Email updates

Keep up to date with the latest news and content from BMC Proceedings and BioMed Central.

This article is part of the supplement: International Conference on Prevention & Infection Control (ICPIC 2011)

Open Access Oral presentation

The concordance of European and US definitions for healthcare-associated infections (HAI)

S Hansen1*, D Sohr1, C Geffers1, P Astagneau2, A Blacky3, W Koller3, I Morales4, ML Moro5, M Palomar6, E Szilagyi7, C Suetens8 and P Gastmeier1

  • * Corresponding author: S Hansen

Author Affiliations

1 Charité - University Medicine Berlin, Berlin, Germany

2 Université Pierre & Marie Curie, Paris, France

3 Medical University Vienna, Vienna, Austria

4 Scientific Institute of Public Health, Brussels, Belgium

5 Agenzia Sanitaria e Sociale Regione Emilia Romagna, Bologna, Italy

6 Hospital Vall d'Hebron, Barcelona, Spain

7 National Centre for Epidemiology, Budapest, Hungary

8 ECDC, Stockholm, Sweden

For all author emails, please log on.

BMC Proceedings 2011, 5(Suppl 6):O2  doi:10.1186/1753-6561-5-S6-O2

The electronic version of this article is the complete one and can be found online at:

Published:29 June 2011

© 2011 Hansen et al; licensee BioMed Central Ltd.

This is an open access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Introduction / objectives

In Europe comparison of infection rates of HAI is restricted since some countries are using CDC definitions while others use HELICS (Hospitals in Europe Link for Infection Control through Surveillance) definitions. As part of the harmonization process of surveillance, ECDC outsourced a study to analyze the concordance between the definitions.


A group with experts from 7 European countries was set up to realize the study. Agreement for bloodstream infection (BSI) and pneumonia (PN) was estimated by Cohens kappa.


The study was performed on 47 ICUs and 6506 patients, 180 PN and 123 BSI cases. Agreement for PN was k= 0.99 (CI95 0.98;1.00). When PN cases were divided in clinically and microbiologically defined PN, kappa values were 0.90 (CI95 0.86;0.94) and 0.72 (CI95 0.63;0.82) respectively. Diagnosis of PN varied among countries: 4 countries predominantly surveyed microbiologically defined PN whereas the others recorded mainly clinically defined PN. Agreement for BSI was k= 0.73 (CI95 0.66;0.80), BSI cases secondary to another infection site (42% of all BSI) were missed by CDC definitions. BSI concordance was perfect (k= 1.00) when only primary BSI cases (HELICS BSI with origin “catheter” or “unknown” and CDC BSI) were analyzed.


Although other methodological differences exist between the two protocols, case definitions per se do not compromise comparability of results and should not be an obstacle for harmonization of European surveillance.

Disclosure of interest

None declared.