Seven of the eight states included in the analysis have policies requiring child welfare agencies to automatically share cases of pregnancy substance use with law enforcement, even if the allegations are unfounded. This practice contributes to investigations that often yield no substantiated findings. Among families referred to law enforcement in the eight states, child welfare agencies were 25% more likely to dismiss or divert pregnancy substance use allegations made against Black families compared to White families. In Kentucky, child welfare workers were 38% more likely to close a case without a finding of abuse or neglect when the newborn was Black compared to White.

Hospital drug tests performed at childbirth can produce false positive results as much as 50% of the time, contributing to unwarranted investigations. Real-world consequences of these errors include the 2021 arrest of Ayanna Harris-Rashid in South Carolina after she tested positive for substances due to legal CBD gummies and a topical hemp-based ointment she used during pregnancy. Similarly, in 2024, Melissa Robinson, a Black elementary school librarian in Alabama, was investigated after a false positive test for cocaine. A second hospital drug test disproved the cocaine allegation against Robinson, and her case was closed due to insufficient evidence.

Robinson described the emotional toll of the experience. “To have such a beautiful experience tainted by something like that, it's difficult,” she said. “Truthfully, it's turned me into somebody different.” Miriam Mack, senior legal counsel at Movement for Family Power, attributed the disparities to systemic bias rather than actual differences in substance use.

“It's not about what you do, what substances you use, how much you're using, whether you're using, but it's about who you are,” Mack said. “It's about the simple fact that you are Black and you have given birth in the hospital.”

Oklahoma officials challenged The Marshall Project’s findings, calling the analysis “misleading and inaccurate” because it excludes children who have a multiracial parent if the other parent is not Black or White. In response, The Marshall Project conducted a subsequent analysis and found that the overrepresentation of Black newborns was consistent in the data whether the children of multiracial parents were categorized as Black, White, or neither.

In contrast to the practices in the eight analyzed states, New Jersey has adopted reforms aimed at reducing unwarranted referrals. The state now requires healthcare providers to confirm the results of drug tests with a second, more precise test before reporting new parents to child welfare authorities. Under existing New Jersey law, a positive drug test alone does not constitute child abuse or neglect and is not required to be reported to child welfare. From mid-2018 to mid-2023, New Jersey child welfare officials still received more than 2,300 reports of newborns allegedly exposed to drugs in the womb.

Additional measures in New Jersey include guidance from the Department of Health directing providers to obtain informed consent before drug testing and the creation of an online portal by the Department of Children and Families that allows healthcare providers to submit information about families without specifically identifying them. Jennifer Kirkman, assistant director at the New Jersey Department of Children and Families, said the changes offer reassurance to new parents: “You don't have to worry about child welfare.” Molly Linhorst, staff attorney with the ACLU of New Jersey, called the reforms. “This is a big deal,” she said. “With these in place, fewer people will be unnecessarily referred to [child welfare] and subject to those invasive, traumatic investigations by the state when they have just given birth.”

The investigation highlights how automated reporting policies and unreliable drug screening can disproportionately impact Black families during a vulnerable time—immediately after childbirth. Over 14,000 Black newborns in just eight states were referred to law enforcement without any finding of abuse or neglect, exposing families to traumatic state interventions based on unverified allegations. The racial disparities persist despite evidence that false positives are common and that allegations against Black families are more frequently dismissed.

New Jersey’s reforms demonstrate an alternative approach that prioritizes verification and informed consent, potentially serving as a model for other states. With more than 25,000 Black newborns referred to law enforcement nationwide over seven years, the findings raise urgent questions about equity in child welfare practices and the need for systemic safeguards against racial bias in perinatal drug reporting.