Prenatal cannabis exposure and the risk of subsequent maltreatment

Abstract

Background

Parental substance use can increase the risk of child maltreatment.

Objective

The purpose of this study was to assess racial bias in newborn drug testing and to investigate the association between prenatal tetrahydrocannabinol (THC) exposure and subsequent child maltreatment.

Participants and Setting.

This retrospective cohort study (n = 35,437) linked University of Michigan Hospital birth data and Michigan Department of Health and Human Services child maltreatment data relative to a 2018 policy change. Prior to 2018, prenatal THC exposure was routinely substantiated as physical abuse; after 2018 THC exposure was investigated but not automatically substantiated.

Methods

We defined prenatal THC exposure as a positive newborn meconium drug test for THC. The primary outcome was a substantiated Child Protective Services (CPS) report of maltreatment before and after the policy change. Demographic variables included parent age, race, ethnicity, zip code and insurance type. Covariates included prenatal urine drug test orders and results, and newborn drug test orders and results. Regression models estimated the rate of subsequent maltreatment and racial disparities associated with newborn testing.

Results

Regression analyses indicated that Black and multiracial newborns were significantly more likely to be tested for substance exposure at birth. Newborns with a test positive for THC only were not more likely to experience maltreatment after the policy change as compared with newborns that tested negative and newborns not tested.

Conclusions

The evidence strongly supports a policy to end routine CPS investigations for cannabis exposure and eliminate racially biased drug testing practices.