Abstract
BACKGROUND: Dog bites can have an array of negative health impacts on victims. Research focusing on the correlates of bites focused on limited sets of variables and produced conflicting findings. OBJECTIVE: To expand knowledge about the correlates of dog bites by exploring a comprehensive set of variables related to the nature of the dog and the circumstances surrounding the bite not commonly explored in extant research. METHODS: Data were drawn from police department reports of dog bites in the city of Detroit between 2007-2015; 478 dog bites were reported. Multiple regression was used to determine the significant correlates of dog bites, focusing on the nature of the dog and the circumstances surrounding the bite. RESULTS: Bites were caused by a neighborhood dog. Thirty-two percent of the reports involved dogs running loose; 25% dogs that had escaped from a fenced or unfenced yard; 9% escaped from their home; and 8% had broken off a chain, were being walked, or were in their own home. Based on multiple regression, the victim was most likely bitten in their own yard by a single neighborhood dog that escaped from its home or yard. Breed of dog was not correlated with bites in multiple regression. CONCLUSIONS: The greatest risk of bites does not come from wandering feral dogs. Based on multiple regression, the victim was most likely bitten in their own yard by a single neighborhood dog that escaped from its home or yard. Human error often contributes to bites.