88 / 218
[
] 88
lance system, which includes Direcção Geral de Saúde
(National Directorate for Health) and Serviço Nacional de
Bombeiros e Protecção Civil (National Service for Civil
Protection) receive a warning in order to enable action to miti-
gate the possible severe consequences of the phenomenon.
Recent heat waves in Portugal
In recent years the population of Portugal’s mainland has been
exposed to several heat waves associated with large excesses
of mortality.
Four heat episodes with a large impact on mortality occurred
in June 1981 with estimated excess deaths of approximately
1,900.
16
Also in July 1991, there were estimated excess deaths
of approximately 1,000.
17
In July/August 2003 estimated
excess deaths were approximately 1,953,
18
and in July 2006
estimated excess deaths were approximately 1,123.
19
Heat episodes with a lower impact on mortality occurred in
Portugal during the following periods: 14-25 July 1990 (esti-
mated excess deaths 690); 19-28 May 1991 (estimated excess
deaths 475); 27 May to 2 June 2001 (estimated excess deaths
441)
20
; at regional level, in Algarve 27 July to 4 August 2004
(estimated excess deaths 80)
21
, 3-7 and 12-16 August 2005
(estimated excess deaths 462)
22
and 4-8 August 2006 (esti-
mated excess deaths 136).
23
The table above shows the calculations leading to the estima-
tion of the number of excess deaths for the two most recent, severe
heat waves. The observed deaths were estimated through a system
that surveys daily mortality. The number of expected deaths was
based on figures from an appropriate comparison period.
The number of excess deaths was then calculated by the
difference ‘observed’ minus ‘expected’ (O-E) and an estimation
of the relative excess was obtained by the ratio ‘observed’
divided by ‘expected’ (O/E).
The heat wave
24
of 29 July-14 August 2003 was a rare and
unprecedented event in terms of both unusually high
maximum and minimum temperatures, and accompanying low
relative humidity. Occurring in the inland territory with a dura-
tion of 17 days, it was the longest recorded heat wave in
Portugal since 1941.
The highest values for both maximum and minimum
temperatures were exceeded in this period. In Amareleja on 1
August, Portugal’s highest ever temperature was measured at
47.3 degrees Celsius. In 97 per cent of meteorological stations,
maximum air temperatures above 35 degrees Celsius were
recorded between 1 and 14 August; maximum air tempera-
tures above 40 degrees Celsius were recorded in two out of
three stations. As to minimum air temperature, 97 per cent of
meteorological stations recorded temperatures above 20
degrees Celsius and 40 per cent above 25 degrees Celsius.
Reference should be made to the heat wave that occurred
between 7 and 18 July 2006. Occurring in the Alentejo region,
In Japan, a study made by Nakai, Itoh and Morimoto,
7
which
investigated heat-related deaths from 1968 through 1994,
concluded that such fatalities were most likely to occur on days
with a peak daily temperature above 38 degrees Celsius, and
that the incidence of these deaths showed an exponential
dependence on the number of hot days. Thus, even a small rise
in atmospheric temperature may lead to a considerable increase
in heat-related mortality. Furthermore, 50.1 per cent of these
deaths occurred in children (four years and under) and the
elderly (70 years and over) irrespective of gender. This clearly
indicates the importance of combating global warming.
Portugal is no exception to this danger. Episodes of excessive
heat have sporadically occurred in Lisbon throughout the 20th
century. The impact of the heat wave of June 1981 on the popu-
lation of the district of Lisbon was first published in 1988.
8
In
fact, multiple episodes of heat waves with different durations
and health consequences have been identified in Portugal
between 1980 and 2006.
In terms of vulnerability, advanced age; cognitive limitations;
existing illness; the consumption of certain medication; hydra-
tion level; isolation and habitation conditions are all relevant.
9
Case-control studies carried out in France found that loss of
autonomy and social isolation played a major role in the risk
factors for the elderly, as did living directly below the roof of
a building, particularly in cities.
10
Whilst there is general agreement that the elderly are most
vulnerable to severe heat impacts, there are cases in which all
age groups have been affected, such as the heat wave of June
1981 in Portugal.
11
Forecasting the effects on mortality
Most European countries have implemented heat wave surveil-
lance and alert systems.
12
However, time of occurrence and
the expected consequences of these heat waves in terms of
duration, intensity and populations affected, are difficult to
estimate precisely.
13
With knowledge based mainly on the incidents of 1981 and
1991, a warning system for heat waves and their effects on
mortality
14
was developed by the National Observatory of Health
of the National Institute of Health, Dr Ricardo Jorge and the
Institute of Meteorology, and has been operational since 1999.
This warning system, the ÍCARO surveillance system of heat
waves, was implemented using a statistical model for the rela-
tionship between high temperatures and mortality in Lisbon.
15
The ÍCARO surveillance system operates yearly from May
to September and calculates an index – the ICARUS INDEX –
that relates predicted deaths (based on three-day forecasts for
air temperature provided by the Institute of Meteorology) due
to the occurrence of high temperatures to those expected
without such climate effect. Whenever a heat wave is predicted
within the next three days, institutions involved in the surveil-
Expected and excess of deaths (95% confidence intervals) during the heat waves of 2003 and 2006 in the Portugal Mainland
Heat waves
Comparison period Expected deaths
Excess of deaths
CI 95% for the excess of deaths Ratio observed/expected
30thJuly to 15th
2000-2001
4499.3
1952.7
(1866.1 ; 2039.3)
1.43
August 2003
7th-17thJuly 2006
July 2004
4163.9
1122.5
(876.3; 1381.0)
1.27
Source: Instituto Ricardo Jorge