Transcript Epidemiological patterns of dengue under climate change

Epidemiology Patterns Of
Dengue In The Caribbean
Under Climate Change
A. M. D. Amarakoon**, Anthony A. Chen,
Michael A. Taylor, Rainaldo F. Crosbourne
Climate Studies Group Mona, UWI, Jamaica
Samuel C. Rawlins, Karen Polson
Caribbean Epidemiology Centre, Trinidad &
Tobago
Wilma Bailey, Charmaine Thomas-Heslop
Department of Geography, UWI, Jamaica
[** SPEAKER]
PROJECT: AIACC-SIS06
The Threat of Dengue Fever Assessment of Impacts and
Adaptation to Climate Change in
Human Health in the Caribbean
An AIACC Project at
The University of the West Indies,
Mona and Caribbean Epidemiology
Centre
THE CARIBBEAN
OBJECTIVES
 To determine the extent of the association
between climate and the incidence of
dengue across the Caribbean Region.
 To explore possible adaptation options
The approaches selected to achieve the
objectives:
 Investigate the influence of climate, through temperature and
precipitation, on the epidemics
 Investigate the seasonality (seasonal variability of the
epidemic)
 Investigate the degree of association of dengue epidemics with
ENSO events
 Examine, briefly, some adaptation options.
Previous studies/events that
influenced the selected approaches:
• Hales et al (1996), Poveda et al (2000),
Gagnon et al (2001)
• Koopman et al (1991), Focks et al (1995)
• Ropelewski and Halpart (1996), Chen et al
(1997), Malmgren et al (1998), Taylor (1999),
Chen and Taylor (2001)
• AIACC V & A workshop, Trieste, Italy, June
2002
DATA & METHODOLOGY
•
•
The data acquired for the CCID project by the CSGM provided the bulk of the
climate data: Temperature (maximum, minimum and mean) and Precipitation,
daily or monthly values
CAREC provided the epidemiology data in the form of reported dengue cases
and vector indices, annual, 4-week period, monthly, quarterly values. More
attention was focused on reported dengue cases
• Data analysis: Time series analysis of annual reported cases and their rates
of change, mean temperature, mean precipitation, temperature and
precipitation anomalies; Study of the climatology of temperature, precipitation,
and reported cases; Performance of statistical significance tests (Fisher’s
exact test using suitable contingency tables) for observed correlations,
wherever applicable.
•
•
ENSO year (El Niño & La Niña) classification: NOAA-CDC MEI index
{EN: 1982/83, 1986/87, 1992/93, 1997/98. LN: 1988/89, 1998+/00}
Supplementary: 1994/95
Main study period: 1980 to 2001
MEI
El Niño
La Niña
Caribbean- Reported Cases
10000
CAREC 4-WEEK ACCUMULATION (1995-2001)
8000
6000
4000
2000
Accumulated reported cases
6000
5000
Average 4-week
period accumulation
4000
3000
2000
1000
Jn
D
0
1
2
3
4
5
6
7
8
9
10 11 12 13
4- Week period
0
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
-2000
-4000
En
-6000
En+1
Annual totals
Rate of change
2000
2002
Some Case Studies: T & T
4000
T & T Annual Reported Cases and Rate of Increase
3000
2000
1000
0
1980
1982
1984
1986
1988
1990
Year
-1000
Annual Totals
Rate of Increase
-2000
-3000
1992
1994
1996
1998
2000
2002
Time Series of Temperature Anomalies: 1980 to 2001
2.5
Temp Anomalies (Piarco)
2
1.5
1
0.5
0
2001
2000
1999
1998
1997
1996
-2.5
1995
-2
Temp Anomalies
1994
-1.5
1993
-1
1992
1991
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
-0.5
Time Series of Rainfall Anomalies: 1970 to 2001
3
Rainfall Anomalies (Piarco)
2.5
2
1.5
1
0.5
0
2000
1998
1996
1994
1992
-2
1990
-1.5
1988
Rainfall Anomalies
1986
-1
1984
1982
1980
1978
1976
1974
1972
1970
-0.5
Time Series of Reported Cases & Rainfall (mm) in 4-Week Periods
1100
900
4-Week period reported cases in T & T: 1995 to 1999
4-Week period rainfall: 1995 to 1999
700
500
300
100
-100
95Jan
96Jan
97Jan
98Jan
99Jan
99Dec
MONTHLY VARIATION OF MEAN T (T & T: 1995-1999)
29.5
1995
1996
1997
1998
1999
29
28.5
MEAN T in oC
28
27.5
27
26.5
26
25.5
0
2
4
6
8
MONTH
10
12
14
A sample of Monthly Variability in House Index: Port of Spain City Co-operation
4.5
4
Average House Index: 1996-2001
3.5
3
2.5
2
1.5
1
Average Index: 1996-2001
0.5
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Some Results For Jamaica
Reported Cases
MONTHLY VARIATION OF RAINFALL (JAMAICA: 1993,
1995, 1997 & 1998)
2000
350
1500
1997
1998
1995
1993
300
1000
Rate of incearse
500
0
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
-500
