The relationship between proximity to homicide and birth outcomes

This study explores the effect of acute in utero exposure to homicides on a range of birth outcomes by testing theories of stress response and critical periods of fetal development. Specifically, this study examines the effect of in utero exposure to homicide on the birth outcomes of infants whose mothers were in close proximity to the homicide compared to infants whose mothers were unexposed to homicide during gestation. This study further investigates how the effect of exposure varies by gestational age at the time of exposure. The data utilized in this analysis are drawn from New Jersey birth records from 1998-2002 and homicide records used in a spatial and ethnographic investigation conducted by the New Jersey Star-Ledger newspaper. The overall analysis in the standard regression model showed that there are quite small, but highly significant positive effects on birth weights. Under a sibling fixed effects rubric these effects go away entirely, indicating that perhaps unobserved familial factors were driving results. Furthermore, the closer a woman and her fetus were to a homicide did not linearly affect the birth outcomes of the fetus. There was also little difference between birth outcomes of mothers residing in low crime and high crime areas. While, mothers in low-crime areas had slightly better birth outcomes, the different was marginal. Falsification tests disproved most significant findings. The questions under investigation would benefit from exploration in locales with greater variation in exposure levels.