How useful was a paediatric physical abuse screening project in a rural Australian Emergency Department?

Abstract

Children with non-accidental injuries have increased risk of future death. There is insufficient evidence for widespread physical abuse screening tool use in the Emergency Department (ED). This study assesses the utility of a physical abuse project that includes the implementation of a screening tool with case-matching from multiple sources. It aims to confirm whether risk-screening in a medium-sized rural Australian ED is reliable and will improve outcomes. Method: 16-month ED retrospective pre/post-implementation study of all injury, burn or poisoning cases 16 years and under. Presentations with potential physical abuse were filtered by ICD10 codes and reviewed. Multivariable logistic regression models compared pre and post implementation cases. Analyses examined outcomes, trends and interrogated the screening tool which formed a Clinical Pathway Algorithm (CPA). Results: 1,469 presentations underwent investigation. 747 pre-implementation and 722 post, Pearson’s chi-square test showed statistically insignificant differences. If tool used, documentation improved (OR 7.73; 95% CI 4.91 to 12.18), child protection service (CPS) referrals increased (OR 5.50; 95% CI 1.82 to 16.61) and hospital admissions decreased (OR 0.42; 95% CI 0.22 to 0.79). Re-presentation rates stayed the same. Increased physical abuse was associated with screening factors including carer behavioural concerns, inadequate supervision, delayed presentation, repeat and unexplained injuries (ORs/Cis in an accompanying figure). Screening tool sensitivity was 62.3%, specificity 79.7%. Conclusion: Implementing this ED paediatric physical abuse project improved safety behaviours and best-practice documentation. The tool improved medical decision making without increased representations. ED clinicians may use similar CPAs to help review safety concerns and facilitate discharge, however resources are needed to investigate referrals flagged due to false positive rates.