Abstract

Importance: The number of children with prenatal opioid exposure to medication for addiction treatment (MAT) with methadone and buprenorphine for maternal opioid use disorder is increasing, but the associations of this exposure with cognitive outcomes are not well understood.

Objective: To examine the strength and consistency of findings in the medical literature regarding the association of prenatal exposure to MAT with early childhood cognitive development, particularly when accounting for variables outside MAT exposure.

Data Sources: A search strategy obtained publications from PubMed, CINAHL, PsycINFO, Web of Science, and Embase from January 1972 to June 2019. Reference lists from identified articles were searched.

Study Selection: Inclusion criteria were cohort studies, studies including children aged 1 to 60 months with at least 2 months of prenatal MAT exposure, studies using standardized direct-observation testing scales, and studies reporting means and SDs. Case reports, case series, historical controls, and reviews were excluded.

Data Extraction and Synthesis: Two authors independently selected studies for inclusion, extracted data, and assessed study quality. Data extracted included demographic characteristics, covariates, sources of bias, and effect estimates. Meta-analysis was performed using random-effects models. This study was conducted according to the Meta-analysis of Observational Studies in Epidemiology (MOOSE) guidelines and the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement. Data extraction and synthesis were conducted between January 2018 and August 2019.

Main Outcomes and Measures: Cognitive test scores and demographic variability between exposed and unexposed groups.

Results: A total of 16 unique cohorts, described in 27 articles and including 1086 children (485 [44.7%] with MAT exposure), were included in a quantitative synthesis. On meta-analysis, MAT exposure was associated with lower cognitive development scores (pooled standardized mean difference, −0.57; 95% CI, −0.93 to −0.21; I2 = 81%). Multiple subanalyses on demographic characteristics (ie, maternal education, race/ethnicity, socioeconomic status, prenatal tobacco exposure, infant sex) were conducted. In the subanalysis of studies with comparable prenatal exposure to tobacco smoke, the association of MAT exposure with cognitive scores was no longer statistically significant and became homogeneous (standardized mean difference, −0.11; 95% CI, −0.42 to 0.20; I2 = 0%).

Conclusions and Relevance: In this study, predefined subanalyses demonstrated how poor recruitment, particularly imbalances in maternal tobacco use, could contribute to a negative overall association of cognitive development test scores with prenatal MAT exposure. Promoting tobacco cessation for pregnant women with opioid use disorder should be prioritized in this high-risk population.