OBJECTIVE To examine medical and demographic factors associated with the firearm-related deaths among children in Kentucky. METHODS Retrospective review and multiple regression analysis. METHODS All firearm-related deaths among children younger than 20 years reported to the Kentucky Office of Vital Statistics, Frankfort, from 1988 to 1993. METHODS None. RESULTS All 320 pediatric firearm-related deaths that occurred in Kentucky from 1988 to 1993 were analyzed. Death rates were calculated for each county in the state. While the overall death rate from firearms was not significantly different between African-American and white children (relative risk [RR], 1.39; 95% confidence interval [CI], 0.98-1.98), the pattern of the types of events was markedly different. African American children were much more likely to have been involved in a homicide; suicides were more frequent in white children. Multiple Poisson regression analysis, controlling for age, race, and gender, identified only 1 variable that was significantly associated with deaths due to firearms. Children in rural Kentucky were at significantly more risk for a firearm-related death than children in urban areas (RR, 1.26; 95% CI, 1.01-1.62) even after controlling for medical system variables (availability of a hospital with 24-hour emergency services, availability of prehospital advanced life support, and availability of 911 service). CONCLUSIONS Children in rural areas of Kentucky are at an increased risk for firearm-related mortality. Prevention and intervention programs that focus only on urban areas may not produce optimum results in the Kentucky setting. Further research is needed to determine factors that are important in rural areas so that interventions specific to them can be planned.