Medical Position Statement
The CDC, American College of Obstetricians and Gynecologists (ACOG), American Academy of Pediatrics (AAP), and the Society for Maternal-Fetal Medicine all advise against any cannabis use during pregnancy or breastfeeding. There is no established safe level of THC exposure for a developing fetus or infant. This article presents the evidence base for that position and additional harm reduction context for those seeking to understand the full picture.
- THC is lipophilic (fat-soluble) and crosses the placental barrier readily; it has been confirmed in fetal bloodstream samples in multiple studies using mass spectrometry.
- Prenatal cannabis exposure is associated with increased risk of low birth weight, preterm birth, NICU admission, and stillbirth in epidemiological studies, though confounders are difficult to fully eliminate.
- The endocannabinoid system plays an essential role in fetal brain development beginning in early gestation; exogenous THC can disrupt CB1 receptor-mediated developmental signaling.
- A landmark Bertrand et al. (2018) study found THC present in human breast milk for up to 6 weeks after the mother’s last use; the infant’s daily estimated THC intake ranged from 0.4 to 8.7μg/kg.
- CBD safety during pregnancy is unknown; the FDA advises against it due to animal data showing developmental effects at high doses and absence of adequate human safety data.
- Cannabis is the most commonly used illicit substance during pregnancy in the US, with prevalence approximately doubling in states following legalization, highlighting the importance of clear clinical guidance.
- Drug testing in prenatal care settings has documented racial disparities in testing frequency and reporting, raising significant health equity concerns independent of the pharmacological evidence.
The Medical Consensus: Why All Major Bodies Advise Against Cannabis in Pregnancy
Cannabis is the most widely used illicit substance during pregnancy in the United States. Data from the National Survey on Drug Use and Health and from state-level birth cohort studies consistently show prevalence of prenatal cannabis use at 7–12% in general population surveys, with higher rates in states that have legalized adult use. Some pregnant people use cannabis to manage nausea (hyperemesis gravidarum), anxiety, or chronic pain — conditions for which conventional medical options may be limited or carry their own risks.
Despite these motivations, every major obstetric and pediatric medical organization has issued clear guidance recommending against cannabis use at any stage of pregnancy. The American College of Obstetricians and Gynecologists (ACOG) Committee Opinion 722 states: “Women who are pregnant or contemplating pregnancy should be encouraged to discontinue marijuana use. Obstetrician-gynecologists should be prepared to offer or refer pregnant and lactating women who use marijuana to substance use disorder treatment programs.” The basis for this recommendation is the combination of plausible biological mechanisms (THC’s ability to cross the placental barrier and interact with fetal cannabinoid receptors) and observational evidence linking prenatal exposure to adverse neonatal and developmental outcomes.
This does not mean that all prenatal cannabis exposure produces catastrophic outcomes — many people who used cannabis in early pregnancy before realizing they were pregnant give birth to healthy infants. The medical concern is probabilistic: prenatal THC exposure increases the risk of specific adverse outcomes in a dose-dependent manner, and no safe threshold has been established. For the full context of how cannabis affects the body pharmacologically, see our how cannabis works explainer and our guide on cannabis drug interactions.
Prevalence and Context
Data from the CDC’s Pregnancy Risk Assessment Monitoring System (PRAMS) documented a statistically significant increase in self-reported prenatal cannabis use following legalization in multiple states. Importantly, rates reported through urine toxicology (which detects use people may not self-report) are typically higher than self-reported rates, suggesting underreporting is common. In California, a large Kaiser Permanente study published in JAMA found a 69% increase in prenatal cannabis use across the period spanning legalization. Use was disproportionately concentrated in the first trimester, most often for nausea relief, which is precisely the period of highest neurological vulnerability for the developing fetus.
THC Crosses the Placental Barrier: The Biological Mechanism
The placenta is not an impenetrable barrier between the pregnant person’s circulation and the fetal circulation. It is a selective organ that actively transports certain substances and passively allows others to diffuse across. The key determinant of passive placental transfer is a substance’s lipid solubility: highly lipophilic compounds cross the placenta readily, while large molecules and charged compounds cross poorly.
THC (delta-9-tetrahydrocannabinol) is extremely lipophilic. Its octanol-water partition coefficient (log P) is approximately 6.9, placing it among the most lipid-soluble drugs known. This means that when THC is present in the maternal bloodstream, it diffuses across the placental lipid membranes into fetal circulation in proportion to the maternal blood concentration. Studies using umbilical cord blood sampling at delivery have confirmed THC and its metabolites in neonates born to mothers who used cannabis during pregnancy. A study published in the New England Journal of Medicine confirmed measurable THC in 63% of cord blood samples from cannabis-using mothers.
