Efforts to assess the changing epidemiology of tuberculosis (TB) in Europe have been limited by differences in definitions and in the quality of tuberculosis surveillance systems between countries (1,2). In order to standardise the surveillance of TB among European countries, consensus recommendations, including a common case definition and a minimum set of variables, were prepared in 1996 by a working group set up by the World Health Organization (WHO) and the International Union Against Tuberculosis and Lung Disease and approved by 37 country representatives (3,4). Based on these recommendations, the EuroTB project started in October 1996 with a one year pilot study. The objectives were: to assess the willingness of countries to take part in a European surveillance system, to collect information about TB cases notified in 1995, to assess the consistency between the information collected and that recommended, and to provide baseline data for future evaluation of the impact of the recommendations. We present here the information collected during the pilot study.

Methods

Health authorities in all 50 countries of the WHO European Region were contacted and asked to appoint one national correspondent.

The European consensus definition of a notifiable case of TB was used (3,4). In countries where laboratories capable of identification of M. tuberculosis complex are widely available, a definite case was defined as a culture confirmed case; in other countries, a patient from whom a sputum smear was positive for acid fast bacilli (AFB) was also considered a definite case. Both definite and other than definite (clinical or radiological signs or symptoms compatible with TB and a clinician's decision to treat with a full course of antituberculous treatment) incident cases notified in 1995 were reportable.

Data requested on each case (3,4) were: year of report, country of report, age, sex, geographic origin according to place of birth, status of the case as new (a patient never previously diagnosed with TB) or recurrent following the national definition, site of disease (pulmonary - involving the lung parenchyma and/or the tracheobronchial tree, extra-pulmonary or both), bacteriological confirmation by culture, and the results of smear examination for AFB on spontaneously produced sputum.

Individual anonymous computerised data were requested. Countries that could not do so were asked to complete standard tables. Given that information collected in the countries in 1995 would probably not always fit the format and definitions of the consensus recommendations, the following alternative classifications were accepted: patient's geographic origin defined according to citizenship; site of disease defined as respiratory, extra-respiratory, or both (respiratory TB includes pulmonary TB and/or pleural TB and/or intra-thoracic lymphatic TB); and bacteriological confirmation based on culture and/or sputum smear results.

Notification rates of incident TB cases (further referred to as incidences) in 1995 were calculated per 100 000 population, based on United Nations demographic data.

Results

1. Feasibility

A national correspondent was identified in 49 of the 50 countries. Forty-eight countries sent data on TB cases notified in 1995: 19 countries provided individual computerised data, 22 provided aggregate data, five provided only a total number of cases without further description, and two sent information too incomplete to be included. Forty-six countries supplied total numbers of TB cases notified in 1995 and 41 provided detailed characteristics of the cases.

Forty-three of the 46 countries reported all new and recurrent cases, two (Greece and Kazakhstan) reported new cases only, and one (Spain) reported only new respiratory cases. Fifteen countries did not report TB cases in foreign citizens, and two of these also excluded cases among prisoners and military personnel.

Table 1 shows the data on notified cases supplied by the 41 countries that provided detailed information. Among these 41 countries, seven described only the new (and not the recurrent) cases: cases notified in these countries were not included in the analysis of disease characteristics.

Among the 19 countries that provided individual computerised data, the following proportions of values for each variable reported were missing:

• 0% to 1% for age and sex,
• 0% to 10% for geographic origin,
• 0% to 86% for new/recurrent status (data from the two countries with more than 30% of missing values were not analysed),
• 0% to 3% for the major site of the disease,
• 70% to 99% for the minor site of disease; this could reflect the absence of minor site or incomplete information,
• 0% to 50% for culture results,
• 0% to 72% for sputum smear results.

Tableau 1 / Table 1 : Données disponibles pour les 41 pays * qui ont fourni des informations détaillées sur les cas de tuberculose, Région Europe de l'OMS, 1995 / Data available from 41 countries* that reported detailed information about tuberculosis cases, WHO European Region, 1995.

* Albanie, Arménie, Autriche, Azerbaïdjan, Bélarus, Belgique, Bosnie-Herzégovine, République Tchèque, Danemark, Estonie, Finlande, France, Allemagne, Grèce, Hongrie, Islande, Israël, Italie, Kirghizistan, Lettonie, Lituanie, Luxembourg, Ex- République Yougoslave de Macédoine, Malte, Moldova, Pays-Bas, Norvège, Pologne, Portugal, Roumanie, Fédération de Russie, Saint-Marin, Slovaquie, Slovénie, Suède, Suisse, Turquie, Turkménistan, Royaume-Uni, Ouzbékistan, Yougoslavie.

* Albania, Armenia, Austria, Azerbaijan, Belarus, Belgium, Bosnia-Herzegovina, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Israel, Italy, Kyrgyzstan, Latvia, Lithuania, Luxembourg, Former Yugoslav Republic of Macedonia, Malta, Moldova, Netherlands, Norway, Poland, Portugal, Romania, Russian Federation, San Marino, Slovakia, Slovenia, Sweden, Switzerland, Turkey, Turkmenistan, United Kingdom, Uzbekistan, Yugoslavia.

2. TB cases notified in 1995

In 1995, 276 811 cases of TB were notified in 46 European countries. The number of cases reported by country varied from one case in Monaco to 96 828 cases in the Russian Federation. The overall incidence was 34.6 cases per 100 000 population. National incidences varied greatly, from 2.7/100 000 population in Malta to 101.9/100 000 in Romania (figure 1). The incidence (per 100 000) was lower than 20 in 22 countries (group 1) and 20 or over in 24 countries (group 2) (figure 1).

In 32 countries that reported the age of all new and recurrent cases (n = 107 096), 4% were children under 15 years of age, 77% adolescents and adults up to 64 years old, and 19% people aged 65 years or over. The overall male to female ratio was 1.8.

Age and sex specific incidences in the two groups of countries are presented in figure 2 showing:

• in group 1 (incidence <20/100 000), a steady increase with age with highest rates in the oldest age group; rates were similar in males and females until the age of 15 to 24 years, after which the incidence in males became twice as high as in females.
• in group 2 (incidence > = 20/100 000), low rates in children; for males the incidence peaked at 45 to 54 years of age; for females, the incidence had a bimodal distribution with highest values at 25 to 34 years and 65 years and over. Sex specific rates differed widely from 35 years of age upwards.

In the 34 countries that reported new or recurrent status, 10% of the cases were recurrent (23 348/226 785), with small variations between countries.

Thirty-two countries reported the site of the disease for all cases (n = 135 743). In 22 countries, 80% of the cases were reported with pulmonary TB (alone or associated with an extra-pulmonary site). In 10 other countries, 90% of the cases were reported with respiratory TB (alone or associated with an extra-respiratory site). In the 24 countries that reported sputum smear results, 37% of all cases were smear positive.

In the nine countries where bacteriological confirmation was based on culture, 55% (11 527/20 776) of the cases were confirmed. In the 18 countries that used either culture or sputum smear results to confirm cases, 52% (31 487/60 988) of cases were confirmed. Among these 18 countries, 12 reported separate results for both examinations and reported 59% of confirmed cases: 43% had positive culture, 13% had positive sputum smear and unknown culture results, and 3% had positive sputum smear and negative culture.

Tableau 2 / Table 2 : Cas de tuberculose par origine géographique, pays de naissance ou nationalité, dans 21 pays, 1995 / Tuberculosis cases by geographic origin, country of birth or citizenship, in 21 countries, 1995.

* Les totaux comprenaient 1101 cas dont l'origine géographique était inconnue
*Totals included 1101 cases with unknown geographic origin.

Discussion

The response was excellent: 49 countries collaborated and data on TB cases notified in 1995 could be collected in 46 countries. These data were collected before the consensus recommendations were implemented.

Differences in case definitions and in the populations covered by the reporting were observed between countries. Several countries reported only new (and not recurrent) cases. Furthermore, differences in criteria used to notify cases as recurrent might bias comparisons between countries of both incidences and proportions of recurrent cases. Several central and eastern countries mentioned that foreign citizens were not included in reports, and some that prisoners and military personnel were notified in specific TB registers only. Not reporting TB in prisoners may introduce a major bias in incidences in some countries (5). Vulnerable population groups such as homeless people, illegal immigrants, and refugees might also not be reported in some countries. The relatively low incidence among children in several countries where the overall incidence is high suggests that differential underreporting may have occurred among children. The issue of underreporting in individual countries was not assessed during the pilot study, however, and may be substantial in some western European countries.

Although notification rates varied widely across Europe, some geographical trends were apparent. Almost all countries with lower incidences (group 1 <20/100 000) were located in western Europe and almost all countries with higher incidences (> = 20/100 000) in central and eastern Europe. Age and sex specific incidences in the two groups of countries were consistent with epidemiological patterns typical of countries with low and high TB incidence. For the reasons discussed above, however, countries should be compared with extreme caution.

The availability of detailed information varied by country and was excellent for age, sex, and site of the disease. The factor least documented was the geographic origin of the patient, and comparisons were complicated by differences in the definition of 'foreign'. The large variations observed in the proportions of cases among 'foreigners' reflect past and current immigration patterns as well as the prevalence of TB in the countries of origin of the migrants.

The methods by which cases were confirmed - by culture alone or culture and/or sputum smear - varied between countries and did not necessarily depend only on the availability of routine culture. Some European countries still rely fully on sputum smear examinations to confirm cases, but it was surprising that only three countries of the European Union required culture confirmation as recommended. The proportions of cases confirmed, whatever definition was used, were low and varied widely between countries. The proportions of pulmonary cases with positive sputum smear were also low. In countries that reported individual results, data on bacteriological status were very incomplete. Some countries may not follow up bacteriological results.

Conclusion

Participation of almost all European countries in the pilot study shows that surveillance of TB in Europe is possible. Disparities in definitions and in availability and completeness of information reflect pre-existing differences in reporting systems and emphasise the need for further standardisation. During its second year, the EuroTB project will establish a routine system for the surveillance of TB case notifications and study the feasibility of a system of surveillance for antituberculosis drug resistance. EuroTB aims to promote the implementation of the consensus recommendations by collecting annual data. Changes in TB reporting systems across Europe will be monitored using the information presented here as a baseline.

Detailed results and a list of co-authors can be found in: Surveillance of TB in Europe. Report on the feasibility study (1996-1997). TB cases notified in 1995, October 1997. This report is available upon request from CESES, Hôpital National de Saint Maurice, 14 rue du Val d'Osne, 94410 Saint Maurice, France and on the website: http://www.b3e.jussieu.fr/ceses/eurotb.

References

1. Raviglione M, Sudre P, Rieder HL, Spinaci S, Kochi A. Secular trends of tuberculosis in Western Europe. Bull World Health Organ 1993; 71: 297-306.
2. Raviglione MC, Rieder HL, Styblo K, Khomenko AG, Esteves K, Kochi A. Tuberculosis trends in Eastern Europe and the former USSR. Tubercle Lung Dis 1994; 75: 400-16.
3. Rieder HL, Watson JM, Raviglione MC, Forssbohm M, Migliori GB, Schwoebel V, et al. Surveillance of tuberculosis in Europe. Recommendations of a Working Group of the World Health Organization (WHO) and the European Region of the International Union Against Tuberculosis and Lung Disease (IUATLD) for uniform reporting on tuberculosis cases. Eur Resp J 1996; 9: 1097-1104.
4. Schwoebel V, Rieder HL, Watson J, Raviglione M for the working group for uniform reporting on TB cases in Europe. Surveillance of TB in Europe. Eurosurveillance 1996; 1: 5-8.
5. Wares DF, Clowes CI. Tuberculosis in Russia. Lancet 1997; 350: 957.