Identification of delayed diagnosis of paediatric appendicitis in administrative data: a multicentre retrospective...

Area of Science:

• Pediatric Emergency Medicine
• Health Informatics
• Diagnostic Accuracy Studies

Background:

• Delayed diagnosis of appendicitis can lead to severe complications.
• Accurate identification of delayed diagnosis in administrative data is challenging.
• A validated tool is needed to retrospectively identify such cases.

Purpose of the Study:

• To derive and validate a tool for accurately identifying delayed diagnosis of appendicitis in administrative data.
• To develop a prediction rule using variables available in administrative datasets.
• To assess the tool's performance in a pediatric population.

Main Methods:

• A cross-sectional study involving 669 pediatric patients across five emergency departments.
• Patients with two emergency department encounters within 7 days, with appendicitis diagnosed at the second, were included.
• Logistic regression was used to derive a prediction rule, which was then validated on a separate cohort.

Main Results:

• The derived tool demonstrated high accuracy, with an area under the curve (AUC) of 0.892 in the derivation group and 0.859 in the validation group.
• At a maximal accuracy threshold, the tool achieved a positive predictive value (PPV) of 84.7% and identified 87.3% of delayed cases.
• A stricter threshold yielded a PPV of 94.9%, identifying 46.8% of delayed cases.

Conclusions:

• The developed tool accurately identifies delayed diagnosis of appendicitis in children using administrative data.
• This tool can be utilized for screening potential missed appendicitis diagnoses.
• It offers a method to identify cohorts of children experiencing delayed appendicitis diagnosis for research or quality improvement.