The Developing Fetal Endocannabinoid System
Perhaps more important than the mere presence of THC in fetal circulation is the fact that the endocannabinoid system is not a passive bystander in fetal development — it is an active participant. CB1 and CB2 receptors are expressed in the human fetal brain beginning in the first trimester of pregnancy. The endogenous endocannabinoid system plays documented roles in neuronal migration (the process by which nerve cells travel to their correct positions in the developing brain), synaptogenesis (formation of synaptic connections), and axon guidance (the directional growth of nerve fibers).
These developmental processes are mediated in part by endocannabinoid signaling at CB1 receptors. When exogenous THC is present, it activates these same CB1 receptors with higher potency and longer duration than endogenous cannabinoids, potentially disrupting the timing and precision of developmental signaling. Fetal brains also lack the protective blood-brain barrier maturation that adults have, making them more permeable to lipophilic substances in the bloodstream.
Animal studies (rodent models primarily) have demonstrated that prenatal THC exposure produces measurable neuroanatomical changes, including alterations in dopaminergic and serotonergic pathway development, reduced dendritic complexity in prefrontal cortical neurons, and altered hippocampal development. Whether these animal findings translate quantitatively to human development at typical use levels remains an active research question, but the biological plausibility of harm is well-established.
Neonatal and Developmental Outcomes: What the Research Shows
The epidemiological literature on prenatal cannabis exposure and neonatal outcomes has grown substantially as legal cannabis use has increased. A 2020 systematic review and meta-analysis published in JAMA Internal Medicine (Correa et al.) synthesized data from dozens of studies covering over 59,000 pregnancies and found statistically significant associations between prenatal cannabis use and several adverse outcomes.
| Outcome | Association | Strength of Evidence | Key Study |
|---|---|---|---|
| Low birth weight (<2,500g) | Significantly increased risk (OR ~1.5) | Moderate-High | JAMA IM 2020 meta-analysis |
| Preterm birth (<37 weeks) | Modestly increased risk | Moderate | Multiple cohort studies |
| NICU admission | Increased in several studies | Moderate | Ontario birth cohort |
| Stillbirth | Association found in some studies; confounders complicate | Low-Moderate | Colorado PRAMS data |
| Neurodevelopmental differences | Attention, executive function, anxiety differences in childhood | Moderate | Ottawa Prenatal Prospective Study (OPPS); ABCD Study |
| Neonatal abstinence symptoms | Tremors, disturbed sleep in newborns | Low-Moderate | Case reports; small cohort studies |
The Confounding Problem: Interpreting the Evidence Carefully
It is important to note that the epidemiological literature on prenatal cannabis exposure is complicated by significant confounding factors. Cannabis use during pregnancy is correlated with tobacco use, alcohol use, poverty, food insecurity, stress, and other factors that independently affect birth outcomes. Most studies attempt to control for these variables statistically, but residual confounding can never be entirely eliminated in observational research. Some researchers argue that after rigorous confounding adjustment, the association between prenatal cannabis use and low birth weight is substantially attenuated, though not eliminated.
The most rigorous longitudinal study of prenatal cannabis exposure outcomes is the Adolescent Brain Cognitive Development (ABCD) Study, which has enrolled over 11,000 US children and is tracking their neurodevelopmental outcomes across childhood. Early published findings from ABCD have found associations between prenatal cannabis exposure and differences in brain structure, increased anxiety and depression symptoms, and lower cognitive performance in children. These findings are correlational and the authors emphasize the role of potential confounders, but the biological plausibility of the observed effects is supported by the developmental mechanisms described above.
THC in Breast Milk: The Bertrand Study and Its Implications
The question of cannabis use during breastfeeding is sometimes treated as distinct from pregnancy, with some sources suggesting that breastfeeding-related exposure is minimal. The best available evidence challenges this assumption. The landmark study by Bertrand et al. (2018), published in the journal Pediatrics, directly measured THC in breast milk samples from 50 cannabis-using lactating women.
Key findings from the Bertrand study:
- THC was detected in 100% of breast milk samples from women who reported using cannabis in the week before collection
- THC was present in breast milk for up to 6 weeks after the mother’s last cannabis use, consistent with THC’s high lipid solubility causing sequestration in milk fat
- Mean breast milk THC concentration was 9.47 ng/mL, with significant variation between individuals
- Estimated infant daily THC intake ranged from 0.4 to 8.7 micrograms per kilogram of body weight, depending on milk consumption and maternal THC levels
- THC-COOH (the primary THC metabolite and urine test marker) was also detected, indicating infant metabolic processing of THC
Critically, the breast milk THC concentrations in this study were measured even when mothers had abstained for several days before feeding (having been told their milk would be collected). The persistence of THC in breast milk for weeks after last use means that “pumping and dumping” for a short period is not an effective strategy for eliminating infant THC exposure — a common misconception. Both the AAP and ACOG guidance specifically note this persistence and advise against cannabis use during breastfeeding.
