| Introduction Albania, a country
with 3.5 million inhabitants, is facing an economic and social crisis. The average per
capita income is less than US$ 1000 per year. Since 1995 unemployment has increased by
2.7% (1). Poverty and migration are major constraints of the society. The country is
divided into 37 administrative districts. Primary health care is delivered through a
network of 1500 health posts (staffed by nurses or midwives) and 602 health centres
(staffed by family practitioners). Secondary health care is served by 53 polyclinics in
district towns, 34 district hospitals, and 10 regional hospitals. Tertiary care is
provided only in Tirana, at the University Hospital Centre (2). The routine national
surveillance system monitors 73 diseases: medical health centres and health posts report
to district epidemiologists, who send information each month to the Institute of Public
Health (IPH) in Tirana. The quality of surveillance is limited by the reliability of data,
underreporting, data quality, data transmission, and feedback (3).
In September 1994 more than a hundred cholera cases were
reported in the country (4); from April to September 1996 an outbreak of paralytic
poliomyelitis in the northern and central parts of the county resulted in 66 cases of
acute flaccid paralysis (AFP) (5). In 1998, the commonest diseases reported were
influenza, unspecified gastroenteritis, common cold, scabies, and hepatitis (6). Hepatitis
is endemic throughout the country: 3139 suspected cases, 260 confirmed cases of hepatitis
A, and 52 confirmed cases of hepatitis B were reported in 1998. The measles immunisation
coverage reported by the Ministry of Health based on the health facilities monthly report
is 89%, but the incidence of measles in children under 5 years of age was 247 per 100 000
in 1998, one of the highest in Europe (7). Zoonoses are a public health concern. In 1998,
58 cases of anthrax, 69 cases of visceral leishmaniasis, and 336 cases of brucellosis were
reported (6).
From March to June 1999, 442 000 Kosovar refugees arrived in Albania. More than
half of them 228 000 have been hosted in Albanian families (Organisation for Security and
Cooperation in Europe (OSCE) Tirana 1999), the rest being settled in camps and collective
centres. The national surveillance system was unprepared to face this situation. An
emergency surveillance system for communicable diseases was set up to detect and control
potential outbreaks among the refugee population if and when they occurred. We present
here the results of the first nine weeks of surveillance among Kosovar refugees in
Albania.
Methods
The population under surveillance included all the Kosovar refugees
present in Albania. Data provided by OSCE were used to estimate the number of refugees per
district each week. As the refugees were not settled only in camps, but also integrated in
Albanian families, the sources of data for the surveillance included all medical agencies
working in camps and the Albanian health facilities caring for local communities. One
medical agency was in charge of the health services for refugees in each camp or
collective centre.
The health events under surveillance were mainly those that constituted
a hazard for outbreaks in the context of a refugee crisis (diarrhoeas, acute respiratory
infection (ARI), scabies, measles, meningitis) (8). Other entities specific to the
situation, such as psychological disorders and war injuries, were also included. Because
of the lack of reliable data on the refugee population, we used cardiovascular diseases
(defined as any cardiac problem or high pressure) - a health event with stable incidence
of reported cases – as a reference to monitor the trends of the other syndromes. A
category ‘others’ was added to record all the consultations excluded from the
list of reportable health events. A list of case definitions was distributed to the health
facilities.
Health units taking part in the system (those caring for refugees)
reported every week to the district epidemiologist the number of new cases and deaths on a
standard form. These forms were transmitted to the National Institute of Public Health
(IPH) in Tirana by fax or by car.
All health facilities also had to fill in a null report form for
measles, AFP, dysentery, neonatal tetanus and meningitis to confirm that these conditions
had not been seen.
The IPH prepared weekly reports including a summary table of the
consultations for the week, the proportional morbidity (calculated as a percentage of the
total consultations), and specific comments on the epidemiological situation. These
reports were distributed in English to the international agencies and in Albanian to the
district epidemiologists.
Data were analysed using an Epi-Info application (9) developed by the
Institut de Veille Sanitaire. Standard analyses included computation of indicators such as
numbers of cases and deaths per week, number of units reporting, and number of districts
reporting.
In the absence of denominators to calculate incidences, proportional
morbidity was used to follow trends in the commonest syndromes. The number of cases per
week was used as an indicator for the rarer but potentially epidemic syndromes and
diseases such as measles, jaundice, meningitis, and diarrhoea with blood.
Data collection began on the 16 April 1999. The data presented refer to
the surveillance period from 16 April (data from week 14) to 6 June (data from week 22).
Results
By the end of week 22 the refugees were present in all 37 districts.
The major concentrations were in Kukes (18%), Shkoder (9%), and Tirane and Tirane city
(22%). The other 33 districts accounted for 51% (map).
The number of health units reporting increased rapidly in the first
five weeks (from 3 to 132) and then remained stable (range 162-181). Thirty-two of the 37
districts took part in the system. The districts that never notified represented 3% of the
refugee population present in week 22 (map).
Nine hundred and twenty reports were received at IPH, corresponding to
189 706 consultations. The Albanian health facilities reported 30% of the cases (56 537).
Conditions in the category ‘others’ accounted for 40% of consultations of
children under 5 and 50% of consultations of people aged 5 years or more. Among children
under 5 years, the commonest reason for consultation was acute respiratory infection
(37%), followed by diarrhoea without blood (17%), and scabies and lice (3%). In the age
group 5 years and over, acute respiratory infection (ARI) accounted for 24% of
consultations, cardiovascular diseases for 10%, diarrhoea without blood 6.7%, and scabies
and lice 3.7%. Consultations for war injuries and severe psychological disorders were
commoner in the older age group (1.3% and 2.8% respectively), than in children under 5
(0.3% for both conditions).
