On 6 June 2017, the World Health Organization (WHO) published updates to its ‘Essential Medicines List’ (EML). Read more here.

Eurosurveillance is on the updated list of the Directory of Open Access Journals and in the SHERPA/RoMEO database. Read more here.

Follow Eurosurveillance on Twitter: @Eurosurveillanc

In this issue

Home Eurosurveillance Monthly Release  1997: Volume 2/ Issue 6 Article 1
Back to Table of Contents
en es fr pt

Eurosurveillance, Volume 2, Issue 6, 01 June 1997
Research Articles
Travel associated legionellosis among European tourists in Spain

Citation style for this article: Galmés Truyols A, Martinez Navarro JF. Travel associated legionellosis among European tourists in Spain. Euro Surveill. 1997;2(6):pii=155. Available online:

A. Galmés Truyols1, J.F. Martínez Navarro2
1. Field Epidemiology Training Programme (FETP) 1995-97.
2. Centre National d'Epidémiologie, FETP, Madrid, Espagne.


Travel associated Legionnaires' disease has caused concern among European countries since the second half of the 1980s because of the morbidity among citizens of the European Union and because of the threat posed to the economies of the Mediterranean countries by the occurrence of the disease among tourists. As a result, the European Working Group for Legionella Infections (EWGLI ) (1) was set up in 1986 coordinated by the National Bacteriology Laboratory in Stockholm until 1993 when this role was transferred to the Public Health Laboratory Service Communicable Disease Surveillance Centre in London. Case reports are sent from patients' countries of residence to countries they have visited.

EWGLI has developed a surveillance scheme based on a computer software program, the European Legionellosis Surveillance Scheme (ELSS). Monthly updates are sent to all collaborators of all available data from throughout Europe since 1987 (2). The aim of this study was to analyse data covering cases of legionellosis associated with travel to Spain, including the Balearic and Canary islands.


An Epi Info analysis was run on the ELSS program database, updated as of 31 December 1995. The numbers of travellers who had arrived from other European countries and stayed in tourist accommodation in Spain were obtained from the Spanish National Statistics Office (INE) (3,4) and used as denominators to calculate rates.

The ELSS program contains two interrelated databases: each record on one database consists of a single accommodation address for a given patient during the incubation period, so that for any one case there are as many records as there are accommodation addresses during a specific trip; the second database contains information about individual patients, with one record per case. These two databases can be linked and cross-referenced using a case-ID field. For the purposes of analysis, Dbase III Plus and Epi Info 6.01 software packages were used. Since all cases reported in Europe are pooled in the same databases, the first task was to separate patients who had travelled to Spain from those who had been to other European destinations. The second step was to code regions (using the designated Spanish Autonomous Region codes) and accommodation addresses.

Hotels and holiday apartments in Spain have a similar structure and management and so we studied both together.

The duration of stay in Spain was calculated for all patients, as were the periods between arrival in Spain and the onset of symptoms and between return to country of permanent residence and the onset of symptoms. These periods were calculated for all hotel stays in the case of travellers who had stayed in more than one hotel.

The incubation period for legionellosis was taken as two to 10 days (5). The disease was said to be confirmed if any legionella was cultured or if a fourfold rise in the titre of antibodies against Legionella pneumophila sg1 titre was observed. A presumptive diagnosis of L. pneumophila sg1 infection was made if the case was diagnosed on the basis of a single high antibody titre or if another method was used, that is the word “other” appeared in the report of the case. Infections with all other Legionella species or serotypes were regarded as presumptive unless diagnosed by culture.

The criterion chosen in this study to define case-clustering in any one hotel was the appearance of more than one case in the same calendar year or the appearance of a single case in two or more successive years.


A total of 281 cases were reported from 1987 to 1995. In 1995 two duplicate cases (repetition of case ID codes) were eliminated. Men accounted for 69% of cases overall, 54% of whom were aged 45 to 64 years. Women accounted for 29% of cases, 46% of whom were aged 45 to 64 years, the most numerous group (table 1). The sex of 2% of cases was unknown.

