Scales of geography, time, and population: the study of violence as a public health problem

OBJECTIVES: In this study, data on violent deaths in the Bronx, New York City, from the 1970, 1980, and 1990 censuses were analyzed. METHODS: The incidence and areal density of intentional deaths were mapped by health area. Simple and stepwise regressions between violent death measures and other factors were performed. RESULTS: The incidence of deaths at levels of those in the highest 1970 quintile spread so that by 1990 only 2 areas saw incidences at levels of the lowest 1970 quintile. Overcrowding, socioeconomic status, population, population change, and drug deaths in simple regressions and overcrowding, socioeconomic status, and low-weight births in stepwise regressions correlated significantly with violent death incidence or density. CONCLUSIONS: Understanding the spatiotemporal development of violence can contribute to public policy on violence.