Cannabis Use in Pregnancy Linked to Child Behavioral Outcomes: A Study on Autism Spectrum Disorder
Clinical Takeaway
Prenatal and periconceptional cannabis use is associated with worse behavioral and developmental outcomes in young children, with effects observed across both those with and without autism spectrum disorder symptoms. Children already showing signs of ASD or developmental delay may face compounding risks when exposed to cannabis during the peripregnancy period. These findings support advising patients to avoid cannabis use from the preconception window through delivery.
#34 Associations of maternal peripregnancy cannabis use with behavioral and developmental outcomes in children with and without symptoms of autism spectrum disorder: Study to Explore Early Development.
Citation: DiGuiseppi Carolyn et al.. Associations of maternal peripregnancy cannabis use with behavioral and developmental outcomes in children with and without symptoms of autism spectrum disorder: Study to Explore Early Development.. Autism research : official journal of the International Society for Autism Research. 2025. PMID: 39660543.
Design: 0 Journal: 0 N: 4 Recency: 2 Pop: 3 Human: 1 Risk: 0
Abstract: Some studies report increased prevalence of autism spectrum disorder (ASD) and associated symptoms with prenatal cannabis exposure. We examined whether associations of maternal cannabis use from 3 months preconception through delivery (“peripregnancy”) with behavior and development in the offspring varied with the presence of ASD symptoms. Children ages 30-68 months with ASD symptoms (i.e., met study criteria for ASD or had ASD symptoms on standardized assessments or community ASD diagnosis, N = 2734) and without ASD symptoms (other developmental delay/disorders or general population sample, N = 3454) were evaluated with the Child Behavior Checklist and Mullen Scales of Early Learning. We examined cannabis use during three time periods: peripregnancy, pregnancy, and only preconception. Peripregnancy cannabis exposure was reported for 6.0% of children with and 4.6% of children without ASD symptoms. Preconception-only cannabis use (versus no use) was associated with more aggressive behavior, emotional reactivity, and sleep problems in children with ASD symptoms, but not in children without ASD symptoms. Cannabis use during pregnancy was associated with increased attention and sleep problems in children with ASD symptoms; these associations did not differ significantly by ASD symptoms. Peripregnancy cannabis use was not associated with child developmental abilities regardless of ASD symptoms. In summary, associations of peripregnancy cannabis use with some behavioral outcomes differed in children with and without ASD symptoms. With rising cannabis use among pregnant women, future studies that examine a range of developmental risks associated with timing and patterns of cannabis use prior to conception as well as during pregnancy could inform clinical guidance.
What This Study Teaches Us
In children with autism symptoms, preconception-only cannabis use was linked to more aggressive behavior, emotional reactivity, and sleep problems, while cannabis use during pregnancy was associated with attention and sleep problems. These associations were not seen or were weaker in children without autism symptoms, suggesting the developing autistic brain may be differentially susceptible to cannabis exposure timing.
Why This Matters Clinically
As cannabis use among pregnant and peripregnancy women rises, clinicians need nuanced counseling data. This study suggests the timing of exposure and the child’s neurodevelopmental status may both matter, which could change how we counsel women planning pregnancy or already pregnant, particularly those with family history of autism.
Study Snapshot
| Study Design | Observational case-control study with retrospective maternal cannabis use reporting |
| Population | 6,188 children ages 30-68 months: 2,734 with ASD symptoms and 3,454 without ASD symptoms, drawn from the Study to Explore Early Development (SEED) |
| Intervention | Maternal cannabis use reported for three periods: preconception only, pregnancy, and peripregnancy (3 months preconception through delivery) |
| Primary Outcome | Behavioral outcomes (Child Behavior Checklist) and developmental abilities (Mullen Scales of Early Learning) in offspring |
| Key Result | Preconception-only use associated with aggressive behavior, emotional reactivity, and sleep problems in ASD+ children (6.0% exposure rate) but not ASD- children (4.6% exposure rate); pregnancy cannabis use linked to attention and sleep problems in both groups |
Where This Paper Deserves Skepticism
This is observational data relying on maternal recall of cannabis use from years prior, which is subject to recall bias and social desirability bias (underreporting in pregnant women especially). The abstract does not specify whether dose, frequency, or THC/CBD ratio was captured, making it impossible to know if ‘cannabis use’ means occasional or chronic exposure. Confounding is a major concern: cannabis-using mothers may differ in socioeconomic status, other substance use, mental health, parenting, or genetics in ways that could explain behavioral differences independent of cannabis. The differential effect by ASD status is intriguing but could reflect reverse causality or unmeasured confounding rather than a true biological vulnerability. No information is provided about how ASD diagnosis was confirmed or whether the ‘without ASD symptoms’ group was adequately characterized.
Dr. Caplan’s Take
I appreciate this paper’s attempt to parse timing of exposure and differential susceptibility, which moves beyond crude ‘cannabis in pregnancy equals harm’ messaging. However, observational studies with retrospective exposure assessment are a weak foundation for clinical guidance, and the lack of dose or product information limits real-world applicability. The finding that preconception-only use affects ASD+ but not ASD- children is hypothesis-generating but needs replication in a prospective cohort with validated exposure measurement before I would counsel patients based on it. What I do take seriously is that we have still-limited data on cannabis safety in periconception windows, and pregnant women deserve honest counseling that we don’t yet have the evidence to reassure them, particularly if there is personal or family neurodevelopmental history.
Clinical Bottom Line
This observational study suggests preconception and pregnancy cannabis use may be associated with behavioral problems in children with autism, but the evidence is limited by recall bias and unmeasured confounding. Pregnant women and those planning pregnancy should be counseled that cannabis safety periconceptually remains understudied, and that neurodevelopmentally at-risk pregnancies may warrant extra caution pending stronger evidence.
| |
Have thoughts on this? Share it:
