LAUSR

OBJECTIVE To identify demographic and urban environmental variables associated with prevalence rates of dog bites per zip code in Detroit. DESIGN Retrospective ecological study. SAMPLE 6,540 people who visited any 1 of 15 hospital emergency rooms in the 29 zip codes in Detroit between January 1, 2006, and December 31, 2013, with a primary complaint of dog bite. PROCEDURES The number of dog bites over the study period was determined per zip code. Data for the human population in each zip code in 2011 and demographic and urban environmental variables were obtained from federal, state, and municipal databases. The prevalence rate of dog bites in each zip code was calculated, and regression analysis was used to identify variables associated with this outcome. RESULTS Results of multivariate analysis indicated that demographic variables (eg, gender, age, and education) accounted for 23.2% (adjusted R2 = 0.232) of the variation in prevalence rates of dog bites per zip code, whereas urban environmental variables (eg, blight, crime with weapons, and vacancy rate) accounted for 51.6% (adjusted R2 = 0.516) of the variation. CONCLUSIONS AND CLINICAL RELEVANCE Findings suggested that demographic variables had poor association with variation in prevalence rates of dog bites per zip code, whereas urban environmental variables, particularly crime, vacancy rate, and blight, were better associated. Thus, public health and education policies need to address these urban environmental issues to lower the prevalence of dog bites in distressed urban areas.