Early-Pregnancy Cannabis Exposure Was Not Linked to Autism or Early Developmental Delays in Two Large Cohort Studies

Two large studies published in JAMA Network Open found no statistically significant association between cannabis use reported or detected in early pregnancy and either autism spectrum disorder (ASD) or several early childhood developmental delays. The findings add important context to an unsettled area of research, but they do not establish that cannabis is safe during pregnancy.

The studies, published October 18, 2024, analyzed electronic health records from Kaiser Permanente Northern California. Both were observational cohort studies supported by grants from the National Institute on Drug Abuse and the National Institutes of Health.

No significant association with autism after adjustment

In the study of maternal prenatal cannabis use and childhood ASD, researchers examined 178,948 singleton pregnancies involving 146,296 individuals whose children were followed through age 12 or younger. Cannabis exposure was assessed when patients entered prenatal care—generally around 8 to 10 weeks of gestation—using self-report, urine toxicology, or both.

Cannabis use was identified in 8,486 pregnancies, or 4.7% of the cohort. ASD was diagnosed in 3.6% of the children. Before adjustment, the cannabis-exposed group appeared to have a higher rate of ASD diagnoses. That difference was substantially reduced after researchers accounted for sociodemographic characteristics, prenatal care, other substance use, and maternal medical and mental-health conditions. In the fully adjusted analysis, prenatal cannabis use was not significantly associated with ASD, with a hazard ratio of 1.05 and a 95% confidence interval of 0.84 to 1.32.

The researchers did, however, identify an unresolved signal involving heavier use. Daily users had a higher estimated risk, but the association was no longer statistically significant after adjustment. Because the daily-use subgroup was relatively small and exposure was measured at only one point in pregnancy, the result should be treated as a reason for further study—not as evidence that frequent use is harmless or definitively linked to autism.

No increased risk of the developmental delays studied

The companion study of early developmental delays included 119,976 pregnancies involving 106,240 individuals. The children were followed to age 5.5 years or younger, and researchers examined diagnoses involving speech and language disorders, global developmental delay, and motor delay.

Cannabis use was documented in 6,778 pregnancies, or 5.6% of the cohort. After adjustment for potential confounding factors, the researchers found no significant association between prenatal cannabis exposure and any of the three outcomes. The adjusted hazard ratios were 0.93 for speech and language disorders, 1.04 for global developmental delay, and 0.86 for motor delay; in each case, the confidence interval included the possibility of no association.

A sensitivity analysis based only on urine toxicology found a modest inverse association with speech and language disorders. That result does not show that cannabis improves development. The researchers noted that the finding differed from analyses based on self-reported use and could reflect differences in how exposure was measured or other unmeasured characteristics of the participants. Toxicology testing also indicated whether THC was detected, not the dose, potency, duration, or route of cannabis use.

What the studies do—and do not—show

Neither study was a randomized trial, so the findings cannot prove that prenatal cannabis exposure has no effect on child development. Cannabis use was measured early in pregnancy rather than repeatedly throughout gestation, and the researchers lacked detailed information about THC concentration, product type, method of consumption, and continued use later in pregnancy. Follow-up and diagnosis may also have differed between families who reported cannabis use and those who did not.

The studies also focused on specific outcomes. They do not rule out possible associations with other childhood outcomes, later-emerging neurodevelopmental or behavioral differences, or adverse pregnancy and newborn outcomes. The authors specifically noted that prior research has associated prenatal cannabis exposure with adverse neonatal outcomes and that further work is needed to examine patterns of use across pregnancy and the strength of cannabis products.

For that reason, the results should be read as reassuring but limited: in these large California cohorts, early-pregnancy cannabis exposure was not associated with ASD or the developmental delays measured during early childhood after statistical adjustment. “Not associated” is not the same as “proven safe.” The Centers for Disease Control and Prevention advises against cannabis use during pregnancy and recommends discussing cannabis use with a health care professional.

The consequences of punitive responses

The medical evidence exists within a broader legal and public-health debate. Pregnancy Justice’s report on pregnancy criminalization documented nearly 1,400 cases involving arrests, prosecutions, or heightened legal restrictions related to pregnancy between 2006 and the Supreme Court’s June 2022 decision in Dobbs v. Jackson Women’s Health Organization. The report found that most cases involved allegations of substance use, often pursued through child-neglect or endangerment theories rather than laws specifically prohibiting drug use during pregnancy.

Those cases raise concerns about the effects of surveillance and criminal penalties on access to prenatal care. People who fear that disclosing substance use could lead to arrest, family separation, or other legal consequences may be less likely to seek medical help. A supportive approach—one that offers confidential counseling, treatment, and help addressing underlying health or social needs—is more consistent with the studies’ limitations and with the need for patients to receive accurate medical guidance.

The two JAMA Network Open studies help narrow one part of the evidence gap, but they do not settle the question of prenatal cannabis safety. Clinicians should explain what the research found, acknowledge what remains unknown, and continue to recommend avoiding cannabis during pregnancy while connecting patients who use it with nonjudgmental care and support.