RAINFALL in mm
Totals
250
200
150
100
50
-1000
0
-1500
-50
0
2
4
6
10
12
14
MONTH
-2000
4-WEEK VARIATION OF CASES (JAMAICA: 1995, 1997,
1998 and 2001)
900
31
700
600
500
1995: cases 1778
1997: cases 17
1998: cases 1255
2001: cases 39
Monthly Mean Temperature in C
1993
1995
1997
1998
30
800
REPORTED CASES
8
29
28
400
27
300
200
26
100
0
-100 0
25
5
10
15
4-WEEK PERIODS
Jn
D
20
24
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
4-WEEK VARIATION OF CASES (SURINAM:1995-2001)
4-WEEK VARIATION OF CASES (BAHAMAS: 1998)
1995: cases 129
1996: cases 677
1997: cases 90
1998: cases148
1999: cases 695
2000: cases 1205
2001: cases 760
REPORTED CASES
250
200
150
100
50
300
250
REPORTED CASES
300
200
1998: cases 336
150
100
50
0
0
0
-50
0
5
10
-50
D
15
20
5
Au
10
4-WEEK PERIODS
4-WEEK PERIODS
SE
N
D
15
20
Statistical Significance Level of ENSO
Associations (* with 1994/95)
REGION
El Niño
(N)
El Niño+1 N &N+1 La Niña
(N+1)
Caribbean
(8 Epeds)
T&T
(8 Epeds)
Barbados
(6 Epeds)
Jamaica
(5 Epeds)
88%
64%
64%
88%
74%
74%
53%
80%
(90%)*
92% **
(94%)*
92% **
(80%)*
90% **
(95%)*
79%
(89%)*
-
-
Results Summary
 In general, across the region, 19-nineties are observed to be more prone to
the epidemic than 19-eighties. There is a periodicity of about 4 to 3 years in
the 19-eighties and 3 to 2 years in the 19-nineties with more frequent
outbursts. May be due to the fact that, in the 19-ninetees, temperatures were
warmer and rainfall was less abundant, for example, as indicated by the
anomalies for T & T. These conditions reduce the incubation period and
increase the disease transmission rate.
 The epidemic shows a well defined seasonality over the region. It occurs
in the latter half of the year. The warmer and drier conditions (less
abundance in rainfall) appear to trigger the epidemic with the onset of the
rainfall, which subsequently & speedily develops. Longer spells of less
abundant rainfall and warmer temperatures appear to enhance the
probability of the epidemic.
 There is a tendency for the spread to get narrower, from south east to
north in the region. Perhaps, this may be due to the warmer & moist climate
(tropical warm moist climate, more suitable for vector breeding and
propagation) that persists in the SE, in contrast to the tropical climate with
seasonal rainfall in the central and the nothern part.
 The periodicity seen roughly agrees with the periodicity of ENSOs
SYNOPSIS
 Significance: The work discussed forms a part of the
retrospective component of the AIACC Dengue Project-SIS06.
May be stated that, exciting features of the dengue epidemic &
evidence of climate influence are seen. Namely;
(i) Periodicity & Seasonality.
(ii) The influence of the temperature and rainfall.
(Iii) Significant association with El Niño episodes (N & N+1
together).
We cannot change the Climate Change!
But adaptation measures could be
provided to minimize the impacts
Impacts on Vector
A: Temperature Increase: Increase in numbers, increased
frequency of blood meals, and expanded spatial
distribution including highland areas. Also increases rate
of extrinsic incubation( period lowers)
B: Precipitation: Either increase or decrease in larval
habitats (very heavy rainfall could flush out habitats).
Humidity increase may increase survival. Flooding, and
hence stagnant water, could increase small habitats.
Droughts could result in possible decrease in larval
habitats, but storage of water increases
COMMON SCENARIOS
(POTENTIAL BREEDING PLACES)
Possible Adaptation Options
 Intensify public awareness through
propaganda and education
 Devise early warning systems coupled to
climate forecasts
 Make the public health sector more efficient
and effective on issues concerning vector
borne diseases (vector control, surveillance,
health education)
 If socioeconomic (SE) conditions, attitudes
and practices are contributing factors, make
attempts to improve/change them.
 Examples of Adaptation: Venice
Trip in June 2002
One of the best
Adaptation Options:
“Public Awareness
& Education”
CONSTRAINTS
Time series of dengue data spanned
only 20 years, which limited the number
of ENSO episodes to 4. The influence
(+ve or –ve) introduced by migration
activities taking place, spraying and
serotypes/dengue types have not been
considered. Need to complete a
socioeconomic & a KAP survey, which
presumably will occur soon.