| Introduction
For many years, Portuguese local health authorities (Delegado de Saúde
Concelhio - DSC) have been responsible for foodborne outbreak investigations,
as part of their legal duties in the field of surveillance and control
of communicable diseases in the community (1). However, written reports
were not mandatory and had no standard format. Foodborne outbreaks (FBO)
were not reportable events although some individual cases were due to
causative agents that were part of the list of statutory reportable
diseases (Doençãs de declaração obligatória
- DDO) (e.g. botulism, brucellosis, salmonellosis). In October 2001,
FBOs were included in the Sistema de Alerta e Resposta Apropriada (SARA).
DSCs were given the responsibility to report FBOs to regional health
authorities and a standard report form was proposed (2). Having adapted
this for Portugal, the northern regional health authority (Delegado
Regional de Saúde do Norte - DRSN) created a formal surveillance
programme of FBOs.
This report briefly describes data generated from this programme, in
the first year of its existence, in the northern health region (Região
de Saúde do Norte) (3.23 million inhabitants) in Portugal.
Material and methods
Initial information on individual cases and/or clustering of cases
was provided to the DSC and/or the DRSN by different sources (hospitals,
DDO, etc). A preliminary assessment was then made in order to confirm
that a FBO had occurred. As part of the above mentioned programme, when
a DSC (at the municipal level) knew of a FBO, they reported it to the
DRSN. That report was first made by phone and letter in a specific written
format. Some FBOs were detected initially by the DRSN, who then contacted
the DSC. The DRSN provided technical advice and guidance in the discussions
with a DSC investigating an outbreak. Ultimately, the DSC was the coordinator
of each outbreak investigation, except when outbreaks involved several
municipalities, in which case coordination was by the DRSN. For each
outbreak, a final written report was produced. After analysis of each
report, feedback was provided to the DSC in the region concerned.
These reports are the source of the data described here. Information
was recorded, processed and analysed, using Epi Info 6.04 (3).
We describe here data concerning operational issues of the programme
and some epidemiological aspects of foodborne outbreaks in northern
Portugal in 2002.
Results
Twenty seven FBOs were studied by DSCs and/or the DRSN, during 2002.
Initial information concerning cases of a suspected FBO was provided
by different entities: hospitals (16/27), senior staff members of institutions
where the FBO had occurred (4/27), DDO (4/27) and other sources (3/27).
The proportion of index cases reported within 72 hours, one week and
two weeks after the date of onset were respectively 59% (16/27), 74%
(20/27) and 100% (27/27) of the outbreaks (Figure 1).
In one of the reports, no data was given concerning the number of persons
affected. In 26 FBOs, 577 people became ill, resulting in an estimated
incidence rate of 17.9 per 100 000, in northern Portugal. The size of
the outbreaks varied from a minimum of one case (botulism) to a maximum
of 154 cases (mean number of persons per outbreak: 22.2). Forty seven
per cent of patients were between 15 and 59 years old, 30% were less
than 15 years and 23% were more than 59 years old. Nine point six per
cent of the patients were admitted to hospital. No deaths were reported.
Whenever available, suspected food items were analysed in the laboratory,
in order to identify the agent and the vehicle of infection. Thirty
five food item samples, from 44.4% (12/27) of the outbreaks, were sent
to be analysed. Laboratory investigations which aimed to isolate aetiological
agents were performed among patients in 81.5% FBOs (22/27) (Table 1).
As a result, the aetiological agent was identified from patients in
63.0% (17/27) and from food items from 18.5% (5/27) of FBOs. In three
FBOs there laboratory evidence was obtained from both patients and food
items (Table 1).
Combining laboratory evidence from patients and food items (Table 1
and 2) it was possible to isolate a putative agent in 70.4% of outbreaks
(19/27), while aetiology was unknown in 22.2% (6/27). Based on clinical
and epidemiological data it was possible to presume a causative agent
in 7.4% (2/27) of the FBOs studied (Table 2). In the FBOs in which laboratory
results were negative both in patients and food items, it was suspected,
based on epidemiological and clinical data, that the aetiological agent
was Norovirus (Tables 1 and 2). In one of the two FBO for which no laboratory
analyses were performed at all (Table 1), the aetiological agent (diarrhoeic
shellfish poisoning toxin) and the vehicle (shellfish) were presumed
based on clinical and epidemiological data (Table 2).