Tableau 1 : Légionelloses chez des touristes européens en Espagne, 1987-1995. Répartition par âge et par sexe
Table 1: Legionellosis among European tourists in Spain, 1987-1995. Distribution by age and sex.

Sexe / Sex
Groupe d’âge Masculin Féminin Inconnu Nombre total (%)
Age group Male Female Unknown Total number (%)
0 - 24 0 3 0 3 (1)
25 - 44 29 16 0 45 (16)
45 - 64 104 37 1 142 (51)
> 64 58 24 0 82 (29)
Unknown 3 1 5 9 (3)
TOTAL 194 (69) 81 (29) 6 (2) 281 (100)

The largest number of cases (53) was reported in 1990 (figure 1). The average of 40 cases were reported each year from 1989 to 1995. Only six cases had been registered before 1989.

The diagnosis of 154 cases (55%) was confirmed by culture or fourfold rise in antibody titre and presumed in 121 cases (43%). The method of diagnosis was unknown in 6 cases (2.1%).

When reported, 21 cases were ill (7%), 123 (44%) had recovered, 28 (10%) had died, and no data were available on the health status of the remaining 109 (39%).

Date of onset of symptoms was known for 265 (94%) cases. Length of stay in Spain was unknown in 60 cases (21%) and less than three days in three cases (figure 2). In 66 cases (23%), analysis of the dates of symptom onset and dates of stay in Spain failed to show whether the patient had been in Spain during the incubation period of the disease. Data on a further 10 yielded periods of time incompatible with having acquired infection in Spain: five having become ill too soon after arrival in Spain and five too long after returning home or reaching another destination.

The 281 cases had stayed at a total of 303 hotels. Two hundred and eight (69%) such stays occurred within the likely incubation periods, 20 stays occurred outside the incubation period and the dates of 75 (25%) were unknown (figure 3). Nineteen cases had stayed at a total of 62 hotels, yet data on the tourist stays during the disease incubation period were available from only 21 of these establishments.

Two hundred and fifty-nine of the 281 reported cases had stayed at hotels. Of these, 240 had been at only one hotel, 19 at more than one (from two to five hotels; 62 in all), and 10 in more than one Autonomous Region. Six stayed in private homes, one in a caravan, and details of accommodation remained unknown for 15 patients.

In all, 186 hotels, 32 of which are classified as apartments on the EWGLI database, were associated with cases and 49 of these hotels were associated with between two and nine cases during the study period. Forty-three of the 49 hotels met the definition of single hotel case clustering in 12 of which all the cases occurred in the same year. When stays during the incubation period were analysed, however, only 38 hotels met the compatibility criteria (78% of all those associated with more than one case) and within this group, case clustering was considered to have occurred in 34 (table 2).

Tableau 2 : Légionelloses chez des touristes européens en Espagne, 1987-1995. Hôtels associés ventilés par Région autonome.
Table 2: Legionellosis among European tourists in Spain, 1987-1995. Associated hotels broken down by Autonomous Region.

Région Autonome Cas Hôtels / Hotels
Autonomous Region Cases
Total Associés à plus d’un cas Répondant aux critères
Total Associated with more d’agrégats de cas (1)
than one case Clustering criteria met (1)
Iles Baléares
Balearic Isles 103 60 19 17
Catalonia 55 33 9 9
Région de Valence
Valencian Region 39 26 9 6
Iles Canaries
Canary Islands 33 27 4 3
Andalousie 34 32 6 6
Reste de l’Espagne
Rest of Spain 4 9 2 2
Autre (2)
Other (2) 10 - - -
Unknown 3 - - -
TOTAL 281 186 49 43
(1) Plus d’un cas dans une année ou un seul cas dans au moins deux années consécutives /
More than one case in any one year or a single case in two or more successive years.
(2) Plus d’une région visitée, sans détermination de celle à l’origine de la contamination /
Visited more than one region, without it being possible to ascertain in which infection took place.

