1
Colloquium Abstract
An Assessment of the Impact of Climate Features on Dengue Fever and its Vectors in Five Caribbean Countries
SC Rawlins 1,,Caribbean Epidemiology Centre(CAREC), FederationPark, Port of Spain, Trinidad1
Introduction
In the last two decades there has been an unprecedented increase in the occurrence and severity of dengue fever (DF) in Caribbean countries (Fig.1). Certainly, any tool useful to predicting outbreaks and implementing enhanced prevention strategies – adaptation – would be most welcome. But first, one must determine what might be the principal causes of these increasingly frequent epidemics.
Dengue outbreaks may be multifactoral, involving social, biological and environmental factors such as poor sanitation, proliferation of water-bearing containers – encouraging increased vector production. But climate factors such as precipitation and temperature may also play a role in enhancing transmission of dengue fever. Hales et al (2) have linked the transmission of dengue fever to environmental features such as temperature and precipitation factors. Rising ambient temperatures may reduce the length of time for mosquito larval development, and thus increase the number of generations of the vector in any one year (3). At the same time, the extrinsic incubation period (EIP) – the time taken for viral replication in the mosquito host - may shorten with higher temperatures (4). Koopman et al (5) reported that decreasing the EIP from its normal 12 days by five days may lead to a threefold higher transmission rate of dengue. Precipitation may affect the abundance of vector habitats available for mosquito production
Fig.1
Already, we have examined the impact of climate change features especially warming conditions on retrospective dengue fever data in the Caribbean region. Thus Rawlins et al (2004) and Amarkoon et al (2004) have shown some association in the increased occurrence of DF and warming periods in some Caribbean Epidemiology Centre (CAREC) Member Countries (CMCs) – Fig. 2.
In this present study we are examining prospectively the occurrence of DF and population features of the vector with varying climate features such as temperature and precipitation. Here, we are examining the patterns of reported dengue fever cases in the Caribbean region – mainly CMCs with a population of just over 6 - 7 millions – to demonstrate any climate-related patterns of disease presentation over 2003 - 2004.
Fig. 2. The Caribbean Region
At the same time, we are reviewing dengue patterns in five CMCs with varying histories of disease occurrence, to try to examine climate or seasonal influence on disease prevalence. It is hoped that any emerging patterns might prove valuable for prediction of future outbreaks which may be useful for adaptation to reduce the population’s vulnerability to the disease and its vector.
Objective: To determine the extent of association between climate features and the incidence of dengue fever and its vector indices across five Caribbean countries.
Design and Methods: Data on reported monthly dengue fever (DF) cases, vector – Aedes aegypti – indices, and climatic indicators were collected prospectively for a 12 month period beginning early 2003 in 5 Caribbean countries – Jamaica, Trinidad and Tobago (T&T), Barbados, St Vincent and the Grenadines (SVG) and St Kitts/Nevis (SKN). Data were analyzed by ANOVA for evidence of climate impact on DF cases and vector indices.
Results:
SVG data showed minor fluctuations of vector breteau (BIs) and house indices (HIs) with precipitation. Dryer periods coincided with low vectorHIs of 9-22%. In wet periods, both indices were higher – HIs (17-72%) and BIs (31-55). Temperature varied slightly (23-32˚C) throughout the year (Fig.3).
SKN vector data showed a bimodal pattern with peaks of HIs and BIs in May-June and October-December, the latter associated with significant precipitation. Mean monthly temperatures only varied between 26 -29˚C (Fig. 4).
Fig. 3 Figure 2-:St. Vincent and the Grenadines-Prospective vector indices and climate
.Fig. 4. Vector indices and climate indicators in St Kitts, 2003.
T&T data for 2002 and 2003 showed distinct dry and wet seasonal patterns. In 2002 dry periods (January-May), there were low BIs (21– 29), and mean monthly DF cases of 290.4. Increases in BIs (32-44), (P=0.000) and mean monthly DF cases (695/month) coincided with the wetter period (June – December). Temperature varied slightly 22-25˚C (min) to 31 -33˚C (max) for the period. The pattern was similar in 2003, but with fewer DF cases, (P=0.016) (Fig. 5).
Fig. 5.Trinidad Climate Dengue and Vector data
Jamaica: Of the six zones studied in the Kingston/St Andrew area, there emerged a pattern of 2 peaks of BIs: April- May (18-29) and September –December (13-16). (Fig.6).
Figure 6. Breteau Indices of Kingston (Jamaica) Zones 1-6, 2003-2004
Barbados: DF cases were reported all year but with a small peak (17.5 %) in January, and 73% of cases reported in the wet season, August – December..
The number of DF cases for SVG, SKN and Jamaica for this year were not sufficient for meaningful analysis.
Conclusions
Patterns from SKN, SVG and T&T data discerned a correlation of seasonal effect on BIs and in T&T on DF transmission. Temperature variations did not seem to be significant, but may have affected the onset of precipitation and in turn, vector production, an increase in indices and DF cases. The data confirm the usefulness of recognizing the wet season as a risk factor for DF transmission in the region. Utilising this knowledge for preparation of early warning systems for DF prevention will be vital.