The Impact of Punitive Prenatal Substance Abuse Policies on Maternal Health Behaviors and on Birth Outcomes

In response to the high prevalence of cocaine use among Americans in the late 1980s and national media attention to infants born addicted to cocaine, state legislatures proposed bills to fund treatment services for pregnant women and their children and debated civil and criminal penalties for prenatal drug use. By January 2000, 48 states and the District of Columbia had enacted targeted prenatal substance abuse policies. These policies ranged from non-punitive drug treatment programs for pregnant women to punitive civil commitment and child abuse statutes. For example, high-profile court cases have highlighted the use of involuntary detention (aka, civil commitment) to penalize rather than to provide treatment to drug-dependent mothers; from 1973 to 2005, more than 400 pregnant women were arrested or forced to receive medical treatment due to illegal drug use (Paltrow and Flavin, 2013). Additionally, in 2014, Tennessee became the first state to explicitly criminalize drug use during pregnancy.

While state legislatures continue to debate civil and criminal penalties for illegal drug use during pregnancy, very little is known about the impact of these punitive policies on maternal health behaviors and on birth outcomes. However, qualitative research suggests that punitive prenatal substance abuse policies deter women from receiving medical care during pregnancy (Terplan et al., 2009), which could negatively affect maternal health behaviors and birth outcomes.

I use the variation in prenatal substance abuse policy implementation across states from 1985 to 2000 to study the impact of punitive policies on birth outcomes and on maternal health behaviors, as measured by birth certificate records in the National Vital Statistics System. I estimate difference-in-differences models that compare (i) outcomes in states with punitive prenatal substance abuse policies relative to states with non-punitive policies and (ii) outcomes in states that adopted punitive prenatal substance abuse policies before 1995 relative to states that adopted punitive policies after 1995. I also estimate triple-difference models that address potential policy endogeneity by using women ages 26 and older as a placebo group; these models rely on the assumption that older women are significantly less likely to be affected by prenatal substance abuse policies because of their low rates of drug use during pregnancy (see Figure 2), but they are equally likely to be affected other health policies such as expansions in public health insurance, for example.

My preliminary results suggest that civil child abuse laws decrease average birth weight by 33.5 grams and increase the probability of early gestation by 0.7 percentage points. Additionally, civil child abuse laws increase the risk of a low Apgar score (indicating fetal distress) by 0.2 percentage points. Finally, in support of the qualitative evidence, I also find that civil child abuse policies decrease the probability of receiving any prenatal care by 2.4 percentage points and decrease the probability of initiating prenatal care during the first trimester by 1.8 percentage points.