Variability in Occult Injury Screening Among Siblings/Household Contacts of Physical Abuse Victims.

Abstract

The objective of this study is to examine radiologic occult injury screening performance/yield among contacts presenting for precautionary medical assessments and assess factors associated with deferred screening. Data were collected retrospectively from charts of contacts younger than 8 years presenting for precautionary evaluation to a level 1 pediatric emergency department January 1, 2018 to March 31, 2023. Demographics, radiologic performance/yield, physical examination, social work-based psychosocial assessment, reasons for deferred imaging, and diagnostic codes were abstracted. Descriptive statistics and χ 2 analysis are reported. Three hundred ninety contacts were identified; 364 (93.3%) were biological siblings. Most (276, 70.8%) were 2 to 8 years old. Statistically significant relationships were identified with age, insurance, and hospital social work assessment and screening. Thirty-four infants (54%) underwent neuroimaging; no studies were abnormal. Of 114 contacts, <2 years old, 97 (85%) underwent skeletal survey (SS); 9 (9%) SS were abnormal. Twenty-seven (24%) returned for follow-up SS; 4 (14.8%) were abnormal. For 2 contacts, an abnormal initial SS was refuted by follow-up imaging. Physical examinations were abnormal for 11% of contacts. Reasons for deferred imaging included contact well appearance, caregiver concerns, and clinician disagreement with indications. Encounter International Classification of Diseases codes varied, commonly reflecting nonspecific screening assessments. Despite national clinical practice guidelines, studies of abusive injury prevalence and radiologic yield among at-risk contacts exposed to unsafe environments are few. Screening evaluations inclusive of physical examination and radiologic studies identify abuse concerns among at-risk contacts. Further study of factors impacting radiologic screening decisions is needed. Considerations to advance epidemiologic research include standardized diagnostic coding and prospective assessment of radiologic yield.