Arthur Lauriot Dit Prevost, Guillaume Clement, Xavier Lenne, Dyuti Sharma, Diane Carriere, Jad Atallah, Kevin Le Duc, Laurent Storme, Amélie Bruandet, Sebastien Mur
{"title":"Contribution of nationwide medico administrative data on congenital anomalies in France: Validation study on congenital diaphragmatic hernia.","authors":"Arthur Lauriot Dit Prevost, Guillaume Clement, Xavier Lenne, Dyuti Sharma, Diane Carriere, Jad Atallah, Kevin Le Duc, Laurent Storme, Amélie Bruandet, Sebastien Mur","doi":"10.1016/j.jeph.2025.203109","DOIUrl":null,"url":null,"abstract":"<p><strong>Introduction: </strong>In France, medico-administrative data are collected through the Programme de Médicalisation des Systèmes d'Information (PMSI), which includes all discharge reports from public and private hospitals, covering more than 20 million hospital stays per year. In parallel, databases on rare diseases are a national priority but remain difficult to implement and maintain. We aimed to assess the validity of PMSI data to identify patients with congenital rare diseases, using congenital diaphragmatic hernia (CDH) as an example. We also aimed to explore the national epidemiological estimates provided by PMSI.</p><p><strong>Methods: </strong>Using PMSI, we identified CDH patients treated in the neonatal intensive care units (NICU) of two French referral centers between 2010 and 2018 based on ICD-10 diagnostic codes. These centers have maintained a prospective database of CDH patients (CDHdb) for over 10 years. We compared PMSI data with CDHdb to assess the accuracy of PMSI in detecting CDH cases. A nationwide extraction was then performed to estimate the incidence and mortality of CDH in France.</p><p><strong>Results: </strong>PMSI identified all 248 CDH patients listed in the CDHdb. However, 37 additional cases were found in PMSI but not in the CDHdb, including 24 neonatal deaths with confirmed CDH. Encoding errors were found in four cases (omphalocele, diaphragmatic eventration), and eight CDH cases were missing from the CDHdb. The sensitivity and specificity of PMSI were estimated at 100 % (98.5 %; 100 %) and 99.9 % (99.9 %; 100 %), respectively. At the national level, 1680 cases were identified, corresponding to an incidence of 2.33 per 10,000 live births, a male-to-female ratio of 1.40, and a 1-year mortality rate of 30.5 % (28.2 %; 32.9 %).</p><p><strong>Discussion: </strong>PMSI appears to be a valid source for identifying CDH patients at birth, with national epidemiological estimates consistent with the literature. However, medico-administrative data require careful validation before being used for rare disease surveillance.</p><p><strong>Conclusion: </strong>Our findings suggest that PMSI could be a valuable tool for studying and monitoring CDH patients-and potentially other congenital anomalies-on a nationwide scale.</p>","PeriodicalId":517428,"journal":{"name":"Journal of epidemiology and population health","volume":"73 3","pages":"203109"},"PeriodicalIF":0.0000,"publicationDate":"2025-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of epidemiology and population health","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1016/j.jeph.2025.203109","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Introduction: In France, medico-administrative data are collected through the Programme de Médicalisation des Systèmes d'Information (PMSI), which includes all discharge reports from public and private hospitals, covering more than 20 million hospital stays per year. In parallel, databases on rare diseases are a national priority but remain difficult to implement and maintain. We aimed to assess the validity of PMSI data to identify patients with congenital rare diseases, using congenital diaphragmatic hernia (CDH) as an example. We also aimed to explore the national epidemiological estimates provided by PMSI.
Methods: Using PMSI, we identified CDH patients treated in the neonatal intensive care units (NICU) of two French referral centers between 2010 and 2018 based on ICD-10 diagnostic codes. These centers have maintained a prospective database of CDH patients (CDHdb) for over 10 years. We compared PMSI data with CDHdb to assess the accuracy of PMSI in detecting CDH cases. A nationwide extraction was then performed to estimate the incidence and mortality of CDH in France.
Results: PMSI identified all 248 CDH patients listed in the CDHdb. However, 37 additional cases were found in PMSI but not in the CDHdb, including 24 neonatal deaths with confirmed CDH. Encoding errors were found in four cases (omphalocele, diaphragmatic eventration), and eight CDH cases were missing from the CDHdb. The sensitivity and specificity of PMSI were estimated at 100 % (98.5 %; 100 %) and 99.9 % (99.9 %; 100 %), respectively. At the national level, 1680 cases were identified, corresponding to an incidence of 2.33 per 10,000 live births, a male-to-female ratio of 1.40, and a 1-year mortality rate of 30.5 % (28.2 %; 32.9 %).
Discussion: PMSI appears to be a valid source for identifying CDH patients at birth, with national epidemiological estimates consistent with the literature. However, medico-administrative data require careful validation before being used for rare disease surveillance.
Conclusion: Our findings suggest that PMSI could be a valuable tool for studying and monitoring CDH patients-and potentially other congenital anomalies-on a nationwide scale.