307 - Healthcare utilization and development among NC infants exposed to marijuana in utero

Publication Number: 307.5116

Brittany J. Raffa, UNC Chapel Hill, Chapel Hill, NC, United States; Paul Lanier, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, United States; Carl J. Seashore, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, United States; Yumei Yang, University of North Carolina at Chapel Hill Gillings School of Public Health, Chapel Hill, NC, United States; Feng-Chang Lin, UNC Chapel Hill, Chapel Hill, NC, United States; Samantha Schilling, UNC-CH, Chapel Hill, NC, United States

Background: Despite marijuana use by 7% of pregnant women, research on health outcomes and healthcare utilization associated with in utero marijuana exposure has gaps and is complicated by confounders.

Objective: To examine the association between in utero marijuana exposure and well child check (WCC) attendance, emergency department (ED) visits, and presence of developmental delay (DD) during the first two years of life.

Design/Methods: For this retrospective cohort study, a merged dataset was created using Carolina Data Warehouse (CDW), a central data repository from UNC Health Care, and NC Medicaid. Infants with a meconium drug screen sent during the birth hospitalization between 04/01/14 and 04/30/22 were identified from CDW and linked with NC Medicaid, which included claims until 04/30/24. Infants were categorized as marijuana exposed (meconium positive for marijuana only) or substance unexposed (meconium negative and urine absent or negative for all substances). Others were excluded (Figure 1).

The primary outcome was WCC adherence; secondary outcomes were ED encounters and presence of DD in the first two years. Covariates included race, ethnicity, sex, gestational age, postpartum depression, and complex care conditions. Additional covariates were included for DD analysis (Table 2). Poisson and logistic regression were used to examine the association between marijuana exposure and outcomes. DD sub-analysis was conducted among children with three years of data.

Results:

Results: Among 7,257 infants with a meconium screen, 5,458 infants (75%) were linked to NC Medicaid. (Figure 1) There were 1,861 infants with a meconium test positive for marijuana. There were 2,318 infants with negative tests for all substances. (Table 1) No difference in WCC adherence or ED visits was observed between marijuana exposed and substance unexposed infants. There was a decrease in odds of DD in the first 2 years (OR = 0.81, p = 0.02) among marijuana exposed infants and no difference in DD at 3 years (OR = 0.99, p = 0.86). (Table 2)

Conclusion(s): This study was conducted in the setting of mandated Child Protective Services (CPS) reporting for positive marijuana screens. Compared to those unexposed, Medicaid insured children who were exposed to marijuana in utero have similar WCC attendance and ED use over the first 2 years, and similar developmental outcomes at 3 years. Because it is unlikely that marijuana exposure promotes positive development, there are likely unmeasured factors contributing to this finding such as a positive home environment. It is also possible that CPS intervention promotes positive development.

Study Population Diagram


Study Population Characteristics


Multivariable Analysis of Outcomes


Study Population Diagram


Study Population Characteristics


Multivariable Analysis of Outcomes