The objective of this study was to identify opportunities to improve identification and evaluation of child abuse by (1) describing rates of initial and recurrent injury in young infants, (2) describing the frequency of a history of prior injury in young children evaluated by a child abuse pediatrician, and (3) identifying practice variation, including use of skeletal surveys as an example, in the identification and evaluation of injury across a network of 6 children's hospitals.Methods: Six children's hospitals identified infants with an initial injury and recurrent injury over a 1-year period using 2 methods: (1) diagnostic code methodinfants 6 months or younger presenting with at least 1 diagnostic code for injury were tracked for 12 months to determine the frequency of recurrent injury, and (2) consult methodall available medical records of children 18 months or younger seen for an inpatient consultation for suspected child abuse were reviewed to identify history of a first injury at 6 months or younger.Results: Using the diagnostic code method, 682 unique infants were identified with initial injuries, most commonly fractures (37.0%), bruising/ ecchymosis (35.9%), and superficial injuries (28.3%). Forty-two infants (6.2%) returned with a second injury, and no demographic factors were significantly associated with the likelihood of a second injury. Using the consult method, 37 of 342 consults (10.8%) were identified as having a history of at least 1 initial injury. Of the initial injuries identified, the most common was bruising/ecchymosis (64.9%). The number of injuries identified with either method varied significantly across hospitals, as did completion of skeletal surveys for infants with bruising (range, 4.5%-71.1%; P < 0.001) and any injury (range, 4.4%-62.7%; P < 0.001).
Conclusions:Our study demonstrates that young infants who experience 1 injury often experience a second injury. There exists significant variability in the identification of injury and the completion of skeletal surveys across a network of 6 children's hospitals. A standardized quality improvement approach may improve identification of injury and reduce the variability in practice observed.