J.-E. Schmidt ¹, A.-E. Tozzi ¹, L. Rava' ¹, S. Glismann ², on behalf of the EUVAC-NET country contacts*
1 Istituto Superiore di Sanità (ISS), Italy;
2 Statens Serum Institut (SSI), Denmark

A questionnaire was mailed out to member states of the European Union (EU) plus Switzerland, Norway, and Iceland, to inquire about the type of information routinely recorded in national pertussis surveillance systems. Information was requested on surveillance methods, type of information recorded for cases of pertussis, vaccination schedule, type of vaccine used, and methods for estimating vaccination coverage. Local surveillance methods, vaccination strategies, and methods to estimate vaccination coverage were found to differ widely across the participating countries. The results of the questionnaire survey show, however, that there are comparable subsets of variables common to many countries. Future activities of the EUVAC-NET project will include defining the homogeneous elements in national systems and to group appropriately those countries with common surveillance features.

Introduction

The EUVAC-NET project was set up to operate a community surveillance network for vaccine preventable disease in the European Union (EU) (1). The diseases initially targeted by EUVAC-NET are measles and pertussis. This report summarises the results of a questionnaire survey on pertussis surveillance systems conducted in 18 countries in preparation for the EUVAC-NET database.

Despite continuing efforts to increase vaccination coverage in indi-vidual countries pertussis remains a disease of substantial public health importance. Surveillance data for 1998 gathered by the European Regional Office of the World Health Organization (WHO) show that the incidence of pertussis in the countries participating in the EUVAC-NET project ranged from 0.1 to 52.3 per 100 000 (2). WHO aims at reducing the incidence of pertussis to below 1/100 000 by 2010 (3).

The results of the questionnaire survey reflect national surveillance systems and control strategies for pertussis as of late 2000. They also highlight changes that have occurred since the publication of a report describing pertussis vaccination practices in Europe in late 1999 (4). A possible structure of the EUVAC-NET database and a common set of variables for a minimal dataset (MDS) for case based reporting will be discussed. In conclusion, implications for the implementation of pertussis surveillance in the European context will be discussed on the basis of the potential for queries on the epidemiology of pertussis in the individual countries and across countries.

Methods

The 18 European countries participating in the project (along with their respective country codes) are listed in table 1. To inquire about the type of information routinely recorded in national pertussis surveillance systems, an 18 item questionnaire was mailed out to contacts in the participating countries. Information was requested on the following subjects.

Surveillance methods:
• type of reporting;
• case definition;
• type of laboratory tests used for case confirmation.

Information recorded for cases of pertussis:
• personal information;
• clinical characteristics and disease outcome; and
• vaccination status.

Information regarding country specific policies and regulations:
• vaccination schedule;
• type of vaccine used;
• methods used for estimation of vaccination coverage.

Table 1. Type of reporting, case definition, and type of laboratory confirmation by country

Results

All questionnaires were returned during the last trimester of 2000. The characteristics of national pertussis surveillance systems were found to be diverse and dynamic.

Surveillance methods

Surveillance methods in place in different countries are illustrated in table 2.

Table 2– Variables recorded in case based surveillance databases by countries

Type of data

With the exception of Austria, Spain, and Iceland all systems record individual data. Iceland was recording aggregate data at the time of the survey (and is reported here as such) but changed to case based reporting as of March 2001. Three countries operate sentinel systems with established denominators: France, Switzerland, and one of the three systems operating in Greece (3). In Germany, data are collected only in the five former East German states and may not be representative for the whole of Germany, since vaccination coverage in those five states has traditionally been high compared with the rest of the country. Out of two systems operated in the United Kingdom (UK) – disease notification (aggregate data) and laboratory reporting (case based data) – only the latter is described in this paper. In Norway, surveillance is restricted to cases younger than 20 years; for cases older than 20 years the collected variables are limited to identification, age, sex, residence, and date of onset. In Denmark surveillance is restricted to children younger than 24 months.

Case definition and laboratory methods

Two broad categories can be identified: countries where a case is defined exclusively on the basis of clinical criteria and those where a case is defined on the basis of a laboratory procedure with or without additional criteria. Four countries follow the WHO clinical case definition (Greece, Portugal, Spain, Switzerland). Two countries (France and the Netherlands) adopt a modified WHO clinical case definition, and all others rely on a physician’s clinical diagnosis.