For those in which the aetiology was confirmed, Salmonella enterica
was responsible for 73.7% (14/19) of the outbreaks and for 80.6% (286/355)
of the cases (Table 2). Serovar Enteritidis was identified in 4 of the
14 outbreaks, and phage types 1 (PT1) and 4 (PT4) were found in two
of those four outbreaks.
Based on epidemiological evidence (results from the analysis of questionnaires),
raw eggs and foods containing raw egg were identified as the vehicle
in 8 of the 27 outbreaks, followed by meat and meat products (3/27),
fish products (2/27), smoked raw ham (2/27), shellfish (2/27) and drinking
water (1/27).
In 24 FBOs it was possible to know the place where the meals had been
prepared. In most cases (41.7%, 10/24) meals were prepared in restaurants,
followed by private homes (33.3%: 8/24) and canteens (25.0%: 6/24).
Thirty one contributing factors were reported by DSCs, in 13 of the
27 FBOs. Inadequate processing, preparing or handling of food were factors
more often (7/31) reported, followed by contamination of drinking water
(5/31), use of contaminated raw material (5/31), preparation of food
items too far in advance (4/31), contamination by personnel (3/31) and
inadequate cooking (3/31).
In 18 of the 27 reports (Table 3) we had information about the control
measures implemented by DSCs in order to prevent further FBOs. Besides
health services, other state departments, with responsibilities in the
areas of environment and economy were contacted by DSCs to be involved
in the control measures, in sixteen FBOs.
Discussion
2002 was the first year of the surveillance programme of foodborne outbreaks
in northern Portugal. Because of this, the number of FBOs studied was
small and data was missing for some variables and observations and thus
conclusions drawn from this study must be cautiously interpreted because
of potential biases.
In this review, declaration of FBOs to health authorities were made
sooner than in a published study in France (4), in which 48% and 68%
of FBOs were reported to Health Authorities within three and seven days
after onset respectively. We question if timeliness was influenced by
the fact that 2002 was the first year of the programme. Our main source
of declaration were hospital doctors (59%), a higher value than reported
in France (28%) (4). Reasons for this difference are not apparent.
The estimated incidence rate of foodborne disease in northern Portugal
(17.9/100000) was between the extremes of European values observed in
1998 (the Russian Federation (3/100 000) and Yugoslavia (219/100 000))
(5). It is believed that only 10% of FBOs in industrialised countries
(6) are reported, but no data are available to estimate the level of
underreporting in our study.
The average number of persons per outbreak in this study (22.2) was
higher than that estimated from data reported by the World Health Organization
between 1993 and 1998 (11.7) (5). The high value found in this study
was influenced by the size of one of the outbreaks (n=154) and possibly
by under-reporting of smaller size outbreaks.
The proportion of patients hospitalised as a consequence of a foodborne
disease in this report (9.6%) is similar to the value reported in France
in 2001 (10%) (4). But, unlike the French study, no deaths were reported
among cases of foodborne disease in our case series. These differences
must be interpreted with caution, because this study included a smaller
number of cases and other data must be available to make a proper comparison
of severity among countries.
The aetiological agent was unknown in 22.2% of the FBOs studied, which
is between the values of 17% and 29.3% found in other studies (Table
4). Comparison with FBOs in France, in 2001 (4), the agent was confirmed
in a higher proportion of outbreaks, and presumed in a lower proportion
of FBOs (Table 4). As in other studies (4,5) Salmonella enterica was
the most common isolated agent (7), and raw eggs and raw egg-containing
foods were found to be important vehicles of agents of FBOs.
Foodborne outbreaks originating from meals prepared at home (33.3%)
were less common than in similar studies in European countries (4,5).
Based on this first year of experience we believe that epidemiological
surveillance and control of foodborne disease outbreaks must be pursued
and reinforced in Portugal. The type of program described here is one
of the important activities of a wider strategy to promote food safety
(8).
Acknowledgements
We would like to thank the Delegados de Saúde Concelhios (DSC)
for their important contribution to this report.
|