In This article

• Large cohort and meta-analyses do not support a consistent increase in overall risk of major congenital malformations with prenatal cannabis exposure.
• Meta-analyses relying on unadjusted estimates tend to show associations between cannabis exposure and birth defects but are likely confounded by sociodemographic and co-exposure factors.
• Better-controlled analyses consistently raise concern for increased risk of several rare anomalies—particularly gastroschisis, omphalocele, and possibly Ebstein’s anomaly—though absolute risks remain low.
• Clinicians should frame cannabis as a modifiable risk factor in pregnancy, especially given additional links to low birth weight, preterm birth, and adverse neurodevelopmental outcomes.
• Counseling should be individualized, recognizing that many pregnant patients use cannabis for symptom management and would benefit from early, preconception discussions about safer alternatives.

Studies examining the risk for major congenital malformations (MCMs) in infants with prenatal exposure to cannabis have yielded inconsistent findings.  While some studies have documented an association between cannabis exposure and a broad range of malformations, large, well-conducted studies have not documented an overall increased risk for malformations in analyses accounting for confounding factors, including maternal age, sociodemographic factors, and other potentially teratogenic exposures.

However, there have been a number of recent studies suggesting that exposure to maternal cannabis use during pregnancy may be associated with several more rare malformations. Two of those malformations, gastroschisis and omphalocoele, are related to the gastrointestinal tract, and the third, Ebstein’s anomaly, is a cardiovascular malformation.  While more study is needed to fully explore this association, the findings raise concerns and should be considered when discussing the use of cannabis during pregnancy.

Avalos et al, 2025: Kaiser Permanente Cohort

In this population-based retrospective cohort study, Avalos and colleagues examined risk for MCMs in singleton births (January 2011–July 2020).  In this cohort, individuals were universally screened for substance use at entry into prenatal care.  Prenatal cannabis use was defined as self-reported use or a positive toxicology test during pregnancy.  Electronic health records and birth certificate data were used to identify 38 specific major congenital malformations (MCMs) within 8 organ systems:

• Out of 363,952 infants, 22,494 (6.2%) were exposed to maternal prenatal cannabis use.
• Overall, 6,094 infants (2.17%) had an MCM.
• Prenatal cannabis use was associated with gastroschisis in unadjusted analyses (RR = 2.00, 95% CI: 1.25-3.19) and in models adjusting for other non-cannabis prenatal substance use (aRR 1.68, 95% CI 1.04–2.71), but not in models adjusting for maternal age or other confounding factors.
• Prenatal cannabis use was associated with omphalocele in unadjusted analyses (RR 3.04, 95% CI 1.42–6.48), in maternal age–adjusted models (aRR 3.54, 95% CI 1.68–7.48), in other substance use–adjusted models (aRR 3.31, 95% CI 1.50–7.31), and in propensity score–adjusted models (aRR 2.92, 95% CI 1.26–6.77).
• Cases of gastroschisis and omphalocele were rare: n = 172 (0.05%) and n = 48 (0.01%), respectively.
• No associations were observed between maternal prenatal cannabis use and any other birth defects.

Summary: In this population-based study including 22,494 infants with prenatal cannabis exposure, maternal prenatal cannabis use was associated with an increased risk for gastroschisis and omphalocele. 

Delker et al, 2023: Meta-Analysis

In this meta-analysis, the authors identified 23 studies analyzing data from birth years 1968–2021. There was considerable heterogeneity across studies, and many included relatively small numbers of exposed infants.  This meta-analysis did not include the study above from Avalos et al (2025).

The following anatomic groups were examined: cardiac (9 studies), oral cleft (3 studies), gastrointestinal (4 studies), genitourinary (3 studies), musculoskeletal (7 studies), and nervous system (5 studies). The key findings included: 

• Eleven studies reported an association between cannabis use and the risk of a MCM. 
• The pooled unadjusted odds ratio was 1.33 (95% CI 1.14–1.56), and the pooled adjusted odds ratio (aOR) was 1.22 (95% CI 1.00–1.50).
• Across most outcomes, pooled unadjusted associations were attenuated after adjustment for potential confounders.
• Two specific MCMs had pooled estimates that did not decrease after adjustment:
  • Ebstein’s anomaly (two studies, aOR 2.19, 95% CI 1.25–3.82).
  • Gastroschisis (five studies, aOR 2.50, 95% CI 1.09–5.74).

Summary: In this meta-analysis, there was inconsistent evidence that prenatal cannabis exposure is associated with an overall increase in risk of birth defects. However, there were two specific outcomes for which risk estimates did not attenuate after adjusting for confounding variables: Ebstein’s anomaly (two studies) and gastroschisis (five studies). 

Delker et al, 2024: Prenatal Cannabis Use Disorder and Gastroschisis 

The Study of Outcomes of Mothers and Infants is a population-based cohort including singleton live births in California from 2007–2019 (n = 5,774,656). This study specifically looked at the association between maternal diagnosis of cannabis use disorder and risk of gastroschisis.  