Eleven systems rely on some form of laboratory procedure for case confirmation, three of which also include epidemiological linkage. A common pattern is the combination of culture and polymerase chain reaction (PCR) with or without additional serological testing. Of the five countries also using enzyme linked immunosorbent assay (ELISA), Austria and France use paired samples, Finland uses a single sample, and Norway and the Netherlands use both.

Information recorded in national databases

Only those countries conducting case based reporting (n=15) are included in this description, which is summarised in table 3.

Table 3. Vaccination schedules, type of vaccine used, and methods for estimating vaccination coverage by country

Personal and clinical characteristics and disease outcome

Age and sex of cases are recorded in all countries. Many of them also record the date of disease onset and the date of disease notification.

Few systems report on the clinical characteristics of pertussis. The symptoms of interest in the questionnaire included duration of cough, presence of paroxysms, inspiratory whoop, and post-tussive vomiting. The two sentinel systems (France, Sweden) cover the whole range of symptoms. Death as a consequence of disease is recorded in nine countries, admission to hospital also in nine countries.

Vaccination status

Eleven countries collect data on the vaccination status of cases. Seven of these indicate the actual number of doses administered. Five record the type of vaccine a patient had received, whole cell vaccine (Pwc) or acellular vaccine (aP).

Information related to country specific policies

Information on vaccination policies is recorded in table 4.

Table 4 - Possible queries by country and by case definition¹

Vaccination schedule and type of vaccine used

The recommended schedule for vaccination suggests similar varia-bility. Three countries limit immunisation to a primary series of three doses during the first year of life. In 15 countries, booster policies are in place. Recent changes in vaccination schedule pertain to these booster policies: four countries have instituted an additional booster at preschool age (Belgium, Iceland, Netherlands, Sweden) (4).

In all 18 countries, aP vaccine is recommended or used. In nine countries, Pwc vaccine has been discontinued altogether. The scenarios for concomitant use of both vaccines are manifold: Portugal uses aP only in the advent of severe adverse reactions to Pwc. In the UK, aP is temporarily being used because of supply problems with Pwc. In Greece, aP is readily available on the market; but there is no official recommendation for its use.

The type of vaccine used has changed in two countries: Belgium has switched to aP for dose 4 (at 13 months). Similarly, Netherlands has switched to aP for dose 5 at 4 years of age.

Estimation of vaccination coverage.

The methods for estimating vaccination coverage differ most between countries. In 13 systems, the number of children who receive three doses in the first year of life serves as the numerator, and single birth cohorts serve as the denominator. Six countries conduct sample surveys at different time intervals. In France, an annual clinical assessment including immunisation status in children aged 24 months is documented on special health certificates (the denominator for the estimation of vaccination coverage being the number of such certificates). In Germany vaccination coverage is estimated by assessing immunisation status of children at school entry.

Finally, we hypothesised which queries a user could run in the various participating countries (table 4). Most importantly, the incidence (per 100 000) of new cases stratified by age, sex, and geographical area can be calculated by all 15 countries that operate case based surveillance systems.

Discussion

The reported survey describes the structure, organisation, and contents of national pertussis surveillance systems in Europe and constitutes a first necessary step to assess the comparability of data on the incidence of disease among participating countries. At this point in time, the burden of disease in EU member states should be measured as accurately as possible. The incidence of the disease by age group and geographical area are essential data for monitoring the trend of the disease across countries implementing different vaccination strategies. Temporal trends are also useful for monitoring the impact of vaccination strategies. In addition, information on mortality and admission to hospital associated with the disease should be collected.

Some difficulties in implementing a global surveillance system for pertussis have already been identified (Global Meeting on Pertussis Surveillance, WHO Headquarters, Geneva, 16-18 Oct 2000). The EU countries operate different vaccination strategies, and coverage varies. Clinical diagnosis alone may not be specific, but laboratory confirmation of cases is not performed in all countries and is not standardised.

To address these issues, an initial assessment of the available data is required to identify comparable subsets common to most EU countries. In this respect, the exercise of identifying the potential queries, which could be run with existing routine data, allows identification of the most striking differences that should be overcome in the future. This approach also shows that some basic information is already available provided that these data are appropriately grouped. Such information might be viewed in relation to the different vaccination strategies.

A solution to the issue of diversity of case definitions and laboratory methods may be to apply different levels of case classification, and to summarise cases under two or three of such levels – for example, suspected or confirmed cases (5, 6).

Additional sources of information may be required to address the issue of obtaining a complete epidemiological picture of pertussis, in particular in countries where data reflecting the severity of disease, such as mortality and admission to hospital, is not part of the routine surveillance system.

