Age-Specific, Gun-Related Homicide Rates in New York City: Investigating the Relative Contribution of Policing, Cocaine Markets, Firearm Availability, Incarceration and Alcohol Consumption to the Homicide Decline during the 1990s in Different Age Groups

Magdalena Cerda, The New York Academy of Medicine
Steven F. Messner, University at Albany, State University of New York (SUNY)
Melissa Tracy, University of Michigan
David Vlahov, The New York Academy of Medicine
Emily Goldmann, University of Michigan
Kenneth Tardiff, Cornell University
Sandro Galea, University of Michigan

We assessed whether changes in neighborhood misdemeanor policing, cocaine consumption, incarceration rate, firearm availability and alcohol consumption had a different relationship with gun-related homicide among youth (15-24), young adults (25-34) and adults (35+) in New York City in the 1990s. We used cross-sectional time-series data for 74 police precincts in 1990-99, and estimated Bayesian hierarchical models. An increase in 5000 misdemeanor arrests was associated with fewer homicides among young adults (median [95% CI]: -18.77 [-27.26, -10.13]) and adults (-3.22 [-5.87, -0.54]). Decreased alcohol consumption was related to declining homicides for young adults and adults, while decreased firearm availability was only associated with decreased homicides for young adults. The only conditions associated with change in homicide among youth were changes in cocaine consumption (0.26 [0.08, 0.45]) and changes in the incarceration rate (0.04 [0.01, 0.06]). This study illustrates the importance of investigating age-specific patterns to understand the mechanisms underlying homicide trends.