Detection Algorithms for Gastrointestinal Perforation Cases in the Medical Information Database Network (MID-NET®) in Japan

Abstract

BackgroundThe Medical Information Database Network (MID-NET®) in Japan is a vast repository providing an essential pharmacovigilance tool.Gastrointestinal perforation (GIP) is a critical adverse drug event, yet no well-established GIP identi cation algorithm exists in MID-NET®. MethodsThis study evaluated 12 identi cation algorithms by combining ICD-10 codes with GIP therapeutic procedures. Two sites contributed 200 inpatients with GIP-suggestive ICD-10 codes (100 inpatients each), while a third site contributed 165 inpatients with GIP-suggestive ICD-10 codes and antimicrobial prescriptions. The positive predictive values (PPVs) of the algorithms were determined, and the relative sensitivity (rSn) among the 165 inpatients at the third institution was evaluated. ResultsA trade-off between PPV and rSn was observed. For instance, ICD-10 code-based de nitions yielded PPVs of 59.5%, whereas ICD-10 codes with CT scan and antimicrobial information gave PPVs of 56.0% and an rSn of 97.0%, and ICD-10 codes with CT scan and antimicrobial information as well as three types of operation codes produced PPVs of 84.2% and an rSn of 24.2%. The same algorithms produced statistically signi cant differences in PPVs among the three institutions. Combining diagnostic and procedure codes improved the PPVs. The algorithm combining ICD-10 codes with CT scan and antimicrobial information and 80 different operation codes offered the optimal balance (PPV: 61.6%, rSn: 92.4%). ConclusionThis study developed valuable GIP identi cation algorithms for MID-NET , revealing the trade-offs between accuracy and sensitivity. The algorithm with the most reasonable balance was determined. These ndings enhance pharmacovigilance efforts and facilitate further research to optimize adverse event detection algorithms. the research ndings increase the probability of incorrect decision-making. Yamana et al. have reported that the number of validation studies in Japan has increased in recent years, driven by the growing need for data quality assurance in post-marketing surveillance and regulatory decisionmaking using RWD. 7 Tanigawa et al. reported that the validity of the identi cation algorithm for inpatients with gastrointestinal perforation (GIP), which consisted of diagnosis codes extracted solely from MID-NET®, had a low positive predictive value (PPV) of approximately 48%. 8 Furthermore, this result was acquired at a single site from among the 10 cooperating medical institutions of MID-NET®; consequently, a detection method for patients with GIP in MID-NET® is not yet available. Therefore, we consider it necessary to develop additional algorithms for the accurate detection of patients with GIP so as to obtain reliable results from outcome studies related to GIP using MID-NET®. GIP can cause life-threatening peritonitis or septicemia if left untreated, 9 and is a serious side-effect induced by nonsteroidal anti-in ammatory drugs and monoclonal antibodies. 10 Accordingly, it is important to develop identi cation algorithms for drug-induced GIP using MID-NE...