(07) Validation and algorithm for children intussusception identification using electronic health records of Ningbo city in China

Background: Intussusception is the second most common cause of an acute abdomen in children, and the most common cause of small bowel obstruction in young infants. In some studies, the rotavirus vaccination has been reported to be associated with an increased risk of intussusception. Therefore, monitoring the risk of intussusception after the introduction of the rotavirus vaccines was recommended by WHO. The electronic health record (EHR) has been utilized commonly in post-market monitoring study. Accuracy of key variables is significant for the data reliability of EHR, and investigators should ensure that the selected outcome variables are validated, in order to maximize study internal validity, and minimize the misclassification bias. Previously, a few studies have confirmed the validity of intussusception through chart review using EHR, but not in China. The objective was to verify the diagnosis and determine the algorithm with best performance for identification of intussusception using Chinese EHR databases, in order to pave the way for safety monitoring of rotavirus vaccine in China.

Methods: Regional Health Information Platform (RHIP) in Ningbo has almost covered the whole permanent children aged below 6 years, and collected nearly all healthcare information of residents from birth to death. Based on RHIP in Ningbo, patients aged 0-72 months from 2015 to 2021 with any related visits of intussusception were included. Potential intussusception visits were identified according to the International Classification of Diseases (ICD), as K56.1 in the Tenth Revision, or the Chinese Term as “Chang Tao”. The validation process of intussusception cases was in line with the Brighton guideline. The cases that meet the definition standard of level one in Brighton criteria were judged as cases with Brighton level one, while the cases that did not meet the definition standard of level one in Brighton criteria were judged as non-cases of intussusception. The algorithms were consisted of diagnostic codes, or keywords in different clinical scenarios, and their performance were evaluated with positive predictive value (PPV) and sensitivity in line with the Brighton guideline.

Results: A total of 2958 patients with 3246 episodes were confirmed as Brighton level one. Algorithm including K56.1 or “Chang Tao” in any diagnose terms and combined with records of enema examinations or records of related surgeries had the highest sensitivity (90.63%, 95% CI 89.58, 91.62%) and high PPV (80.14%, 95% CI 78.81, 81.42%), while the highest PPV (87.35%, 95% CI 86.08, 88.56%) and high sensitivity (76.40%, 95% CI 74.90, 77.85%) were observed when further requiring any value of the longitudinal cutting length and transverse cutting diameter of concentric circles to exceed 30 mm in the results of imaging examinations.

Conclusion: The intussusception can be well identified and validated internally in EHRs of Ningbo. The electronic algorithm of any diagnose terms of intussusception combined with air enema or related surgery could be alternative method to capture the occurrence of intussusception in RHIP in Ningbo.

Keywords: algorithms; China; electronic health records; intussusception; post-market monitoring; verification study.