Child maltreatment in U.S. emergency departments: Imaging and admissions.

Objective: We report imaging and admission ratios for children with definitive and suggestive maltreatment in a national sample of emergency departments (EDs).

Methods: Using the 2012 Nationwide Emergency Department Sample (NEDS), we generated national estimates of ED visits for children <10 years with both definitive and suggestive maltreatment. Outcomes were admission/transfer ratios for children <10years and screening ratios by skeletal surveys and head computed tomography (CT) for children <2 years with suspected physical abuse. We compared hospitals with low, medium, and high pediatric ED volumes using multivariable logistic regression.

Results: The 2012 national estimate of U.S. ED visits (children <10years) with definitive maltreatment is 14,457 (95% CI: 11,987-16,928). Suggestive child maltreatment was seen in an additional 103,392 (95% CI: 90,803-115,981) pediatric ED visits. After controlling for patient case mix, high volume hospitals had a significantly higher adjusted odds ratio (AOR) of admission/transfer among definitive cases (AOR=1.74, 95% CI: 1.08-2.81), and medium volume hospitals had a higher odds of admission/transfer among suggestive cases (AOR=1.24, 95% CI: 1.02-1.50) when compared with low volume hospitals. In hospitals with reliable reporting of imaging procedures, high volume hospitals reported skeletal surveys (age <2 years) significantly more often than low volume hospitals, AOR=3.32 (95% CI: 1.25-8.84); the AORs for head CT did not differ by hospital volume.

Conclusions: Low volume hospitals were less likely to screen by skeletal survey, but head CT ratios were not affected by ED volume. Low volume hospitals were also less likely to admit or transfer.