major birth defects

Moderate Evidence
15 studies|Observational design cannot prove causation. Cannabis exposure was measured at prenatal care entry, not throughout pregnancy. Could not account for potency, frequency, or route of use. Very small numb

Removing cannabis as a urine drug screening indication in a maternity unit eliminated racial disparities in testing and CPS reporting without missing other substance use

Moderate Evidence
15 studies|Single center in one Midwestern city. Cannot fully separate the effect of removing cannabis indication from the clinical decision support component. Unmeasured confounders may exist.

What We Still Don't Know

  • Only 0 randomized controlled trials exist out of 84 studies — most evidence is observational or from reviews.
  • Long-term prospective studies tracking outcomes over 5+ years are largely absent from the literature.
  • Research on diverse populations (different ages, ethnicities, and medical backgrounds) remains limited.

Evidence Breakdown

Distribution of study types in this research area. Higher-tier evidence (meta-analyses, RCTs) provides stronger conclusions.

Meta-Analyses & Systematic Reviews(Tier 1)
4 (5%)
Observational & Cohort(Tier 3-4)
33 (39%)
Reviews & Scoping(Tier 4)
9 (11%)
Case Reports & Animal(Tier 5)
9 (11%)
Other
29 (35%)

Key Studies

The most impactful research in this area.

Meta-Analysis Links Prenatal Cannabis to Smaller Babies

This meta-analysis pools evidence from multiple studies to provide a clearer picture of prenatal cannabis risks. The significant findings for birth size outcomes add confidence to what individual studies have suggested.

2025

Meta-analysis found cannabis use associated with 53% higher all-cause mortality, but evidence quality was low

This is the first meta-analysis to synthesize evidence on cannabis use and all-cause mortality. The finding of a 53% increased risk is notable but comes with substantial caveats: high heterogeneity, potential confounding, and the association disappearing in patient populations suggest the relationsh

2025

Cannabis Users Had 48% Higher Risk of Cardiovascular Events in Large Meta-Analysis

Whether cannabis causes cardiovascular disease has been debated for years. This meta-analysis of nearly 1.9 million people provides the strongest pooled estimate to date, showing a significant association with composite cardiovascular events even if individual endpoints fall short of significance.

2025

Cannabis Use Linked to 29% Higher Heart Attack Risk and Double the Cardiovascular Death Risk

This is one of the most comprehensive quantitative assessments of cannabis cardiovascular risk. The elevated cardiovascular death risk (2.1x) is particularly concerning given increasing cannabis use and the often-overlooked cardiac effects.

2025

CBD and THC Can Interfere With How Your Body Processes Other Medications

With millions of people using CBD products alongside prescription medications, understanding drug interactions is critical for safety. The strong inhibition of CYP3A4 alone is clinically significant — this enzyme metabolizes drugs like blood thinners, statins, and anti-seizure medications.

2026

Cannabis use disorder increased post-surgical cardiovascular risk by 26% in a study of 289,000 patients

This is one of the largest studies linking cannabis use to post-surgical cardiovascular complications. The critical finding that risk depends on both the type of cannabis use (disorder vs recreational) and baseline cardiac risk provides actionable information for surgical risk assessment.

2025

Research Timeline

How our understanding of this topic has evolved.

2020–present

84 studies published. Includes 4 meta-analyses, 7 strong-evidence studies.

About This Consensus

This consensus synthesizes 84 peer-reviewed studies: 4 meta-analyses (5%), 33 observational studies (39%), 9 reviews (11%), 9 case studies (11%), 29 other study types (35%). Studies span from the earliest available research through 2025. Evidence strength ratings reflect study design, sample size, and replication across multiple research groups.

This page synthesizes findings from 84 peer-reviewed studies. It is not medical advice. Always consult a healthcare provider for personal health decisions.