Effects of Prenatal Cannabis Exposure (PCE) on Language Development in Children Open Access

Karthikeyan, Manjushri (Spring 2025)

Permanent URL: https://etd.library.emory.edu/concern/etds/6w924d50k?locale=en
Published

Abstract

Despite cannabis being the most commonly used illicit drug amongst pregnant women in the United States, literature on the effects of prenatal cannabis exposure (PCE) is limited. Previous studies allude to associations between maternal cannabis use during pregnancy (MCDP) and challenges in achieving child neurodevelopmental outcomes, such as delays in speaking their first words. However, findings in this field remain mixed. This study seeks to investigate the relationship between PCE and language development in children aged 8 to 12 further– distinguishing between expressive and receptive language outcomes, while also understanding how PCE may influence language development after controlling for confounding environmental factors. We used an initial cohort of 11,598 mothers and children from the Adolescent Brain Cognitive Development (ABCD) dataset to note maternal reporting of cannabis use. Expressive and receptive language development outcomes were assessed using picture-vocabulary and oral-reading assessments developed by the NIH-Toolbox– conducted at baseline when children were aged 8-10 and during a 2-year follow-up when the children were aged 10-12. A total of 23 maternal, familial, and environmental risk factors were accounted for using propensity scores to address what made a mother more likely to smoke cannabis during her pregnancy. The propensity scores were incorporated in a series of logistic regression models assessing reading outcomes at baseline, reading outcomes at the 2-year follow-up, picture vocabulary outcomes at baseline, and picture vocabulary outcomes at the 2-year follow-up. While there was no robust effect of MCDP on the reading or picture-vocabulary outcomes, the propensity for mothers to use cannabis during their pregnancy predicted performance on both assessments across both time points. Overall, the strongest MCDP predictor of a decrease in language ability was the propensity for being exposed to cannabis during pregnancy. This was associated with maternal age, unplanned pregnancy, maternal immediate family drug use, alcohol use during pregnancy, tobacco use during pregnancy, maternal rule-breaking, biological mother, paternal immediate family drug use, parental impulsivity psychopathologies, neighborhood safety, and race. These findings set a foundation for future PCE methodology, clinical markers and neurobehavioral initiatives addressing language development, and targeted public health projects addressing risk factors from our propensity scores.

Table of Contents

Abstract

Acknowledgments

Table of Contents

Introduction 1

The Significance of Teratology Research and the Impact of Further Marijuana Legalization 1

Neuronal Mechanisms and the Consequences of Prenatal Cannabis Exposure (PCE) 2

Receptive and Expressive Language Development in Children 4

Hypothesis and Study Goals 5

Methods 6

ABCD Dataset, Sample Origin, and Participant Descriptives 6

Independent Variables of Interest and Assessments 7

ABCD Parent Demographics Survey 7

ABCD Developmental History Questionnaire 8

Propensity for MCDP Derivation 8

Demographic covariates 9

Dependent Variables of Interest and Assessments 10

Picture Vocabulary Test and Oral Reading Assessment 10

Data Preparation and Analysis Plan 11

Results 11

Discussion 13

Limitations 17

Future Recommendations 18

References 20

Appendices 37

Tables 37

Table 1. Screenshot of questionnaires and data dictionary of variables cleaned from the ABCD 5.0 Data Repository. 37

Table 2. ABCD Study Sample Characteristics. 39

Table 3. Reading and Picture Vocabulary Outcomes associated with MCDP and Propensity Score. 40

Table 4. Familial risk factors/confounders associated with PCE from propensity score analysis (Ikeda et al., 2024). 41

Figures 42

Figure 1. Path model depicting the association between prenatal cannabis exposure and delays in first-word outcomes while accommodating for the following covariates: sex, race, and income (Ikeda et al., 2024). 42

Figure 2. ROC Curve for propensity score model (Ikeda et al., 2024). 43

About this Honors Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
School
Department
Degree
Submission
Language
  • English
Research Field
Keyword
Committee Chair / Thesis Advisor
Committee Members
Last modified

Primary PDF

Supplemental Files