Objectives: This study aimed to develop a performance assessment tool for the history-taking components of the medical evaluation of physical abuse in young children by (1) determining the consensus-based injury history and social components for documentation, (2) identifying preliminary performance standards, (3) assessing current level-specific performance using the created tools, and (4) evaluating reliability and validity of the created tools.
Methods: The Physical Abuse Assessment Tool (PHAAT) was developed in 2 steps: (1) a modified Delphi survey was used to identify the injury history and social components for documentation in a medical evaluation for physical abuse, and (2) level-specific ("novice," "competent," "expert") practice standards (minimum passing scores) were created using the identified components via the Angoff method. To evaluate validity, reliability, and level-specific performance of the PHAAT, a chart review of 50 consecutive cases from each of the 3 levels was performed.
Results: Seventy-one child abuse pediatricians and 39 social workers participated in the modified Delphi survey, and 67 child abuse pediatricians and 27 social workers participated in the Angoff method. The resulting PHAAT included 2 checklists for use based on presence or absence of a history of an injurious event. One-way analysis of variance shows significant differences in performance based on team level (P < 0.001), indicating construct validity. Intrarater and interrater reliability evaluations showed strong (rs = 0.64-0.92) and moderate to strong (intraclass correlation coefficient = 0.81-0.98) correlations, respectively.
Conclusions: Initial evaluation suggests the PHAAT may be a reliable and valid practice assessment tool for the medical evaluation of physical abuse.