Skip to Main content Skip to Navigation
Master Thesis

Étude de la discordance des diagnostics infectieux entre les urgences et les services de médecine interne au CHUGA : recherche de facteurs prédictifs et évaluation d’impact

Abstract : Introduction: Hospitalizations for infections in internal medicine through emergency departments represent a significant part of the total amount of hospitalizations. Emergency physicians make an initial diagnosis that influences the progess of the treatment after the admission. Discrepancies with the internal medicine’s diagnosis are described in literature, with potential medical and economic consequences. The main objective of this study, which focuses on patients with a preliminary or definitive diagnosis of infectious origin, is to assess the rate of diagnostic discordances between the diagnosis hinted at the emergency department and the diagnosis retained in the internal medicine department. The secondary objectives are to identifying the factors associated with these diagnostic discrepancies and to measure its consequences. Material and Method: We carried out a monocentric retrospective study on hospitalized patients in internal medicine at the Grenoble Alpes University Hospital, via an emergency department, from March, 22nd 2021 to June, 22nd 2021 included, with a suspected diagnosis of infection in the emergency department and/or retained in internal medicine. We determined whether there was a difference between the type of infection reported at the exit of the emergency department and the type of infection at the exit of the internal medicine department, defining the diagnostic discrepancy. The intrinsic characteristics of the patients, related to the crossing to the emergencies and the hospital characteristics of morbidity-mortality were collected in order to study its association with the diagnostic discrepancy. Results: Over the period of inclusion, 559 patients were hospitalized in internal medicine. Among these patients, 280 came from emergencies and had an infectious diagnosis held in emergencies and/or internal medicine. They were mainly men (59.3%), with a median age of 80, a Charlson score above 6 and they were taking 7 medication a day. COVID infection was the post-hospital diagnostic of the internal medicine’s service in less than one-third of the patients included. A diagnostic discordance rate was found in 65 of 280 patients, that is 23.2%. Among the most frequent diagnostics made in the emergencies, it included labelled diagnostics of non-infectious origin (26 patients), respiratory infections (14 patients) and urinary tract infections (4 patients). The intrinsic characteristics of the patients that can predict diagnostic discrepancy include a higher age (p-value 0.01) and a higher Charlson score (p-value 0.034). Among the characteristics from the visit to the emergencies, there is a longer transition time to emergencies (p-value 0.016). The conclusion of COVID infection at the end of internal medicine hospitalization was significantly associated with the lack of diagnostic discordance (p-value < 0.001). Regarding the study of the impact, we highlight a statistically significant association with the cessation of antibiotic therapy on arrival in the service (p-value < 0.0001). Conclusion: Diagnostic discordance is a public health problem that affects all services of the hospital but also ambulatory medicine. The integration of current knowledge into our practice, as well as the proliferation of robust studies on the prevalance and on the search for causal factors, are necessary for the development of interventional studies in the future.
Document type :
Master Thesis
Complete list of metadata
Contributor : Jean-Hugues Morneau Connect in order to contact the contributor
Submitted on : Monday, October 25, 2021 - 2:14:59 PM
Last modification on : Wednesday, October 27, 2021 - 3:37:46 AM


  • HAL Id : dumas-03385001, version 1


Marie Vangout. Étude de la discordance des diagnostics infectieux entre les urgences et les services de médecine interne au CHUGA : recherche de facteurs prédictifs et évaluation d’impact. Médecine humaine et pathologie. 2021. ⟨dumas-03385001⟩



Record views


Files downloads