Nathan R Manley, Dih-Dih Huang, Richard H Lewis, Tiffany Bee, Peter E Fischer, Martin A Croce, Louis J Magnotti

Publicly available firearm data is difficult to access. Trauma registry data is excellent at documenting patterns of firearm-related injury. Law enforcement data excels at capturing national violence trends to include both circumstances and firearm involvement. The goal of this study was to utilize publicly available, law enforcement data from all 50 states to better define patterns of firearm-related homicides in the young. All homicides in individuals < 25 years old in the United States over a 37-year period ending in 2016 were analyzed: infant < 1 year old, child > 1-9 years old, adolescent 10-19 years old, young adult 20-25 years old. Primary data files were obtained from the Federal Bureau of Investigation and comprised the database. Data analyzed included homicide type, situation, circumstance, month, firearm type and demographics. Rates of all homicides and firearm-related homicides per 1 million (M) population and the proportion of firearm-related homicides (out of all homicides) were stratified by year and compared over time using simple linear regression. 171,113 incidents of firearm-related homicide were analyzed (69% of 246,437 total homicides): 5,313 infants, 2,332 children, 59,777 adolescents and 103,691 young adults. Most were male (88%), black (59%) with a median age of 20 years old. Firearm-related homicides peaked during the summer months of June, July and August (median = 1,156 per year; p = 0.0032). Rates of all homicides (89 to 53 per 1M population) and firearm-related homicides (56 to 41 per 1M population) decreased significantly from 1980 to 2016 (β = -1.12, p < 0.0001 and β = -0.57, p = 0.0039, respectively). However, linear regression analysis identified a significant increase in the proportion of firearm-related homicides (out of all homicides) from 63% in 1980 to 76% in 2016 (β = 0.33, p < 0.0001). For those 25-years-old and younger, the proportion of firearm-related homicides has steadily and significantly increased over the past 37-years, with three out of four homicides firearm-related in the modern era. Despite focused efforts, reductions in the rate of firearmrelated homicides still lag behind those for all other methods of homicide by nearly 50%. That is, while the young are less likely to die from homicide, for those unfortunate victims, it is more likely to be due to a firearm. This increasing role of firearms in youth homicides underscores the desperate need to better direct prevention efforts and firearm policy if we hope to further reduce firearm-related deaths in the young. Level III, epidemiological study.