Cannabis use in pregnancy linked to preterm birth, low birth weight

Pregnant patients walk into Jamie Lo's clinic at Oregon Health & Science University and ask her a question she hears often: they've quit smoking and drinking, but is cannabis still safe? Until now, many have assumed the answer was yes. Lo, an obstetrician specializing in high-risk pregnancies, knows better. An updated systematic review published this week in JAMA Pediatrics, led by Lo and her team, finds that cannabis use during pregnancy increases the odds of preterm birth, low birth weight, and infant death—findings that challenge a widespread perception of safety around the drug.

The research draws on 51 observational studies involving 21.1 million people. Researchers identified eight new studies since their last update, enough to shift the certainty of evidence from "very-low-to-low" to "moderate" for low birth weight, preterm birth, and babies born small for their gestational age. The review also found increased odds of newborn mortality, though the certainty there remains lower. What makes this update significant is that it includes a larger proportion of studies examining people who use cannabis alone, without also using nicotine or alcohol—isolating cannabis as the variable and strengthening the case that the drug itself carries risk.

Cannabis remains one of the most commonly used substances in pregnancy, despite being illegal under federal law. Unlike alcohol and nicotine, which have seen declining use among pregnant people, cannabis use continues to climb. Many patients resist giving it up because it addresses real symptoms: nausea, insomnia, and pain are genuine burdens of pregnancy, and cannabis offers relief. In states like Oregon, where the drug is legal for medical and recreational use, patients have easy access and little legal barrier to continued use.

The mechanism of harm is becoming clearer. Animal studies using nonhuman primates exposed to THC, cannabis's main psychoactive compound, show consistent damage to placental function. Ultrasound and MRI imaging revealed reduced blood flow to the placenta, decreased oxygen availability, and lower amniotic fluid volume. Lo translates this into clinical terms: when the placenta doesn't function properly, fetal development and growth suffer. The placenta is the lifeline between mother and baby; compromise it, and you compromise the pregnancy.

Yet the reality Lo faces in her clinic is that abstinence, while ideal, isn't realistic for many patients. Some cannot or will not stop using cannabis during pregnancy. Recognizing this, Lo has adopted a harm-reduction approach. She counsels patients to reduce both the amount and frequency of their use, explaining that even partial reduction can lower the risk of adverse outcomes. It's a pragmatic stance born from years of caring for pregnant people in difficult circumstances—acknowledging that perfection is the enemy of progress, and that some reduction in exposure is better than none.

The findings arrive at a moment of tension in American medicine and law. Cannabis remains a Schedule 1 substance federally, yet states continue to legalize it. Pregnant people in those states face a gap between what the law permits and what the evidence suggests is safe. Lo's work, along with that of her co-authors at OHSU and the Portland VA, narrows that gap by providing clearer evidence of harm. The question now is whether that evidence will shift clinical practice and patient behavior—and whether it will inform policy in states where cannabis is legal but pregnancy outcomes remain a public health concern.