The developmental significance of the THC doses infants receive through breast milk is not yet fully characterized by research. The precautionary principle argues strongly against any avoidable THC exposure to a developing infant brain, regardless of dose.
CBD Safety in Pregnancy: Unknown Risk
Some pregnant people consider CBD products as a “safer” alternative to THC-containing cannabis. CBD does not produce intoxication, is widely marketed as a wellness product, and in many legal markets can be purchased without a medical recommendation. However, CBD safety during pregnancy is explicitly unknown, and the FDA has issued clear guidance advising against its use.
Animal studies have shown that CBD at doses substantially higher than those typically used by humans can affect fetal development, including disrupting male reproductive system development in rodent models. Whether these animal findings are relevant to human pregnancies at typical CBD doses is unknown — and that uncertainty is precisely the problem. Adequate human clinical safety data for CBD use during pregnancy simply do not exist. The FDA has cited this absence of data, combined with the animal safety signals, as the basis for advising pregnant and breastfeeding people to avoid CBD products entirely.
Additionally, CBD inhibits CYP450 liver enzymes that metabolize many medications commonly used in obstetric settings. For people using CBD alongside prenatal vitamins, folic acid supplements, or any prescription medications, there is a potential for drug interactions that are poorly characterized in the pregnancy context. Our full cannabis drug interactions guide covers the CYP450 mechanism in detail.
Social Equity: Drug Testing in Prenatal Care and Racial Disparities
Any honest discussion of cannabis and pregnancy must address the significant racial disparities in how prenatal cannabis testing and reporting are conducted. Research published in JAMA Pediatrics and other journals has consistently documented that Black women are tested for drugs during prenatal care and delivery at substantially higher rates than white women, even controlling for self-reported substance use. When positive tests are reported to child protective services, Black women and their infants face disproportionately harsher consequences.
This disparity exists independently of the pharmacological evidence about THC. It reflects systemic biases in how clinical suspicion is distributed, which populations are subjected to routine versus selective toxicology screening, and how positive results are acted upon. Some researchers and clinicians have called for universal consent-based testing protocols and standardized reporting criteria to reduce racial bias in prenatal drug testing practices.
From a harm reduction perspective, these disparities affect how pregnant people who use cannabis perceive and interact with the healthcare system. Fear of legal consequences or child protective service involvement can deter people from disclosing cannabis use to their providers — which impairs the ability of clinicians to provide appropriate counseling and support. Confidential, non-punitive prenatal care environments produce better outcomes for both birthing people and their infants.
Harm Reduction for Those Who Continue Use
A harm reduction framework acknowledges that some people will continue to use cannabis during pregnancy despite medical recommendations against it. In this context, providing accurate, non-judgmental information about risk minimization is more effective for health outcomes than abstinence-only messaging that people experiencing pregnancy symptoms (nausea, anxiety, pain) may find difficult to follow.
Evidence-based harm reduction principles for pregnant people who use cannabis:
- Reduce dose: Lower THC exposure means less fetal exposure. If continuing to use, the lowest effective dose is preferable to heavy or frequent use.
- Avoid smoking: Cannabis smoke contains combustion byproducts (carbon monoxide, benzene, particulates) that independently reduce fetal oxygen delivery. Vaporizers, tinctures, or edibles avoid combustion-related fetal harm.
- Avoid high-potency concentrates: Concentrates delivering 70–90% THC produce proportionally higher fetal THC exposure. Flower (15–20% THC) is lower-exposure than concentrates.
- Do not use synthetic cannabinoids: K2, Spice, and similar synthetic cannabinoids have documented severe fetal toxicity and must be avoided entirely.
- Be honest with your prenatal care provider: Accurate disclosure enables appropriate monitoring (e.g., additional growth scans if low birth weight risk exists) and access to support resources.
- Discuss alternatives for underlying conditions: Many conditions people use cannabis to manage in pregnancy have safer pharmacological and non-pharmacological alternatives. Nausea/hyperemesis gravidarum, for example, has FDA-approved treatments including vitamin B6, doxylamine, and ondansetron that may be used under obstetric supervision.
WATCH: Cannabis and Pregnancy — Medical Evidence Overview