From 1989 to 1995, the highest mean annual incidence occurred in the Valencian Region (0.85 cases per 100 000 European tourists) and the lowest in Andalusia (0.31 per 100 000). Rates varied considerably from year to year owing to the small number of cases. The only region with a stable incidence was the Balearic Isles (0.50/100 000; 95% confidence interval (CI) 0.03-1.12), from which 101 cases were reported, a little over a third of the overall total (table 3).

Tableau 3 : Légionelloses chez des touristes européens en Espagne. Cas et taux moyens annuels par Région Autonome, 1989-1995.
Table 3: Legionellosis among European tourists in Spain. Cases and annual mean rates by Autonomous Region, 1989-1995.

Taux moyen annuel / Annual mean rate
Cas / Cases (cas pour 100 000 touristes européens /
cases per 100 000 European tourists)
Région Autonome / Taux / Rates I.C. 95% / 95% C.I.
Autonomous Region
Région de Valence
Valencian Region 37 0.85 (0-2.86)
Iles Baléares
Balearic Isles 101 0.50 (0.03-1.12)
Catalonia 55 0.35 (0-1.00)
Iles Canaries
Canary Islands 32 0.32 (0-1.16)
Andalousie 33 0.31 (0-1.11)
Reste de l’Espagne
Rest of Spain 4 0.03 (0-0.51)
Autre (1)
Other (1) 10
Unknown 3
TOTAL 275 0.38 (0.16-0.65)
(1) Plus d’une région visitée / Visited more than one region

Distribution of cases by country of residence showed that Swedish tourists had the highest rate of illness (mean annual rate of 2.52 cases/100 000 European tourists) and Germans had the lowest (0.06/100 000). As with regional case distribution, rates by country of origin proved unstable from year to year, and statistical significance was observed only for British cases, with 184 diagnosed cases from 1989 to 1995 (70% of the total) and a rate of 1.12/100 000 (95% CI 0.33-2.10) (table 4). Both rates were calculated for years 1989-95, given that until 1989, only 6 cases had been reported.

Tableau 4 : Légionelloses chez des touristes européens en Espagne. Cas et taux moyen annuel par pays d’origine, 1989-1995
Table 4: Legionellosis among European tourists in Spain. Cases and annual mean rates by country of origin, 1989-1995

Taux moyen annuel / Annual mean rate
Nombre de cas / (cas pour 100 000 touristes européens /
Number of cases cases per 100 000 European tourists)
Pays d’origine Taux / Rates I.C. 95% / 95% C.I.
Country of origin
Sweden 31 2.52 (0-9.20)
Denmark 15 1.60 (0-8.89)
Royaume Uni
United Kingdom 184 1.12 (0.33-2.10)
Holland 24 0.94 (0-3.96)
Norway 7 0.73 (0-13.99)
Germany 9 0.06 (0-0.38)
Other 5 0.01 (0-0.20)
TOTAL 275 0.28 (0.16-0.65)


Cases of legionellosis among European visitors to Spain show a similar age and sex distribution as reported elsewhere (6).

The differences in rates of legionellosis between travellers from different European countries to the same region is remarkable (table 5), and suggests the existence of an information bias, due to differences in national surveillance systems or degrees of participation in the European system. We were unable to adjust rates we calculated for country of residence for lack of appropriate denominators distributed by age and sex.

Tableau 5 : Légionelloses chez des touristes européens en Espagne. Taux pour 100 000 visiteurs, par pays d’origine et par région visitée, 1989-1995
Table 5: Legionellosis among European tourists in Spain. Rates per 100 000 visitors, by country of origin and region visited, 1989-1995