• Cannabis use disorder was measured using diagnosis codes at any visit during pregnancy or at delivery.
• The prevalence of cannabis use disorder was about 1% (n=50,435).
• The prevalence of gastroschisis was 0.14% among those with cannabis use disorder and 0.06% among those without cannabis use disorder.
• There were positive associations between cannabis use disorder and gastroschisis in multivariable models (aRR 1.3, 95% CI 1.0–1.7).
• The association varied by maternal age and was largest among individuals older than 34 years (aRR 2.5, 95% CI 1.0–5.8).

Summary: In this study focusing on individuals with cannabis use disorder, researchers observed a positive association between cannabis use disorder and gastroschisis, with the strongest associations observed in mothers older than 34 years.

How to Interpret the Data

In interpreting data on cannabis and the risk of major congenital malformations, a few key studies stand out and provide clinically relevant information about the association between prenatal cannabis exposure and birth defects.

The retrospective cohort study from Avalos and colleagues is one of the largest to date and assesses exposure to cannabis at the first prenatal visit using both self-report and urine toxicology screens. While the study is not prospective, an important strength is that exposure data were collected prior to knowledge of pregnancy outcomes. The large sample size, including 22,494 exposed infants, allowed researchers to control for multiple potential confounding factors. In adjusted analyses, prenatal cannabis exposure was associated with an increased risk of omphalocele across several models and showed a more fragile association with gastroschisis that did not persist after full adjustment.

Some meta-analyses, for example the one published by Tadese et al in 2024, have reported associations between prenatal cannabis exposure and a wide range of major malformations.  However, these analyses provide unadjusted estimates of risk and do not fully account for potential confounding factors, including maternal age, sociodemographic variables and other teratogenic exposures, including alcohol and tobacco. These unadjusted associations likely overestimate the teratogenic risk of cannabis itself.

In contrast, the meta-analysis conducted by Delker and colleagues (2023) attempted to more thoroughly control for potential confounding factors. In the 23 studies examining associations between prenatal cannabis exposure and MCMs, the authors note that, after adjusting for potential confounders, many of the initially observed associations were no longer statistically significant.

Overall, Delker and colleagues conclude that there is inconsistent evidence to support a broad association between prenatal cannabis exposure and major congenital malformations. That said, in this meta-analysis, two specific anomalies—Ebstein’s anomaly and gastroschisis—remained significantly associated with prenatal cannabis exposure even after adjustment, suggesting that these particular defects warrant further investigation.

In the third study focusing on a large cohort of individuals with a diagnosis of cannabis use disorder, an approach that would enrich for pregnant individuals with higher intensity cannabis exposures, Delker and colleagues observed an association between cannabis exposure and risk for gastroschisis, with the risk being about threefold greater in older mothers. 

When assessing the literature regarding risk for malformations, we look for reproducible patterns. The most concerning and consistent signal across better-controlled studies is the association between prenatal cannabis exposure and gastroschisis, with additional concern for omphalocele and, to a lesser extent, Ebstein’s anomaly. 

Taken together, these studies highlight a pattern of concern for rare abdominal wall and select cardiac defects in the setting of prenatal cannabis exposure, while not supporting a consistent increase in the overall rate of major congenital malformations.

Clinical Guidelines

While the data regarding cannabis use and the risk of malformations are mixed, there are a number of studies suggesting that prenatal cannabis exposure may be associated with increased risk of some rare malformations, specifically gastroschisis and possibly omphalocele and Ebstein’s anomaly. Although the relative risks may be elevated, these malformations are rare, so the absolute increase in risk for an individual patient remains small.

When these potential teratogenic risks are considered alongside data linking prenatal cannabis exposure with other negative outcomes, including low birth weight, preterm birth, small for gestational age, and adverse neurodevelopmental outcomes, it remains prudent to urge pregnant and birthing people to avoid the use of cannabis during pregnancy.

Many individuals regard cannabis as relatively safe and may not actively consider stopping its use during pregnancy. We need to provide patients with concrete, balanced information about risk so that they can make well-informed decisions.

A major challenge in counseling about cannabis use during pregnancy is that reasons for use vary.

• For individuals using cannabis recreationally, it is easier to argue that continued use is not worth the potential risks to the fetus.
• However, most individuals who continue to use cannabis during pregnancy report doing so to manage symptoms such as pain, insomnia, stress, depression, and anxiety.
• Individuals with heavier and more consistent use of cannabis during pregnancy are more likely to have a cannabis use disorder and may require additional support to achieve abstinence.

Thus, discussions about cannabis use during pregnancy need to be nuanced and tailored to each patient. Ideally, these conversations should take place long before conception, allowing time to consider and implement alternative treatments, either non-pharmacologic or pharmacologic, when indicated. Framing the discussion around shared decision-making, with attention to the patient’s symptoms, values, and available evidence, can help support safer choices in pregnancy.

—Ruta Nonacs, MD PhD