Monitoring measles outbreaks using emergency department data in France. (13th November 2019)

• Record Type: Journal Article
• Authors: Noel, G; Franke, F; Forgeot, C; Piarroux, J; Viudes, G; Caserio-Schönemann, C
• Is Part Of: European journal of public health. Volume 29(2019)Supplement 4
• Publication Date: 2019-11-13
• Subjects: Epidemiology -- Europe -- Periodicals; Public health -- Europe -- Periodicals; 362.109405
• DOI: 10.1093/eurpub/ckz187.072
• Languages: English
• ISSNs: 1101-1262

Abstract

Abstract: Background: Measles' outbreaks still occur in industrialised countries and reactive monitoring of the disease is needed. In France, measles' monitoring is based on a mandatory notification (MN). Methods: In France, for each emergency department (ED) visit, an Electronic ED Abstract (EEDA) is transmitted to the OSCOUR® network included in national Syndromic Surveillance System. EEDA include the diagnosis at discharge using International Classification of Disease (ICD-10). This study aims to evaluate the correlation between the number of measles MN and the number of ED visits related to measles and to calculate the positive predictive value (PPV) of diagnosis at discharge of measles in ED. The study period covered the third wave of the 2008-2011 French outbreak. From 2010 to 2011, we analyzed EEDA transmitted from 32/53 ED of the PACA region and regional measles MN. Using weekly aggregated data, the correlation between the number of measles cases given by both MN and ED sources was assessed (Pearson coefficient). In the 4 ED (2 pediatric) of the teaching hospital of Marseille, electronic medical files coded as measles were reviewed to evaluate the PPV of the diagnosis coded according to the national MN case definition (gold standard). Results: 2, 949 MN were transmitted and 1, 306/1.3 million visits in ED were coded as measles. The peak of the outbreak was reported the same week by EEDA and MN. The correlation between EEDA and MN was strong (0.91; p < 0.001). Hospitalization rate was 20.7% and 21.0% according to EEDA and MN (p = 0.81). Among 363 medical files coded as measles in the 4 ED, 14 cases (3.9%) did not fulfilled the national case definition (PPV: 96.1 [IC95%: 94.2-98.1]), 86 (23.7%) fulfilled criteria for "clinical case" and 263 (72.5%) for "confirmed case". Conclusions: Most of the time, ED visits coded as measles fulfilled diagnosis criteria. Our results suggested that EEDA (better reactivity) could be complementary to MN (better completeness) to monitor measles. Key messages: Most of the time, ED visits coded as measles fulfilled diagnosis criteria: PPV of coded medical diagnosis in ED for measles is high. Reactivity of ED data monitoring could usefully complete mandatory notification for measles surveillance.