Associations of the Dengue Epidemic with Precipitation and Temperature, and Its Seasonality

Dengue Epidemics-its association with Precipitation and Temperature, and its

Seasonality in some Caribbean Countries

AMD Amarakoon, AA Chen, SC Rawlins, MA Taylor

CSGM, Department of Physics, The University of the West Indies, Kingston 7, Jamaica

The University of the West Indies, St. Augustine, Trinidad. 2

ABSTRACT

The variability in the reported cases of dengue was studied in relation to the variability in precipitation and temperature. The attention was focused on Trinidad and Tobago, and Barbados. The data covered the period 1980 to 2000. Annual and monthly/four weekly variability of the reported cases as well as those of the climate parameters (precipitation and temperature) were investigated. Results show that the incidences of dengue in Trinidad and Tobago, and Barbados were higher in the last decade compared to that in the previous decade. The annual patterns of reported cases were nearly periodic and compared closely with the periodicity of ENSO events. Monthly/four weekly variability of the reported cases indicated a well defined seasonality in the epidemics. The observed variability in the precipitation and temperature revealed that there was a tendency toward less abundant rainfall and higher temperatures in the last decade compared to the previous decade. A comparison of the pattern of variation of the reported dengue cases with those of the climate parameters indicated that warmer temperatures and less abundance in rainfall appeared to be influencing the epidemics. ENSO events, temperature and precipitation are predictable, although with inherent inaccuracies. With improved predictability, the observed seasonality could be a useful tool in the design of early warning systems for the prevention and control of dengue epidemics in our communities. 3

Introduction

It is a known fact that millions of people are under the threat of dengue fever (DF) in the tropics and subtropics. Vector-borne diseases such as malaria have been eliminated from the Caribbean, but over the last 22 years (1980-2001), forty three thousand cases of dengue (CAREC, 2002) have been reported. The incidence has been very high in the last decade. About forty one thousand cases have been reported in the last decade, and the indications are that the frequency and the intensity of outbreaks of the epidemic are on the rise. Studies done in the other regions of the globe (Jonathan et al., 1998; Martens et al., 1997; Gagnon et al., 2001; Hales et al., 1966; Poveda et al., 2000; Focks et al., 1995;

Koopman et al., 1991) have associated epidemic outbreaks and the transmission of the disease with climate variability (evidenced by the occurrences of El Niño Southern

Oscillation/ENSO events), through temperature increases and availability of stagnant and stored mosquito-accessible water during droughts as well as after rains. These conditions appear to enhance mosquito breeding and dengue transmission rates. Recently published studies (Peterson and Taylor et al., 2002; Chen and Taylor, 2002), indicate that in the

Caribbean, there is a warming trend, drier than normal conditions with warmer temperatures in the latter half of El Niño years, and increase in precipitation in the early part of the following year. Thus, we can expect an association of dengue outbreaks in the

Caribbean with precipitation and temperature which are indicators of climate variability.

In this paper, we present the nature of the influences of precipitation and temperature on dengue outbreaks in Trinidad and Tobago, and Barbados. We also examine the 4 seasonality of the epidemic in these two countries in the light of precipitation and temperature climatology.

Data and methods

Data on reported cases of dengue were provided by the disease surveillance unit, CAREC in Trinidad, and were derived from reports submitted from the various countries’

Ministry of Health. The data which spanned the period 1980 to 2001 were in the form of annual totals for 1980 to 2001 and also on a 4-week period/monthly basis for 1995 to

2001. The climate data (precipitation and temperature) were obtained from the data depository of the Climate Studies Group Mona at the Department of Physics, the

University of the West Indies, Mona Campus, Jamaica. The climate data available were on a daily or a monthly basis. The data were converted to annual averages or monthly/4-week averages as required for use in the analysis.

Attention was focused on Trinidad and Tobago, and Barbados. Dengue data indicated that the prevalence of dengue in Trinidad and Tobago was very high (about 50% of the reported cases), and in Barbados moderate to high (about 20% of the reported cases).

Also these two countries had fairly well documented long term series of climate data.

Hence they were chosen for this study.

