Maternal Cannabis Use Before (Not During) Pregnancy Was Linked to Higher Autism Symptoms in Children | Research

The takeaway

In a prospective study of 4,380 children, maternal cannabis use before pregnancy (but not during) was associated with higher autism symptoms across childhood, while continued maternal tobacco use during pregnancy was independently linked to autism symptoms at age 6.

Obstetricians, developmental pediatricians, prenatal health researchers

Preconception, not prenatal, cannabis use was associated with higher autism symptoms in children

What the researchers found

Maternal cannabis use before but not during pregnancy was associated with higher CBCL autism symptoms across childhood (beta: 0.33, CI: 0.02-0.63). Paternal cannabis use was also linked to higher CBCL symptoms (beta: 0.27, CI: 0.05-0.50) but was explained by maternal psychopathology. Continued maternal tobacco use during pregnancy was associated with SRS autism symptoms at age 6 (beta: 0.03, CI: 0.003-0.05), but paternal tobacco was not, suggesting intrauterine rather than familial effects.

Why it matters

This study separates the effects of preconception vs. prenatal cannabis exposure and maternal vs. paternal use, helping distinguish between direct in utero effects and familial/genetic confounding. The finding that preconception (not prenatal) cannabis use predicts autism symptoms suggests epigenetic or selection effects rather than direct fetal exposure.

The numbers in context

4,380 children; maternal preconception cannabis: beta 0.33 (CI: 0.02-0.63) for CBCL autism; paternal cannabis: beta 0.27 (CI: 0.05-0.50) for CBCL; no association with SRS autism; maternal continued tobacco: beta 0.03 (CI: 0.003-0.05) for SRS autism; no paternal tobacco association

How the study worked

Prospective population-based cohort (n=4,380) measuring parental cannabis and tobacco use via questionnaires and maternal cannabis metabolites via urinalysis. Autism symptoms measured using CBCL at ages 1.5, 3, and 6, and SRS at age 6. Linear mixed models and linear regression used.

What this study cannot tell us

Self-reported substance use (though supplemented with urinalysis for cannabis), observational design cannot prove causation, autism symptoms measured by mother report not clinical diagnosis, effect sizes are small, cannot rule out all confounding

How to read the evidence

Large prospective cohort with biological verification of cannabis exposure and longitudinal symptom measurement; strong design but observational

When this study was published

Published 2025

The bigger picture

The pattern of results suggests that cannabis-related autism risk may operate through family-level factors (genetics, parental psychopathology) rather than direct in utero exposure, while tobacco appears to have a genuine intrauterine effect since only maternal (not paternal) use predicted symptoms.

Questions still open

Why would preconception cannabis use predict child autism symptoms if prenatal use does not? Could epigenetic changes from cannabis use persist beyond cessation? Does maternal psychopathology (which explains the paternal cannabis association) independently affect child neurodevelopment?

Common questions

Does cannabis use during pregnancy cause autism?

In this study of 4,380 children, cannabis use during pregnancy was not associated with autism symptoms. Surprisingly, cannabis use before pregnancy was linked to higher symptoms, suggesting the association may reflect family-level factors rather than direct fetal exposure.

Is tobacco use during pregnancy linked to autism?

Yes. Maternal continued tobacco use during pregnancy was associated with autism symptoms at age 6, and since paternal tobacco showed no association, the effect appears to be from direct in utero exposure rather than shared family factors.

Read the original research

The association of preconception and prenatal cannabis and tobacco exposure with autism symptoms in offspring: A population-based longitudinal study.

Neurotoxicology and teratology, 112, 107561

View on PubMed ↗Publisher’s paper ↗

Citation

Cajachagua-Torres, Kim N; Boer, Olga D; Louwerse, Anneke; Ghassabian, Akhgar; Reiss, Irwin K M; Jaddoe, Vincent W V; El Marroun, Hanan. (2025). The association of preconception and prenatal cannabis and tobacco exposure with autism symptoms in offspring: A population-based longitudinal study.. Neurotoxicology and teratology, 112, 107561. https://doi.org/10.1016/j.ntt.2025.107561