Eleven deaths were reported among children under 5 and 23 in the
population 5 years and over. In children under 5 years, six deaths were attributed to
‘other’ conditions, and three to ARI. In children under 5 years 11 deaths were
from cardiovascular diseases, eight from ‘other’ conditions, and three from ARI.
After the first week of surveillance, the proportional morbidity for
the commonest syndromes remained stable (figures1-2) and no outbreaks were detected.
Thirty-nine suspected cases of measles were reported in children
under 5 and 31 in the older age group. The numbers of cases increased in the first three
weeks. The largest number of measles cases (16) was reported in week 19 (figure 3).
A total of 107 cases of diarrhoea with blood were reported during the period
studied. Kukes district accounted for 47 % of the cases (50 cases) but the overall
proportional morbidity (0.1%) was no higher than elsewhere in the country (figure 4).
No cases of AFP, tetanus or cholera were
reported through the zero report form.
Actions
During the nine weeks of surveillance some clusters of suspected cases
of diarrhoea with blood, hepatitis, and measles were detected in specific refugee
settlements. Active case finding and visits to the field were conducted to assess the
situation and to take appropriate control measures.
In response to the measles cases notified in Kukes in the first weeks
of data collection an accelerated immunisation campaign took place among Kosovar and
Albanian children in Kukes and Has districts from 23 to 28 April. The immunisation
coverage estimated after the campaign through a survey was 80.9% (95% confidence interval
(CI) 74.0-87.9) among Albanians and 90.1% (95% CI 86.0-94.3) among Kosovars (10). The
increasing number of reports received could have contributed to the peak in week 19, but
the number of units reporting began to increase in week 18 and continued until week 20
with no increase in the number of measles cases reported. After this peak, the IPH asked
the district epidemiologists to conduct active case finding. A cluster of four cases was
investigated in a collective centre in Laç district in week 21: all the children of the
collective centre, including the four cases, had been immunised a month before. No more
cases were reported in the following weeks.
From week 18 to week 20, 18 cases of diarrhoea with blood were reported
in Kukes. Eight cases of shigellosis were confirmed among whom seven were positive for Shigella
flexneri and one for S. sonnei. An individual notification form for diarrhoea
with blood was distributed to all health facilities in the district. In week 21, the local
laboratory performed stool cultures from 13 cases of diarrhoea with blood, only one of
which was positive for S. flexneri. The laboratory in Kukes was supplied with tubes
and transport medium in order to send faecal specimens to the IPH central laboratory in
Tirana to perform a quality control test.
Discussion
During nine weeks of surveillance (from 16 April to 6 June 1999), no
outbreaks occurred among Kosovar refugees in Albania. This could be explained because
their health was generally good, they had not been displaced for very long, and the
hygiene in the camps and in the homes of host families was acceptable.
A syndrome-based surveillance system was set up for the emergency
situation. Its high sensitivity led to several false alerts, but the system was designed
to facilitate early warning of epidemics in order to control the spread of the diseases in
this vulnerable population. Medical staff found the system acceptable because of its
simplicity: only diseases with the potential for producing epidemics with high morbidity
and mortality were monitored.
The distribution of the refugees all over the country and their
accommodation into local families made it difficult to collect demographic data,
especially by age or mode of housing. Thus incidence could not be calculated. Following
the trends of the syndromes through proportional morbidity or absolute numbers of cases
may cause bias, leading to under- or overestimation. Once the situation is
stabilised, an attempt to calculate incidence should be undertaken. In Albania, with the
rapid evolution of the crisis and the return of refugees to Kosovo, this will not be
possible.
Few deaths have been reported but they have
to be interpreted with caution because the reports come from outpatient consultations.
Other sources of data (hospital records or specific mortality surveys) should be used to
estimate mortality.
The rapid increase in the numbers of the reporting units in the first
weeks and the participation of both international medical agencies and Albanian health
facilities allow us to suppose that we may achieve comprehensive reporting. Difficulty in
obtaining the number of Albanian health facilities directly involved in the health care of
refugees and in monitoring the international medical agencies working in the country makes
it hard to estimate the representativeness of the system.
This emergency surveillance system has been unique in enabling Albanian
health facilities and local epidemiologists to work together. It was a challenge to
integrate collaborating services and individuals: the implementation of the system has
brought about close collaboration and exchange between international agencies and Albanian
health authorities. The participation of international agencies in the emergency has
provided a stimulus for the improvement of the overall national epidemiological
surveillance system, and may contribute to an improvement in the health system in Albania.
* EPIET, a programme funded by DGV of the commission of European Communities.
Update: Immediately after the signature of the peace accord (10 June), the return
of refugees to Kosovo began despite that UNHCR and NATO tried to delay this return until
the first week of July. In July, a repatriation plan started. By 8 July, 362 812 refugees
had already returned to Kosovo from Albania (Emergency Management Group, Tirana) and 116
411 were still in the country. Many camps have closed and international agencies have left
Albania to start working in Kosovo.
We continue to carry out the activities related with the emergency surveillance system
even though the population under surveillance is much less and the number of reporting
units has decreased. No outbreaks have been detected.
On the basis of the emergency surveillance system experience and methodology, a
surveillance alert system will be developed in the country for early detection of
outbreaks among the Albanian population. |