P0226 - Context of Injury among abusive and accidental injuries in children

Background: Child physical abuse causes significant morbidity. Previous research has established key physical examination and radiographic findings that are associated with abusive injury. A gap in the literature, however, is the role of additional injury circumstances and details, such as the context in which the injury occurred, in distinguishing abusive from accidental injuries.

Hypothesis: Are there historical elements of the context in which a child’s injury occurs that are more likely associated with either abusive or accidental mechanisms?

Methods: We performed a retrospective review of the medical records for 303 children evaluated by the child abuse consultation team at Yale-New Haven Hospital for head injuries or fractures that occurred between June 1, 2008 to June 30, 2017. We extracted information from the medical record regarding the context of injury: the elements related to the circumstances of the injury itself and the events leading to medical presentation for evaluation of the injury. These included: who was present at the time of the injury; who else may have witnessed or heard the injury or saw the child following the injury; and was there a delay in seeking medical care. Chi-square and Fisher’s exact tests were used to evaluate differences between abusive and accidental injuries, which were determined by an expert physician panel blinded to the context of injury for each case.

Results: There were 303 patients: 201 had a head injury, including either skull fractures or intracranial bleeds, and 102 were cases of non-skull fractures. Eighty patients were determined to have abusive mechanisms of injury, and 168 were found to accidental mechanisms. In 54 cases, the expert panel came to an indeterminate rating as to the mechanism of injury. Of the patients, 73% were younger than 12 months old. There were no differences between those children with abusive, accidental or indeterminate ratings with respect to age, race, ethnicity or gender (Table 1).

In children with accidents, the mother was more likely to be present at the injury event t(64.8% vs 21.5%, p=.001), and a second person was more likely to observe the event (55.9% vs. 27.5%, p <.001), hear the event (43.1% vs. 8.8%; p<.0001), or see the child immediately after the injury (75.4% vs. 47.5%; p<.001). Conversely, in children with abusive injuries, a higher percentage had a delay in seeking care >24 hours (86.3%. vs. 51.6%, p<0.001), the presence of the father (61.3% vs. 38.4%, p=.005), or multiple caregivers in the home at the time of the injury (65% vs. 43%, p=.003 ).

Conclusion: There are certain historical elements of the context in which a young child’s injury occurs that may be helpful in distinguishing abusive from accidental injuries.