V.M. Ringheanu1, J.J. Strubel1, M. Anapolsky1, K.M. Sutyak1, N.B. Hebballi1, J.M. Joly1, M. Broussard1, T. Fisher1, K.P. Lally1, K. Tsao1 1McGovern Medical School at UTHealth Houston, Department Of Pediatric Surgery, Houston, TX, USA
Introduction: Conversion of the electronic medical record (EMR) by healthcare systems often brings clinical benefits but also transitional challenges which may create unanticipated harm. In our tertiary children’s hospital, we previously used the widely accepted, EMR-integrated Pediatric Appendicitis Score (PAS) which guided patient disposition and need for additional imaging. Upon EMR transition, PAS will be replaced by the Pediatric Appendicitis Risk Calculator (pARC) with unknown validation. This study aims to investigate the level of agreement between the PAS and pARC scoring systems and identify incongruencies where one may be favored.
Methods: A retrospective review was conducted of patients aged 3-17 years who underwent appendectomy for pathologically confirmed appendicitis at our tertiary children’s hospital from May 2021 to March 2024. Demographics, patient characteristics, lab values, imaging results, and operative details were collected. PAS and pARC scores were retrospectively calculated and assigned to low, moderate, or high-risk categories based on the recommended guidelines. Descriptive statistics and Chi-square were used to compare risk category frequencies between the two scores. Cohen's kappa statistic (κ) was used to assess agreement between PAS and pARC, evaluating differences based on age, sex, and WBC count.
Results: Among 1,182 children, the median age was 11 years (IQR 8-14) with 63% male, 25% white, 60% Hispanic, and 98% with no significant past medical history. Median PAS and pARC scores were 8 (IQR 6-9) and 70% (IQR 44-84%), respectively. PAS categorized 32 (3%) patients as low-risk, 266 (23%) as moderate, and 884 (75%) as high while pARC categorized 182 (15%) patients as low-risk, 269 (23%) as moderate, and 731 (62%) as high (p<0.001). Overall, there was fair agreement between PAS and pARC (κ=0.25) (Figure 1a). However, agreement varied by age and sex: fair for <12 years (κ=0.30) vs. slight for ≥12 years (κ=0.19); fair for males (κ=0.35) vs. slight for females (κ=0.11) (Figure 1b). Considering both age and sex, females ≥12 years had the lowest agreement (κ=0.01) (Figure 1c). Those with WBC>10,000 x 109 cells/L had fair agreement (κ=0.26) vs. slight for those with WBC<10,000 x 109 cells/L (κ=0.02) (Figure 1d).
Conclusion: Although fair agreement existed between systems for pathology-proven appendicitis, pARC categorized significantly fewer patients as high-risk. Clinically challenging cohorts, such as adolescent females and those with normal WBC count, had almost no agreement, with pARC underestimating compared to PAS. Careful scrutiny is needed in all aspects of EMR transitions, especially those with direct patient impact.