Surveillance of influenza through EISS (European Influenza Surveillance Scheme) during the 1999 to 2000 winter shows that influenza affected most of the 11 participating countries and was particularly active in December 1999 and January 2000. Influenza A(H3N2) virus, responsible for one of the three largest epidemics in the past ten years in some countries of the EISS area, circulated widely. Widespread activity was observed in Belgium, the Czech Republic, France, Germany, Italy, the Netherlands, Spain, and Switzerland. The absence of a common case definition continues to hinder data comparison, and one of the key priorities for developing EISS is now to standardise the main indicators collected and the epidemic thresholds.

Outbreaks of influenza may arise suddenly, disturb social life and working patterns, increase drug consumption, and increase the death rate (1). Early detection of the circulation of influenza virus enables appropriate, timely action to be taken in response to substantial outbreaks of influenza. Many European countries have set up national networks for influenza surveillance since the 1950s. Many of these networks have gradually integrated the collection of clinical morbidity and virological data in general populations of patients consulting general practitioners in order to provide sensitive and specific warnings of disease activity. In the last 15 years, the development and optimisation of the existing surveillance systems has improved their sensitivity and made earlier warning possible. Influenza knows no boundaries, and communication between national networks in European countries has developed through a series of collaborations. The Eurosentinel experience, which lasted from 1987 to 1991 (2), was followed by the ENS-CARE Influenza system, which operated under the aegis of the World Health Organization (WHO) regional office for Europe, funded by the Directorate General (DG) XIII of the European Union (EU) from 1992 to 1995 (3). Subsequently, in 1995, the European Influenza Surveillance Scheme (EISS) was set up. It has been operating in its current form since 1996, supported financially by the Health and Consumer protection Directorate General (DG SANCO /F4) of the European Commission since 1999. The objectives of EISS are to:

• facilitate the rapid exchange of information on influenza activity with the help of sentinel physicians and virological laboratories
• link medical data with virological data from the same population
• provide health authorities with a continuous picture of the clinical impact of influenza in Europe
• contribute to defining vaccine components
• give high quality data based, if possible, on standardised and comparable indicators at a European level.

This article describes the data collected through EISS during the 1999 to 2000 winter, when an influenza A (H3N2) virus, antigenically close to A/Sydney/5/97 and A/Moscow/10/99 variants, circulated widely. The virus was responsible for one of the three largest epidemics in the past ten years in some countries (France, Great Britain, and Italy) of the EISS area.

Sentinel networks for the surveillance of influenza, composed of physicians who collect virological specimens, exist in all countries that contribute to EISS. The catchment population or the number of participating physicians indicate the extent to which the national networks are likely to form representative samples (table). Only some European networks use precise case definitions. The main indicator of disease activity varies from one country to another - the rate of acute respiratory illness per consultation (France and Germany), the rate of influenza-like illness per consultation (Belgium, Switzerland, and Denmark), or the rate of influenza-like illness in a defined population (all other countries). Epidemic thresholds also vary from one national system to another. Nevertheless it is possible to follow the evolution of influenza activity within and between countries between week 40 of one year and week 15 of the next year.

National reference centres in each EISS country are responsible mainly for virological surveillance. Virological data reported result from rapid diagnosis tests (immuno-enzymology or immunofluorescence) and from cell culture with precise identification. Some centres also perform routine reverse transcriptase polymerase chain reaction (RT-PCR) tests.

Data on influenza activity are sent to national coordinators each week during the surveillance season. After analysis by national experts, the data are sent to other participating countries electronically using the internet on the Friday of the same week. EISS data processing has been described in detail elsewhere (4)

During the 1999-2000 season, 11 countries took an active part in EISS: Belgium, the Czech Republic, Denmark, France, Germany, Great Britain, Italy, the Netherlands, Portugal, Spain, and Switzerland. Sweden joined EISS at the beginning of 2000.

Influenza A virus was first detected in swabs obtained by sentinel physicians and reported (figure) in week 40 of 1999 in France, and successively in week 41 in Belgium, Great Britain and Czech Republic, 43 in Germany and Portugal, 45 in Spain and the Netherlands, 46 in Denmark and Switzerland, and finally 47 in Italy. The subtype A(H3N2) predominated in all participating countries; the strains analysed were antigenically linked to the vaccine variant A/Sydney/5/97(H3N2) or to A/Moscow/10/99(H3N2). Influenza A(H1N1) viruses were identified very sporadically except in Spain, where they were first detected in week 45 of 1999 and were isolated mostly in weeks 45/99, and weeks 2, 5, and 9 of 2000. Type B viruses were first detected in Belgium in the last week of 1999, then in Great Britain (02/00), Germany (04/00) and Italy (05/00). Influenza B circulated significantly only in Great Britain, towards the end of the activity period of influenza (weeks 02/00, 04/00, and 05/00). Strains of influenza B that were analysed were antigenically close to the vaccine variant B/Yamanashi/166/99.

The rates of acute respiratory illness or influenza-like illness per consultation or influenza-like illness/100 000 population generally peaked at the same time as the rates of detection of influenza viruses in general (non hospital) practice. Maximum rates were generally recorded between weeks 51/99 and 05/00 with a peak in weeks 01 and 02 of 2000, although some national variations were seen (figure).

