Introduction
Spain is among the European Union (EU) countries that receive the largest
number of tourists. It is estimated that 30 million tourists (1) visit
Spain every year (the population of Spain is 40 million). In Spain, guidelines
for the prevention and control of legionnaires´ disease (LD) have
existed since the 1980's. Some of these guidelines have been developed
by the health authorities in the tourist areas and are specifically targeted
at the prevention of the illness in the accommodation sites. In 2003,
national law was passed in Spain (2) that targeted installations that
could be possible sources of infection and which included both preventive
and control measures should cases or outbreaks appear.
Methods
Information received through EWGLINET, whose functions have been described
elsewhere (3), and notifications of cases and clusters of legionellosis
received by the National Epidemiology Centre (NEC) through the National
Epidemiological Surveillance Network (NESN) have been analysed. Reporting
of legionnaires' disease by physicians is mandatory in Spain . The NESN
gathers, on a weekly basis, any cases of legionnaires' diagnosed and
reported by physicians, accompanied by a minimum dataset that includes
demographic, clinical and epidemiological information about the case.
When a cluster or outbreak occurs, the health authorities of the affected
region send the NEC a report which includes a summary of the epidemiological
and environmental investigation carried out.
Spain has adopted the 2001 definitions of cluster and sporadic cases
(4) and the procedures for the notification and follow-up of clusters
in foreign travellers set down by the EWGLINET Guidelines for the Prevention
and Control of Legionellosis (5). Travel is defined as staying away
from home for one or more nights in accommodation such as hotels, campsites
etc., in the 10 days before the onset of illness.
In this paper, the combined analysis of the clusters associated with
travel, including foreign cases as well as national travellers from
the year 2001 to the end of July 2003, is presented.
Results
From January 2001 to July 2003, 46 clusters were notified and 135
people were affected; 74 were foreign travellers and 61 were Spanish
(table). Twenty six out of the 46 clusters included only foreign citizens,
14 included Spanish and six had cases of both origins. Fifty cases were
citizens of the United Kingdom, Netherlands and Sweden. These cases
accounted for 69% (51 out of 74) of cases of foreign origin.
The mean age was 62 years (ranging from 25 to 89). No differences were
observed between foreign and national travellers in this respect. The
male-female ratio was 2.4/1 in nationals and 3.3/1 in foreign travellers.
All but 11 cases were diagnosed by urinary antigen test. Four cases
were diagnosed by culture. Seven cases were probable cases.
Ten deaths were recorded, however the outcome information was not available
for 76 (56.3%) cases. The case fatality rate was higher for those who
stayed in Apartments or Apart-hotels, but the numbers are too small
to draw any conclusion.
The clusters detected were small, ranging in size from two or three
cases (37 out of the 46 clusters) to eight cases (2 out of the 46).
Most clusters were related to hotels (31 out of 46). Clusters at other
accommodation sites were less frequent: seven clusters were related
to apartments, three to campsites and five to spa resorts. Only Spanish
people were affected in these spa resorts due to the fact that these
are not located in tourist areas. The mean length of stay in the accommodation
sites was 8.5 days (0 to 61 days). This figure was 9.1 days for foreigners
(1 to 61 days) and 7.3 days for Spanish (1 to 23 days). Regarding the
duration of the cluster (time between the first and the last case notified)
this was less than six months in 20 of the clusters (43.5%).
The mean time between the onset and notification by EWGLINET for the
clusters of foreign travellers was 47 days (range was from 7 to 450
days). A reduction in this mean time was observed in 2003 compared to
2001 (26 days versus 65).
Accommodation sites related to the clusters are located on the mainland
Mediterranean coast and the Spanish islands of the Mediterranean and
Atlantic (figure). Clusters associated with both foreign and Spanish
travellers demonstrated a similar location pattern. Only eight clusters
were located outside of the main Spanish tourist areas.
The microbiological results of the environmental investigation were
positive for 25 clusters (54% of the total). The reports stated that
the microorganism was L. pneumophila serogroup 1 for thirteen of these
positive results. No differences were observed in the percentage of
positive results between clusters that lasted less than six months and
longer clusters. No differences were observed in this regard when comparing
clusters of foreigners to those including only Spanish travellers. With
respect to size, in those clusters with two or three cases, the percentage
of positive results was approximately 50%. All results were positive
in those clusters which had four cases or more except in one cluster
of eight cases.
However, only detailed information about the inspection carried out
at the accommodation site for 25 clusters was received. The most frequent
deficiencies were related to incorrect temperatures, both in cold and
hot water systems, followed by inadequate chlorination. In one hotel
the guests were moved out.
Discussion
The adoption in January 2001 of the new definition for clusters signified
an increase of 37% (17 out of 46) in the number of travel associated
clusters related to Spain. In addition, the introduction of new procedures
in July 2002 has resulted in a significant increase in the burden of
work at all levels in the process of notification, investigation and
control. However, these changes have been adopted smoothly by the Spanish
health authorities. Up to the present, EWGLINET forms A and B have been
sent in promptly for all the accommodation sites inspected and all the
sites have complied satisfactorily with the control measures required
by the environmental health officers.
One limitation observed was that Spanish regional health authorities
have replaced the previous more comprehensive reports of the environmental
results of the investigation with just the forms A and B. This change
has caused a significant loss of information.
The difference between the number of clusters of Spanish and foreign
travellers may be accounted for by the type of holidays and accommodation
sites used. No other major differences were observed. The small number
of clusters associated with campsites merits further investigation.
It is worthy of mention that 13 clusters 40.6% (13 out of 32) were composed
of citizens of different countries and that these clusters would never
have been identified without the work of EWGLINET. Also, the improvement
in the reduction of the time delay for notification has to be mentioned.
The increasing use of the urinary antigen test has impaired the possibility
of comparing clinical and environmental isolates. A weakness in the
EWGLINET procedures, which could be considered for future modifications,
is the limited epidemiological information that is presently collected.
More detailed information related to risk exposure would help with the
investigation of clusters in the country of origin of the infection.
|