Legionnaires' disease in France: sensitivity of the mandatory notification has improved over the last decade

• Published in: Epidemiology and infection Vol. 141; no. 12; pp. 2644 - 2649
• Main Authors: CAMPESE, C., JARRAUD, S., SOMMEN, C., MAINE, C., CHE, D.
• Format: Journal Article
• Language: English
• Published: Cambridge, UK Cambridge University Press 01.12.2013; Cambridge University Press (CUP)
• Subjects: Antigens; Bacterial diseases; Bacterial diseases of the respiratory system; Biological and medical sciences; Disease Notification; Disease Notification - methods; Disease Notification - standards; Epidemiological Monitoring; Female; France; France - epidemiology; Fundamental and applied biological sciences. Psychology; Health Services Research; Hospitalization; Hospitals; Human bacterial diseases; Humans; Infectious diseases; Laboratories; Legionnaires disease; Legionnaires' Disease - epidemiology; Life Sciences; Male; Medical sciences; Microbiology; Original Papers; Pneumonia; Surveillance; France; Europe; ANE, France; Legionnaires' disease; public health surveillance; Epidemiology; surveillance system; Infection; Lung disease; Pneumonia; Sensitivity; Respiratory disease; Bacteriosis; Legionellosis; Legionnaires disease
• ISSN: 0950-2688; 1469-4409
• DOI: 10.1017/S0950268813000502

The notification rate of Legionnaires' disease (LD) in France was 2·4/100 000 population in 2010, varying across regions with an increasing rate from west to east. Two sources [mandatory notifications (MN) and a survey of hospital laboratories] were used in a capture–recapture study to estimate the number of LD cases and the sensitivity of the MN system at national and regional levels in 2010. The number of missed cases was estimated using Chapman's method. The estimated sensitivity of MN was 88·5% (95% CI 88·0–89·0) and ranged from 70% to 100% by region. The estimated incidence was 2·7/100 000 population. Sensitivity of the MN system improved since the previous capture–recapture estimates (10% in 1995, 33% in 1998). This study confirmed that the observed west–east gradient is not related to regional notification disparities. Ecological studies should be conducted to better understand the observed spatial variations in LD incidence.