In the analysis, pattern of the variation of annual totals of reported dengue cases and also their annual rates of increase were compared with the annual standardized anomalies

(SA) of precipitation and temperature. The rate of increase mentioned here is equal to the reported cases in a given year minus that in the previous year and provides more features of the pattern of variation of the disease. The period of study in this component was from 5

1980 to 2000. The seasonality of the epidemic was studied by examining the 4-week period/monthly reported cases of dengue for the period 1995 to 1999. Four weekly data sets were used for Trinidad and Tobago while monthly data were used for Barbados. The choice of 4-week periods or monthly was based on the available format of data. However, this choice should not cause a problem as a month is almost a 4-week period. A standardized anomaly (SA) as used in this paper means the following:

(measured value – average value)/standard deviation. In the case of Trinidad and Tobago climate data used were obtained from the meteorological station at Piarco and Barbados climate data were from the meteorological station at Husbands. Even though these data sets did not cover the entire islands, they were good enough to be used as representative samples to understand the trends in the climate associations considering, (i) the geographical locations of the counties/districts that reported dengue cases and the meteorological stations at Piarco and Husbands and (ii) the topography of the study regions. The counties/districts are not too far from the meteorological stations and the variability in the topography across the study regions (Trinidad and Tobago, and

Barbados) is not very significant so as to cause a significant variability in the climate across each region. As examples, although the results are not given in this paper, the precipitation pattern at Piarco in Trinidad is almost the same as that at St. Augustine and the precipitation pattern at Husbands in Barbados is almost the same as those at

Claybury, Edgecumbe and Apes Hill. 6

Results

The results of this study are presented in Figures 1 to 5 and Table 1. Figures 1 and 2 illustrate the variation of annual reported dengue cases in Trinidad and Tobago, and

Barbados respectively. Both figures exhibit higher incidences of dengue in the last decade. Figure 3 illustrates the variation of annual standardized anomalies of precipitation and temperature using observed data from Piarco meteorological station in

Trinidad. Figure 4 illustrates the same but for the Husbands station in Barbados. Both figures show that in the last decade there was also a tendency toward less abundant rainfall (rainfall below normal or close to normal) and higher temperatures in comparison to the previous decade. A comparison of these patterns of anomalies with the epidemic patterns illustrated in Figures 1 and 2 suggests that warmer temperatures and less rainfall appear to be influencing the epidemics. In further support of this, we note from a careful examination of the curves (in Figures 1 and 2) representing the rate of increase, that the annual pattern is nearly periodic. The periodicity was about 2 to 3 years in the last decade and 3 to 4 years in the previous decade. Almost all the peaks can be identified with El

Niño episodes that occurred in 1982/83, 1986/87, 1992/93, 1994/95 (weak El Niño), and

1997/98. The periodicity seen is nearly that of El Niño episodes. Below normal rainfall, particularly in the latter half of the year, and higher temperatures are known associations within the Caribbean with El Niño events.

Figure 5, which is for Barbados, illustrates the typical monthly variability of reported cases, temperature and rainfall. The period covered here is from 1995 to 1999.

Interestingly, a clear seasonal pattern in the dengue cases is visible. The epidemics appeared to occur in the latter part of the year. This was also the case with Trinidad and 7

Tobago and many other countries in the region. Details pertaining to this feature and the nature of the influence of precipitation and temperature are given in Table 1. Note from

Table 1 and Figure 5 that the warming occurs first, then rainfall, and then the dengue epidemic. Thus we see a simple pattern among onset of the epidemic, warming, and precipitation. The epidemic peak actually occurs when the rain is receding with the temperature remaining high until the onset of the epidemic. Also worth noting, from

Figure 5, is the fact that the epidemic is weak (in 1996 and 1998) when rains last for a longer time (broader rainfall peaks). This is the case with Trinidad and Tobago as well.

Perhaps, too much of rain may be washing the mosquito larvae away.

Discussion

We have presented a simple analysis of the association of dengue epidemics with precipitation and temperature. Two countries in the Caribbean region were considered.

Trinidad and Tobago, and Barbados. The results indicate that there is close association of the epidemic patterns with temperature and precipitation. Warmer temperatures and less abundance in rainfall appear to be influencing the epidemics. As warmer temperatures and less abundance in rainfall, especially in the latter half of the year, are associated with

El Niño episodes, we can expect a higher probability of dengue epidemic occurrences during El Niño periods. From the results it was also clear that the epidemic has a well defined seasonality. ENSO events, temperature and precipitation are predictable, although with inherent inaccuracies. With improved predictability, the observed seasonality could be a useful tool in the design of early warning systems for the prevention and control of dengue epidemics in our communities. 8

References

1. Communications with the disease surveillance unit, CAREC, Port of Spain, Trinidad 2002.

2. Jonathan AP, Martens WJM, Focks DA, Jetten TH. Dengue fever epidemic potential as projected by general circulation models of global climate change. Environmental Health Perspectives 1998; 106: 147-153.

3. Martens WJM, Jetten TH, Focks DA. Sensitivity of malaria, schistosomiasis and dengue to global warming. Climate Change 1997; 35: 145-156.

4. Gagnon AS, Bush ABG, Smoyer-Tomic KE. Dengue epidemics and the El Niño Southern Oscillation. Climate Research 2001; 19: 35-43.

5. Hales S, Weinstein P, Woodward A. Dengue fever in the South Pacific: Driven by El Niño Southern Oscillation. Lancet 1966; 348: 1664-1665.

