Introduction: Designing a tool that can differentiate those at risk of child abuse with great diagnostic accuracy is of great interest. The present study was designed to evaluate the diagnostic accuracy of Escape instrument in triage of at risk cases of child abuse presenting to emergency department (ED).
Method: The present diagnostic accuracy study performed on 6120 of the children under 16 years old presented to ED during 3 years, using convenience sampling. Confirmation by the child abuse team (pediatrician, a social worker, and a forensic physician) was considered as the gold standard. Screening performance characteristics of Escape were calculated using STATA 21.
Results: 6120 children with the mean age of 2.19 ± 1.12 years were screened (52.7% girls). 137 children were suspected victims of child abuse. Based on child abuse team opinion, 35 (0.5%) children were confirmed victims of child abuse. Sensitivity, specificity, positive and negative likelihood ratio and positive and negative predictive values of this test with 95% CI were 100 (87.6 - 100), 98.3 (97.9 - 98.6), 25.5 (18.6 - 33.8), 100 (99.9 - 100), 0.34 (0.25 - 0.46), and 0 (0 - NAN), respectively. Area under the ROC curve was 99.2 (98.9 - 99.4).
Conclusion: It seems that Escape is a suitable screening instrument for detection of at risk cases of child abuse presenting to ED. Based on the results of the present study, the accuracy of this screening tool is 99.2%, which is in the excellent range.
Keywords: Child abuse; decision support techniques; diagnosis; emergency service; hospital; risk assessment.