Suède / Danemark / Royaume Uni / Hollande Norvège / Allemagne /
Sweden Denmark United Kingdom Holland Norway Germany
Région de Valence / Taux / Rates - - 2.01 1.77 - -
Valencian region I.C. 95% / 95% C.I. - - 0-7.02 0-33.70 - -
Iles Baléares / Taux / Rates 6.32 2.77 1.12 0.25 10.34 0.07
Balearic Isles I.C. 95% / 95% C.I. 0-42.99 0-26.17 0-2.8 0-13.11 0-127.40 0-0.79
Catalogne / Taux / Rates 2.03 0.41 2.07 1.69 1.34 -
Catalonia I.C. 95% / 95% C.I. 0-21.61 0-21.40 0-7.46 0-9.39 0-70.04 -
Iles Canaries / Taux / Rates 24.77 11.83 3.54 36.77 - 0.30
Canary Islands I.C. 95% / 95% C.I. 0-233.81 0-321.87 0-18.11 0-84.50 - 0-8.254
Andalousie / Taux / Rates 1.06 1.05 0.92 0.05 - 0.05
Andalusia I.C. 95% / 95% C.I. 0-28.96 0-28.73 0-3.88 0-1.50 - 0-2.71

The crude rates showed that tourists visiting the Balearic Isles and the Valencian Region were the most greatly affected. As above, adjusted rates could not be calculated .

The high proportion of hotels associated with more than one case plus the appearance in many such establishments of patients in successive years suggested that their control measures are inadequate.


The results obtained highlight the need for EWGLI to adopt stricter case reporting criteria, especially with regard to the dates when cases stayed at particular places of accommodation, and dates of onset of symptoms, and the compatibility of those dates with the incubation period of legionellosis. Countries where the disease is diagnosed need to investigate risk factors more thoroughly and forward detailed information to the countries in which cases have travelled. Countries associated with the appearance of cases should monitor the maintenance of control measures over time, particularly in hotels repeatedly associated with cases. ELSS program records should be updated continually, by filing epidemiological and environmental data of interest. We would argue that all accommodation shown by subsequent investigation to be irrelevant to the development of the disease should be deleted from the registry. Similarly, if a patient has stayed at several establishments and environmental studies enable the case to be linked with just one of these, the establishments no longer under suspicion should be removed from the register.


1. Joseph CA, Hutchinson EJ, Dedman D, Birtles RJ, Watson JM, Bartlett CLR. Legionnaires' disease surveillance: England and Wales 1994. Commun Dis Rep CDR Rev 1995; 5: R180-3.
2. European Working Group of Legionella Infections. User guide for operating the computer database of the European Surveillance Scheme for Travel Associated Legionnaires' Disease. London: PHLS, 1993.
3. Encuesta sobre la Estructura de los Establecimientos Hoteleros. Madrid: Instituto Nacional de Estadística, 1992.
4. Movimiento del Viajero en Establecimientos Turísticos. Resumen Anual. Madrid: Instituto Nacional de Estadística, 1991.
5. Yu VL. Legionella pneumophila. In: Mandell GL, Douglas JE, Dolin R, eds. Enfermedades infecciosas. Principios y Prácticas. 3ª Ed. New York; Churchill Livingstone, 1990: 1865-1876.
6. Hutchinson EJ, Joseph CA, Bartlett CLR on behalf of EWGLI. EWGLI: a European surveillance scheme for travel associated legionnaires' disease. Eurosurveillance 1996; 1 (5): 37-9.

Back to Table of Contents
en es fr pt

The publisher’s policy on data collection and use of cookies.

Disclaimer: The opinions expressed by authors contributing to Eurosurveillance do not necessarily reflect the opinions of the European Centre for Disease Prevention and Control (ECDC) or the editorial team or the institutions with which the authors are affiliated. Neither ECDC nor any person acting on behalf of ECDC is responsible for the use that might be made of the information in this journal. The information provided on the Eurosurveillance site is designed to support, not replace, the relationship that exists between a patient/site visitor and his/her physician. Our website does not host any form of commercial advertisement. Except where otherwise stated, all manuscripts published after 1 January 2016 will be published under the Creative Commons Attribution (CC BY) licence. You are free to share and adapt the material, but you must give appropriate credit, provide a link to the licence, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.

Eurosurveillance [ISSN] - ©2007-2016. All rights reserved

This website is certified by Health On the Net Foundation. Click to verify. This site complies with the HONcode standard for trustworthy health information:
verify here.