Marijuana is not formally classified as a teratogen, meaning it has not been shown to cause the kind of structural birth defects typically associated with known teratogens like alcohol or thalidomide. However, that distinction is narrower than it sounds. Prenatal cannabis exposure is linked to a range of adverse outcomes for pregnancy and child development, and every major medical organization advises against using it during pregnancy.
What “Teratogen” Actually Means
A teratogen is a substance that causes recognizable physical malformations when a developing embryo or fetus is exposed to it. Alcohol, for example, is a well-established teratogen because it produces a specific pattern of facial abnormalities, organ defects, and growth problems known as fetal alcohol syndrome. Marijuana does not produce this kind of signature pattern of birth defects. A review in Drug and Alcohol Dependence Reports states plainly that cannabis “has no documented teratogenic effects” in the classical sense.
But the absence of that label does not mean cannabis is safe during pregnancy. Many substances cause harm to a developing baby without meeting the strict definition of a teratogen. The real question most people are asking is whether marijuana can hurt a pregnancy or a baby’s development, and the answer to that is more complicated.
How THC Reaches the Fetus
THC, the primary psychoactive compound in marijuana, readily crosses the placenta. Once it does, it binds to the same receptors in the placenta and developing fetal organs that the body’s own endocannabinoid system uses. This system plays a role in brain development, immune function, and cell growth. Research in rhesus macaques published in Scientific Reports has shown that chronic THC exposure alters placental function and development, raising concerns that even if THC doesn’t cause visible malformations, it may disrupt the biological signaling that guides fetal growth.
Links to Specific Birth Defects
While marijuana doesn’t cause the broad spectrum of defects associated with classic teratogens, it isn’t entirely off the hook for structural problems. A large study published in Birth Defects Research found that prenatal cannabis use was associated with a roughly threefold increase in the risk of omphalocele, a defect where abdominal organs protrude through the belly button. There was also an elevated risk of gastroschisis, a similar abdominal wall defect, though that association weakened after adjusting for the mother’s age.
Both conditions are rare. In the study population, gastroschisis occurred in about 0.05% of births and omphalocele in about 0.01%. No associations were found between prenatal cannabis use and any other structural birth defects. So the risk of physical malformation exists but is small and limited to specific conditions.
The Bigger Concern: Brain and Behavior
The more consistent evidence points not to visible birth defects but to effects on the developing brain. A cohort study of 250 children published in JAMA Pediatrics found that children exposed to cannabis before birth scored about 6 points lower (roughly 0.4 standard deviations) on tests of attention and impulse control at age 5, compared to unexposed children. They also showed poorer planning ability and more aggressive behavior during observed interactions. These differences held up after researchers adjusted for confounding factors.
Interestingly, the same study found no differences in caregiver-reported behavior or in lab measures of cognitive flexibility and emotion regulation. This means the effects were real but selective, showing up on certain structured tests and behavioral observations rather than across every measure of development.
The Adolescent Brain Cognitive Development (ABCD) Study, which followed nearly 11,500 children aged 9 to 11 across 22 U.S. sites, adds a longer-term picture. Children whose mothers used cannabis after learning they were pregnant showed more attention problems, externalizing behavior, social difficulties, and psychotic-like experiences. They also had lower birth weight, reduced brain volume (both white and gray matter), and lower cognitive scores. These associations persisted even after accounting for confounding variables.
One important nuance from the ABCD data: children whose mothers used cannabis only before knowing they were pregnant, and then stopped, showed no meaningful differences from unexposed children once confounders were considered. This suggests that timing and duration of exposure matter significantly.
Why the Research Is Hard to Untangle
One of the biggest challenges in studying marijuana’s effects during pregnancy is that cannabis use rarely happens in isolation. Tobacco and alcohol are both established teratogens, and they frequently co-occur with cannabis use during pregnancy. A study in JAMA Network Open noted that this overlap makes it genuinely difficult to separate the health risks of cannabis from those of other substances.
Researchers try to control for this by adjusting for tobacco use, alcohol consumption, age, race, income, and other socioeconomic factors. The strongest studies use propensity score weighting or multiple adjustment models to isolate cannabis-specific effects. Still, no observational study can fully eliminate confounding, and ethical constraints mean no one will ever run a randomized controlled trial giving pregnant women THC. The evidence we have is strong enough to raise serious concern but not strong enough to draw the kind of clean causal lines that exist for alcohol.
Today’s Cannabis Is Not the Same Product
Much of the older research on cannabis and pregnancy was conducted when THC concentrations in typical marijuana flower were far lower than they are today. Smoked flower now typically reaches 10% to 30% THC. Concentrates used with vaporizers often exceed 80% THC, and some products reach 90%. This matters because THC is the compound that crosses the placenta and interacts with fetal development. The dose a fetus is exposed to today from a single session of vaping concentrates could be many times higher than what earlier studies measured, and the long-term consequences of these higher exposures are largely unknown.
What Medical Organizations Recommend
The American College of Obstetricians and Gynecologists (ACOG) released updated guidance in 2025 stating that there is no medical indication for cannabis use during pregnancy or the postpartum period. ACOG’s position acknowledges that cannabis use during pregnancy has been associated with spontaneous preterm birth, low birth weight, NICU admissions, and postnatal neurocognitive and behavioral problems. The organization also notes a “misperception in the general public that cannabis is safe for use during pregnancy due to its increasing availability, legalization, and social acceptability.”
The CDC echoes this, stating that cannabis may be harmful to a baby regardless of how it’s consumed: smoking, vaping, dabbing, edibles, or topical creams and lotions. Smoking cannabis carries additional risk because the smoke contains many of the same toxic and carcinogenic compounds found in tobacco smoke.
ACOG’s guidance also highlights a troubling inequity: Black and minority women face disproportionate rates of drug testing and mandatory reporting to child protective services. The organization recommends universal screening through interviews or self-reporting rather than biological testing like urine or hair samples, which can’t accurately assess severity of use and have wide detection windows that may not reflect current behavior.
The Bottom Line on Classification
Marijuana is not a teratogen in the way that word is traditionally used. It does not cause the recognizable pattern of physical birth defects that defines substances like alcohol, certain medications, or infections like rubella. But it does cross the placenta, interact with fetal brain development, and correlate with measurable cognitive and behavioral differences in children. It is also associated with pregnancy complications including preterm birth and low birth weight. The fact that it falls outside the strict teratogen category should not be confused with a finding of safety.