Development and validation of a scoring system to predict esophagogastroduodenoscopy necessity

Area of Science:

• Gastroenterology
• Medical Diagnostics
• Clinical Decision Support

Background:

• Esophagogastroduodenoscopy (EGD) is a common diagnostic procedure.
• Accurate prediction of EGD necessity is crucial to optimize resource allocation and patient care.
• Current methods for determining EGD indication may lack precision, leading to potential misapplications.

Purpose of the Study:

• To develop and validate a novel scoring system for predicting the need for esophagogastroduodenoscopy (EGD).
• To enhance the accuracy of EGD indication and minimize unnecessary procedures in clinical practice.
• To compare the performance of the developed scoring system against physician judgment.

Main Methods:

• A scoring system was developed using data from 865 outpatients scheduled for EGD.
• Patients provided clinical symptoms, medical history, and endoscopic findings.
• A parallel test algorithm was selected, and the system was validated on a separate cohort, with performance compared to six physicians using sensitivity and negative likelihood ratio (-LR).

Main Results:

• The final scoring system included 21 variables (age, symptoms, medical history).
• The system achieved a sensitivity of 79.0% (training) and 83.9% (validation) with -LR of 0.627 and 0.615, respectively.
• The scoring system demonstrated superior sensitivity (84.0% vs. 68.7%) compared to physicians.

Conclusions:

• A user-friendly and effective scoring system for predicting EGD necessity was successfully developed and validated.
• The developed system shows promise in improving the accuracy of EGD indication in clinical settings.
• This tool can aid clinicians in making more informed decisions regarding EGD referrals.