Dog bites are consistently among the top 15 causes of nonfatal injuries in the general population and top 10 among 5-9 year olds.1 Children are more likely to be bitten than adults, and these bites are more likely to involve the head and neck region with potential for long-term aesthetic, psychological, and functional consequences.2 This serious public health issue has yet to be adequately addressed in terms of policy and education among vulnerable populations. In designing educational programs, there is a need to fully understand all potential risk factors and provocations. Currently, we lack sufficient knowledge of the circumstances surrounding the injury. This study aims to answer the following: In all patients presenting to a tertiary care hospital for treatment of a facial dog bite, are there identifiable risk factors associated with the injury? In particular, we wish to focus on those provocations and behavioral issues that can be modified in an effort to decrease the incidence of this devastating injury. This will help focus efforts on improved education and outreach programs and identify vulnerable populations who would most benefit. This is a retrospective study analyzing the facial dog bites treated at Massachusetts General Hospital from 1997-2018. Inclusion criteria, in addition to the diagnosis of facial dog bites, were records that included specific information about the circumstances surrounding the bite and descriptors of the type and location of the injury. Incomplete records resulted in exclusion. The predictor variables included: demographics of the victim (e.g., age and gender), circumstantial information regarding the injury (e.g., location of injury, relationship of victim to dog, provocational circumstances), and dog characteristics (e.g., dog breed, fate of attacking dog, and dog’s history of aggressive behavior). Data analyses were performed using IBM SPSS Statistics 25. Descriptive statistics, such as mean, range, and standard deviations, were calculated for applicable study variables. To measure the association among variables of interest, either bivariate t-tests or chi-square testing was performed for the appropriate variables. A P-value of .05 was considered statistically significant. A total of 321 patients were identified with a mean age of 29.5 years (range 0.7-81 years). There were 141 males and 180 females. The majority of patients were adults [> 18 years of age, n = 223(69.5%)]. A preponderance of dogs [n = 281(87.5%)] were known (P < .00001), and provocation was recorded in 207 cases (64.5%, P < .00001). Bites were preceded by the following behaviors: playing with, feeding, and placing face close to the dog. Pitbulls led in number of bites [n = 26 (8.5%)]. Location on the face was predominantly the middle or lower third [n = 299 (93.1%)]. The type of injury (often mixed) included 284 (88.5%) lacerations, 72 (22.4%) punctures, 48 (15.0%) avulsions, and 20 (6.2%) involving contiguous structures including the mental nerve, ear, gingiva, nasal cartilage, and Stenson’s duct. This study highlights the role of provocation in dog bite injuries and suggests that vulnerable populations may benefit from efforts aimed at improved education, outreach programs, and policy changes. Future initiatives will focus on mandatory documentation and reporting to ensure complete collection of information. More effective methods of promulgating educational information is required. Partnering with veterinarians, pediatricians, rescue and adoption agencies, and pet supply companies will help bring this about.