6. Poveda G, Graham NE, Epstein PR, Rojas w, Quiñones ML, Valez ID, Martens WJM. Climate and ENSO variability associated with vector-borne diseases in Colombia. In: Diaz HF, Markgrtaf V, ed. El Niño and the Southern Oscillation. Cambridge University Press, 496 pp; 2000.

7. Focks DA, Daniels E, Haile DG, Deesling LE. A simulation model of the epidemiology of urban dengue fever: Literature analysis, model development, preliminary validation, and samples of simulation results. American Journal of Tropical Medicine and Hygiene 1995; 53: 489-506.

8. Koopman JS, Prevots DR, Marin MAV, Dantes HG, Aquino MLZ, Longini IM Jr, Amor JS. Determinants and predictors of dengue infection in Mexico. American Journal of Epidemiology 1991; 133: 1168-1178.

9. Peterson TC, Taylor MA, Demeritte R, Duncombe DL, Burton S, Thompson F et al. Recent changes in climate extremes in the Caribbean region. Journal of Geophysical Research 2002; 107(D21), ACL 16: 1-9.

10. Chen AA, Taylor MA. Investigating the link between early season Caribbean rainfall and the El Niño+1 year. International Journal of Climatology 2002; 22: 87-106. 9

Table 1. Seasonality of the Epidemics and the Nature of Association with

Climate Parameters. ______Country Year Epidemic Peak Temperature Precipitation

Peak Peak ______

Trinidad and 1995 August (weak) Apr. to Nov Jun. to Sep.

Tobago 1996 September (strong) Apr. to Dec. May to Oct.

1997 December (strong) May to Dec. July and Nov.

1998 July to Sep. (strong) March to Nov. May to Sep.

1999 September (weak) Apr. to Dec. Jul. to Oct.

Barbados 1995 October (strong) Apr. to Nov. Jul. to Oct.

1996 September (weak) Apr. to Nov. May to Nov.

1997 November (strong) Apr. to Nov. June to Nov.

1998 Aug. to Sep. (weak) Apr. to Oct. Jul. to Nov.

1999 November (weak) Apr. to Nov. Jun. to Nov.

______

3000 Annual Totals Rate of Increase 2000

0 1 1 1 1 1 1 1 1 1 1 2 2 9 9 9 9 9 9 9 9 9 9 0 0 8 8 8 8 8 9 9 9 9 9 0 0 0 2 4 6 8 0 2 4 6 8 0 2 -1000 Year

-3000 11

2000 Annual Totals 1500 Rate of Increase

0 1 1 1 1 1 1 1 1 1 1 2 2 9 9 9 9 9 9 9 9 9 9 0 0 8 8 8 8 8 9 9 9 9 9 0 0 -500 0 2 4 6 8 0 2 4 6 8 0 2 Year -1000

-2500 12

2.5 Rainfall Anomalies Temperature Anomalies 2

0 1 1 1 1 1 1 1 1 1 1 2 9 9 9 9 9 9 9 9 9 9 0 8 8 8 8 8 9 9 9 9 9 0 -0.5 0 2 4 6 8 0 2 4 6 8 0

-1 Year

-2.5 13

Rainfall Anomalies Temperature Anomalies 2

0 1 1 1 1 1 1 1 1 1 1 1 9 9 9 9 9 9 9 9 9 9 9 7 8 8 8 8 8 9 9 9 9 9 9 1 3 5 7 9 1 3 5 7 9

-1 Year

1200 32

Reported Dengue Cases Monthly Rainfall 1000 31 Monthly Temperature T 30 e 800 m p 29 e 600 r 28 a t 400 u 27 r e 200 26 (ºC)

0 25 Jan-95 Jan-96 Jan-97 Jan-98 Jan-99

-200 24 15

Figure captions

Figure 1. Annual variability of the reported cases of dengue and the rate of increase in Trinidad and Tobago.

Figure 2. Annual variability of the reported cases of dengue and the rate of increase in Barbados.

Figure 3. Annual variability of the standardized rainfall and temperature anomalies in Trinidad. Standardized anomaly = (measured value-average value)/standard deviation

Figure 4. Annual variability of the standardized rainfall and temperature anomalies in Barbados. Standardized anomaly = (measured value-average value)/standard deviation

Figure 5. Monthly variability of the reported dengue cases, rainfall and temperature from 1995 to 1999 in Barbados 16

Acknowledgements

The authors wish to thank Miss Roxanne Stennett for processing some of the climate data used in this work, while she was an undergraduate student. This work is supported by a grant for Project SIS06 from the Assessment of Impacts and Adaptation to Climate

Change (AIACC) initiative, through TWAS and the START Secretariat in New York, and constitutes a part of the study under the retrospective component of the AIACC

SIS06 project.