The EISS system describes five levels of influenza activity: no activity, sporadic activity, local activity, regional activity, and widespread activity. The maximum levels reached in the 1999-2000 season were defined as follows:

• widespread activity in Belgium, the Czech Republic, France, Germany, Italy, the Netherlands, Spain, and Switzerland
• regional activity in Denmark and Great Britain
• sporadic activity in Portugal.

The data reported by EISS participants from week 40 of 1999 to week 6 of 2000 and included in the database while this article was being written show that influenza affected most participating countries and that the influenza activity in the general population was mainly due to influenza A(H3N2). Influenza was particularly active in December 1999 and January 2000 in most countries, but occurred later in the Czech Republic and Portugal. In most EISS countries peak influenza activity was seen between weeks 52/99 and 02/00.

Once again, high quality data were collected swiftly by EISS, confirming its reliability and capacity to provide early warnings of influenza activity. Health systems differ widely between the 11 EISS countries and the definition of ‘epidemic’ varies, but it is possible to follow the evolution of influenza activity in real time throughout Europe. Virological data collected in the general population by sentinel physicians in participating countries are remarkably coherent despite health systems being as different as those in Belgium, France, Germany, the Netherlands, Spain, and Switzerland for example.

The absence of a common case definition continues to hinder data comparison. One of the key priorities for developing EISS is to standardise the main indicators collected and the epidemic thresholds. The system also collects other data such as, for instance, the age distribution of the indicators of disease, the circulation of influenza viruses in hospitals and in the community, and in some countries on the circulation of respiratory syncytial virus. The reporting forms that accompany specimens taken by general practitioners are currently being harmonised. Collaboration between EISS laboratories will allow quality assurance controls to be performed at a continental level in the future.

The EISS network is growing from year to year while encouraging the implementation or the development of national surveillance systems for influenza that have to fulfil strict criteria for eligibility.

This article was written on behalf of all the participants in EISS: Aymard M (F), Bartelds AIM (NL), Cohen JM (FR), Heckler R (DE), Heijnen M-L (NL), de Jong JC (NL), Fleming DM (UK), Havlickova M (CZ), Lina B (FR), Manuguerra J.-C. (FR), Marinho Falcao I (PT), Mosnier A (FR), Mueller D (CH), Noone A (UK), Perez-Brena P (ES), Pregliasco F (IT), Rebelo de Andrade H (PT), Samuelsson S (DK), Schweiger B (DE), Snacken R (BE), Thomas Y (CH), Uphoff H (DE), Valette M (FR), Vega T (ES), van der Velden K (NL), van der Werf S (FR), Watson J (UK), Yane F (B), and Zambon M (UK).

Germany
ArbeitsGemeinschaft Influenza (AGI), Marburg; Robert Koch Institut, Berlin; Niedersächsisches Landesgesundheitsamt, Hannover

Belgium
Scientific Institute of Public Health - Louis Pasteur, Bruxelles

Denmark
Statens Serum Institut, Copenhagen

Spain
Instituto de Salud Carlos III, Madrid; Sentinel Networks of Madrid, Castilla y Leon, Pais Vasco, Guadalajara, and Andalucia

France
Groupes Régionaux d'Observation de la Grippe (GROG), Open Rome, Paris; Institut Pasteur, Paris; Centre Hospitalo-Universitaire, Lyon

Great-Britain
Royal College of General Practitioners (RCGP), Birmingham; PHLS Communicable Disease Surveillance Centre (CDSC), London; PHLS Central Public Health Laboratory, London; Scottish Centre for Infection and Environmental Health, Glasgow

Italy
Istituto di Virologia, Milano; Centro Interuniversiterio di ricerche sull’ Influenza (CIRI), Genova; Istituto Superiore di Sanita, Roma

Netherlands
Netherlands Institute for Primary Care (NIVEL), Utrecht; National Institute of Public Health and the Environment (RIVM), Bilthoven; Erasmus University, Rotterdam

Portugal
Instituto Nacional de Saúde, Lisboa; Direcção Geral da Saúde, Lisboa

Czech Republic
National Institute of Public Health, Praha; National Influenza Centre, Praha

Switzerland
Swiss Federal Office of Public Health, Bern; Hôpital Cantonal Universitaire, Geneva

1. Ashley J. Smith TDK. Deaths in Great Britain associated with the influenza epidemic of 1989/90 Population Trends 1991; 65: 16-20

2. Snacken R. Lion J. Van Casteren V. Comelis R, Yane F, Mombaerts M, et al. five years of sentinel surveillance of acute respiratory infections (1985-1990): the benefits of an influenza early warning system. Eur J Epidemiol 1992; 8: 485-90 

3. Snacken R. Bensadon M Strauss A. The CARE telematics network for the surveillance of influenza in Europe. Methods inf Med 1995; 34: 518-22.

4. Snacken R. Manuguerra JC, Taylor P. European Influenza Surveillance Scheme. Methods inf Med 1998; 37: 266-70.