A Standardized Scoring System to Predict Unnecessary Transfers to Children's Hospitals for Appendicitis Evaluation

Background: Acute appendicitis is common in the pediatric population and results in frequent inter-hospital transfers. Transferred patients do not have more advanced disease or higher complication rates but undergo more diagnostic testing, incurring higher costs. No scoring systems exist to predict which patients do not require transfer to a dedicated children's hospital for appendicitis evaluation. Methods: A retrospective review was performed of patients ages 5 to 17 years transferred for appendicitis evaluation over a 5-year period. Data included demographics, presenting symptoms, vitals, and laboratory results. Multiple logistic regression analysis was developed on a derivation set (n = 1099) and tested on a sub validation set (n = 215). Odds ratios were used to assign weights. Area under the curve analysis was used to assess accuracy. Results: A total of 1354 patients were transferred; 439 (33.4%) patients were diagnosed with appendicitis. Age ≥10 years (2 points), male gender (2 points), Hispanic ethnicity (1 point), nausea (1 points), vomiting (1 point), absolute neutrophil count ≥7.5 (4 points), and lymphocyte percent ≤20 (1 point) were associated with appendicitis [12 points total]. A score of ≤5 points predicted unnecessary transfer (P < 0.001) with AUC values of 0.802 and 0.772 in the derivation and validation sets, respectively. Sensitivity of 90% and negative predictive value of 90.9% were achieved with score ≤5 points. Discussion: The majority of transferred patients did not have appendicitis. Our scoring system provides a screening tool to reduce unnecessary transfers and inappropriate resource utilization.