Summary & Analysis: Dog bites to humans–demography, epidemiology, injury, and risk

To best understand this article in the context of the literature on growling, snarling, snapping, and biting behavior (incidence and correlates), please see National Canine Research Council’s complete analysis here.

Article Citation:
Overall, K. L. & Love, M. (2001). Dog bites to humans—demography, epidemiology, injury, and risk. Journal of the American Veterinary Medicine Association, 218(12), 1923-1934.

National Canine Research Council Summary & Analysis: 

This paper is included as one of the few attempts at a comprehensive review of the dog bite literature. In the 2001 special report, Overall and Love summarized the existing data on dog bites. Much like our current review, the authors discussed incidence of dog bite injuries and factors related to the incidents, including prevalent human and canine demographics, epidemiology, and risk factors. The purpose of the report was to curate any useful data that correlated with dog bites, perhaps to be utilized for legislative purposes. They concluded that the only robust findings in the literature were that children, particularly boys, were disproportionately affected, that human behavior is a major factor, and that more owned dogs than strays were involved. The authors also aimed to dispel any myths surrounding aggression and breed by pointing to objective data. This is not an original research article, but rather a review of the relevant literature at the time. The paper highlighted the inconsistencies in how dog bite data is collected and analyzed between various studies and so presented most of their findings as attempts to establish a plausible range of conclusions, rather than definitive answers.

The data reported is generally accurate and often involves averaging of statistics from diverging papers without making judgments about which study cited might contain the most reliable findings. The authors engaged in some of the typical hyperbole to establish the importance of the issue; seemingly dramatic changes in health care expenditures to treat dog bites since the 1940’s were cited without mentioning that simple inflation would account for more than half of the change, and that when compared to the increases in health care expenditures generally, dog bite treatment costs increased at only half the rate of general costs. They speculated that veterinarians may lose income if people become frightened of the risk of dog bites and consequently own fewer dogs, an outcome that has failed to materialize in any period in the last 70 or 80 years in the U.S.

Overall and Love (2001) noted that estimates of dog bite incidence in the U.S. vary widely and range from 500,000 to 4.7 million bites per year and provided an extremely useful list of studies showing this divergence, which appears to be primarily a contrast between locally based records of bites reported to authorities and randomized health surveys such as Sacks et al. (1996). Based on the number of bites medically treated each year, the authors estimated that 17% of bites are reported; however, this presumes that all medically treated bites are reported. They also estimated that 1-2% of the medically treated bites are severely injurious (require hospitalization). 

Since this paper was written before the practice of visual breed identification of dogs of unknown origin had been discredited (Voith et al., 2009; 2013; etc.), the authors included various reports of incidence related to breed and engaged in some behavioral speculation based on these identifications. However, this information is heavily qualified by recognized difficulties in relative population size of various breeds and obvious media bias in reporting. The authors recognized that data surrounding dog bites may be biased – those that are more severe or those that occur from an unfamiliar dog may be more likely to be reported or treated. They noted that some data, such as human and environmental factors surrounding a bite incident, are rarely documented.

Lastly, the authors expressed a frustration regarding the inconsistent and/or lacking definitions of key terms. For example, many studies do not define what constitutes a bite and do not differentiate between injurious and non-injurious incidents; minor bruises and severe punctures are all too often grouped together as “bites.”

Link to Full Text of the Original Article: