EuroTB

On average, three contacts per country were necessary to obtain and verify the data. All communication tools were used: telephone, fax, electronic mail and conventional mail. Communication was difficult with some countries, particularly some Asian republics of the former USSR. This may have hindered the validation of information reported from these countries.

A national correspondent was identified in 49 of the 50 countries of the WHO European region (98%). No contact could be established in Ukraine.

Forty eight countries sent data on tuberculosis cases notified in 1995. Among these 48 countries, 19 (40%) provided individual computerised data, 22 (46%) provided aggregate data, five (10%) provided a total number of cases without further description and two countries (4%), Georgia and Tajikistan, sent information which either was too incomplete or could not be validated on time. One country, Croatia, could not send data because of administrative constraints.

After validation, a total number of tuberculosis cases notified in 1995 was included for 46 countries, and data from 41 countries were included in the overall description of the cases.

Among the 46 countries reporting a total number of cases notified in 1995, 43 (93%) included all incident (new and recurrent) definite and other than definite cases as recommended, two (Greece and Kazakhstan) reported only the total number of new cases, and one (Spain) reported only the total number of new respiratory cases.

Among the 41 countries providing detailed information on the characteristics of the cases, eight (19%) provided this information on incident new cases only. Data for those countries are presented separately. Through contacts with national correspondents and by examining the data provided, it was observed that 15 of these 41 countries (37%) did not report tuberculosis cases in foreign citizens; of these, two also mentioned that prisoners and military personnel were not included.

Data availability was evaluated among the 41 countries providing detailed information on notified cases:

information on age was available as exact age (in years) or according to recommended age groups in 33 countries (80%) and according to different age groups in 7 countries (17%). One country, Turkey, could not provide this information for 1995.

information on gender was available in all but one country (Turkey).

geographic origin of the patient was reported by 21 countries (51%), of which eight reported the country of birth, two reported the origin as born in the country or abroad, six reported the country of citizenship and five reported the origin as national or foreign citizen. Among the two latter groups, the Netherlands included Dutch patients born abroad to mothers of foreign origin in the category of "foreign citizens", and Israel included foreign patients applying for Israeli nationality in the category of "national citizens".

new or recurrent status of the case was provided by 34 countries (83%). Only a few countries provided the national definition of a new and a recurrent case. Definitions were not consistent across countries.

site of disease was reported in 39 countries (95%): 13 provided the specific major site (among which 10 also provided the minor site), 10 reported cases as pulmonary or extra-pulmonary, and 16 reported cases as respiratory or extra-respiratory. Information from the two remaining countries, Azerbaijan and Turkmenistan, could not be included because of validation problems.

bacteriological confirmation of the cases was available in 34 countries (83%). The type of confirmation varied across countries and was not necessarily dependent on the availability of level II laboratories in the country : nine countries used the positive culture (as recommended) and 25 used either positive culture or positive sputum smear for defining a case as bacteriologically confirmed.

results of sputum smear examination were reported by 27 countries (66%).

Completeness of information was evaluated for the 19 countries providing individual computerised data.

In countries which provided variables describing patient's characteristics, the proportion of missing values was very low, with a maximum of 10% in France for the variable geographic origin.

new or recurrent status of the case
In five countries (Denmark, France, Finland, Italy, Sweden), information was missing for more than 10% (28% to 94%) of the cases. It was decided to include information only for those countries with less than 30% of the cases with unknown status.

site of the disease
The proportion of missing values was very low for the major site, with a maximum of 3% in Luxembourg. For the minor site, the proportion of missing values was high (70% to 99%) in all 10 countries providing this variable. However, this could reflect the absence of minor site as well as incomplete information.

bacteriological information
For the variables indicating whether bacteriological examinations were performed, the proportion of missing values varied across countries, ranging from 0% to 36% for the culture, and from 0% to 61% for the sputum smear examination. Cases for whom it was not known whether the examination had been done may have been erroneously coded as "examination not done", as suggested by the absence of missing values for these variables in a few countries. In most countries, whether the examination was done for an individual patient was known only if a result (positive or negative) was reported. Thus, the proportion of missing results was 0% among cases reported as having the examination performed in those countries (12 of the 16 countries reporting sputum smear results, eight of the 17 countries reporting culture results). In some countries, only positive results were reported: this reporting pattern was observed in Czech Republic and France for the sputum smear examination, and in Denmark, Finland, Luxembourg and Norway for the culture.