Police Violence, Maternal Stress and Child Health: Exploring the Association between Police Killings and Poor Birth Outcomes

Background: Police violence is endemic to the United States, but public awareness of this crisis has exploded in recent years in response to multiple high-profile police killings coupled with the proliferation of social media. In addition to the direct trauma of experiencing police violence, this violence is linked to stress and poor mental health at the individual and community levels. Maternal stress is in turn associated with poor birth outcomes. The objective of this analysis was to evaluate the association between police killings and poor birth outcomes (low birth weight and premature birth) at the U.S. county level. Methods: Data on police killings in the United States from 2013-2017 were sourced from Mapping Police Violence, geocoded and aggregated by county and year, and linked to natality data from CDC WONDER. We used negative binomial regression to model counts of low birth weight and premature babies born in relationship to the degree of police violence in the year prior. Final models included N=2,246 county-years of data. All models controlled for county fixed effects and included a one-year time lag. Adjusted models additionally controlled for county-level poverty, unemployment, health insurance, education, age, income inequality, and violent crime. Results: The number of police killings in a county was not significantly associated with the total number of low birth weight births in the following year (IRR=0.998, p=0.147). When results were stratified by race, this association became slightly negative (Black: IRR=0.998, p=0.028; white: IRR=0.997, p=0.002). There was also a slight inverse association between the count of police killings in a county and the total number of premature births in the subsequent year (IRR=0.998, p=0.047). Again, this association became more prominent when stratified by race (Black: IRR=0.998, p=0.098; white: IRR=0.997, p=0.003). Conclusions: Police killings have mixed associations with low birth weight and preterm birth at the U.S. county level. Further investigation is needed to better understand these relationships, particularly those stratified by race. There is also an urgent need for a legitimate federal system of capturing police violence data to support the assessment of its public health burden