Finally, surveillance of vaccination coverage in Europe is problematic. Given the variety of approaches to its estimation, the standardisation of methods would be desirable.

Conclusions and recommendations

Given these survey results, the EUVAC-NET database for pertussis should comprise three separate sub-databases:

(1) case based data;

(2) aggregate data;

(3) non case related information.

Only two countries record aggregate data (Austria, Spain). Participation in EUVAC-NET could be an incentive to consider a shift to case based reporting. The following sets of variables are proposed for an MDS of the EUVAC-NET database for case based reporting.

(1) Age

(2) Gender

(3) Area of residence

(4) Date of disease onset

(5) Vaccination status

(6) Admission to hospital

(7) Death as disease outcome

(8) Case classification (suspected/confirmed).

In order to interpret these variables correctly, additional information regarding national policies and regulations on pertussis, such as vaccination schedule and coverage assessment methods, would be helpful. Many of them are already available from other sources (6-9).

The tasks to be accomplished in the future by EUVAC-NET with respect to pertussis include the following:

• The study of available data to highlight differences in surveillance systems across countries, but also the identification of homo-geneous information and parts of the national databases that can be pooled with those from other countries.

• In depth study of the differences in case definitions observed in participating countries, and implementation of initiatives aimed at standardising laboratory methods for confirmation of the disease.

EUVAC-NET will further encourage and promote homogeneity of the data collected by each individual surveillance system. This project also provides an opportunity for promoting case based surveillance in countries still dealing with aggregate data, and for implementing efficient methods to assess vaccination coverage.

Close monitoring of the epidemiology of pertussis is required to assess the impact of different vaccination programmes and eventually adjusting vaccination strategies in order to meet the WHO goal of reducing the incidence of pertussis to below 1/100 000 by 2010. 

R. Strauss, Ministerium für Soziale Sicherheit und Generationen, Austria
• . Ronveaux, Institut Scientifique de la Santé Publique, Belgium
S. Glismann, Statens Serum Institut, Denmark
I. Davidkin, National Public Health Institute, Finland
D. Levy-Bruhl, Institut de Veille Sanitaire, France
G. Rasch, Robert Koch-Institute, Germany
T. Panagiotopoulos, Institute of Child Health, Greece
H. Briem, Directorate of Health, Iceland
D. O'Flanagan, National Disease Surveillance Centre, Republic of Ireland
L. Vellucci, Ministerio di Sanità, Italy
P. Huberty-Krau, Inspection Sanitaire, Luxembourg
H. Blystad, Statens institutt for folkehelse, Norway
G. Freitas, Direcção-Geral da Saúde, Portugal
C. Amela, Instituto de Salud Carlos III, Spain
A. Tegnell, Smittskyddsinstitutet, Sweden
H. Zimmermann, Gesundheitsministerium, Switzerland
H. De Melker, Rijksinstituut voor de Volksgezondheid en Milieu, Netherlands
J. White, Public Health Laboratory Service Communicable Disease Surveillance Centre, England and Wales

1. Glismann S, Rønne T. Tozzi A. The EUVAC-NET project: creation and operation of a surveillance community network for vaccine preventable infectious diseases. Eurosurveillance 2001; 6: 94-8.

2. World Health Organization (WHO), Regional Office for Europe, Computerized Information System for Infectious Diseases (CISID). http://cisid.who.dk.

3. World Health Organization (WHO), Regional Office for Europe. Health 21: the health for all policy framework for the WHO European Region. European Health for All Series No. 6, 1999.

4 Therre H, Baron S. Pertussis immunisation in Europe - the situation in late 1999. Eurosurveillance 2000; 5: 6-10.

5. World Health Organization. WHO recommended surveillance standard. Geneva: World Health Organization, 1997 (WHO/EMC/DIS/97.1).

6. European Union Charter Group. Case definitions for infectious diseases under EU-wide surveillance. Revised June 2000. (http://www.smittskyddsinstitutet.se/archive.asp?theSection=4).

7. Osborne K, Weinberg J, Miller E. The European Sero-Epidemiological Network (ESEN). Eurosurveillance 1997; 2: 29-31.

8. Istituto Superiore di Sanità. Extended inventory on means and resources for infectious disease control. Rome: ISS, Italy. 2000. (http://iride.cineca.org).

9. Scientific and Technical Evaluation of Vaccination Programmes in the European Union. (http://www.euvax.org).