Marijuana and Pregnancy: What Does the Science Say?

The best available evidence suggests fears about fetal risk, while not totally unwarranted, are often overblown.


It's 2021 and more U.S. women are using—or at least admitting to using—marijuana while pregnant. In 2016, around 7 percent of pregnant women in a national survey said they used marijuana in the past month, up from 3.4 percent in 2002 and 3.9 percent in aggregated data from 2007-12. Should public health authorities be worried?

Probably not. Shaky science makes measuring the effect of a mom's marijuana use on fetal development difficult, but the best available evidence suggests fears about fetal risk, while not unwarranted, are often overblown.

Studies on marijuana use during pregnancy are inconsistent and inconclusive. But cannabis is not known to be teratogenic—that is, to cause birth defects—in humans. The bulk of scientific evidence suggests that risks posed to developing fetuses are relatively minor and babies exposed to marijuana in utero still fall within normal ranges of outcomes.

A 2020 review looked at longitudinal studies on "the impact of prenatal cannabis exposure on multiple domains of cognitive functioning in individuals aged 0 to 22 years" and found that "evidence does not suggest that prenatal cannabis exposure alone is associated with clinically significant cognitive functioning impairments." Researchers did note some differences—"those exposed performed differently on a minority of cognitive outcomes (worse on < 3.5 percent and better in < 1 percent)" — although "cognitive performance scores of cannabis-exposed groups overwhelmingly fell within the normal range."

A 2016 review of studies on potential ties between in utero marijuana exposure and adverse birth outcomes—things like low birth weight and preterm delivery to miscarriage and stillbirth—found "maternal marijuana use during pregnancy is not an independent risk factor for adverse neonatal outcomes after adjusting for confounding factors." Instead, any increases in adverse outcomes appeared "attributable to concomitant tobacco use and other confounding factors."

Sussing out the effects of marijuana from the effects of other risky behavior presents a problem. "Illegal behaviors tend to cluster so women who use marijuana are more likely to use other drugs," Emily Oster, author of Expecting Better and Brown University economist, explained in her Substack Parent Data. "Reporting on use is hugely biased and we are likely to identify only a (very non-random) subset of women who use."

As marijuana becomes increasingly accepted, studies on its effects are likely to improve. "The best new evidence on this comes from a 2019 study out of Canada," Oster writes. Matching women who used cannabis with demographically similar women who didn't, researchers did "find evidence of worse birth outcomes among the cannabis users," including "an increased risk of prematurity and NICU transfer. The increases are moderate but statistically significant: preterm birth occurred in 10% of cannabis users and 7% of non-users." An August 2020 study from the same authors found marijuana use correlated with slightly higher incidences of intellectual disability and learning disorders, as well as higher chances of having autism spectrum disorder. "The percent increase is large—about 50%—and significant," Oster points out, though the researchers do note that the overall incidence rate is still small.

Though the researchers tried to demographically match participants between groups, it can still be hard to totally compensate for the ways marijuana users may differ from non-users. For instance, marijuana users may be more likely to have partners who use, too, and dads who smoke pot prior to conception may have sperm changes that correlate with higher autism risk.

Ultimately, correlational studies like these raise important possibilities but have major limitations. Some are done not during pregnancy but by later asking women to recall behaviors during their pregnancy, which ups the likelihood of unreliable information. They also often fail to differentiate between consumption methods, timing, and dosage, so we can't know if any and all marijuana use triggers a specific outcome or if it's only produced by certain methods of consumption, THC content, frequency, or at certain stages of pregnancy.

The biggest problem is that it's hard to isolate specific factors like marijuana consumption. The population of women who not only use marijuana during pregnancy but are also willing to admit to researchers that they do may differ from those who don't.

An Australian study published in 2020 found differences in birth size and length, head size, and gestational age at birth in the offspring of women who had still used marijuana at or after 15 weeks. But "the mean age and socio‐economic status of women who continued to use cannabis were lower, and their mean anxiety and depressive symptom scores higher than for other participants; the proportions who consumed alcohol, used other illicit drugs, or were smoking at 15 weeks of pregnancy were also higher." (Notably, "neonatal outcomes for babies of women who quit before or during early pregnancy were not significantly different from those for infants of women who had never used cannabis.")

A review of evidence published in February 2020 "points to the possibility of lower birth weight, diminished IQ and more behavior problems among children whose mothers used cannabis during pregnancy, but notes it is very difficult to separate the marijuana use from other demographics or other variables," writes Oster.

Another 2020 study purported to show children exposed to marijuana in the womb had a range of "attention, thought, and social problems" later in life by looking at data on 9- to 11-year-olds in a large longitudinal study. But since women who admit to using marijuana during pregnancy are different in myriad ways from those who don't, there could be a range of differences in the heritable traits or early childhood experiences of their offspring, which could be responsible for subsequent cognitive differences.

When public health officials in the U.S. say that "no amount of marijuana use during pregnancy…is known to be safe," this should be taken with a heaping grain of salt. Phrases like "known to be safe" or "proven to be safe" are frequently applied to the consumption of all kinds of substances—alcohol, artificial sweeteners, certain foods, prescription and over-the-counter drugs, even nutritional supplements—during pregnancy. This is because studies directly testing the safety of these substances for pregnant women can't ethically be done. Hence, a huge range of legal and illegal substances aren't "known to be safe" for developing fetuses, but that doesn't mean they're known to be harmful.

We know some substances to be unsafe during pregnancy—drinking lots of alcohol, smoking tobacco, consuming too little folic acid. Marijuana isn't one of them.

With the limited evidence available, it may make sense for most pregnant women to avoid marijuana to minimize possible risks to their offspring. But the best choice for one woman and her baby won't be the best choice universally. For women who have extreme morning sickness that makes getting adequate nutrients through food and vitamins difficult, and for whom marijuana mitigates nausea, using cannabis might make sense. Likewise, women with certain mental health conditions helped by marijuana may deem it safer than their usual prescription drugs.

Ultimately, pregnant women should make these decisions for themselves, in consultation with their doctors. Unfortunately, the state doesn't always see it this way.