Climate Variability and Dengue Hemorrhagic Fever in Ba Tri District, Ben Tre Province, Vietnam during 2004–2014

Background Currently, dengue fever/dengue hemorrhagic fever (DF/DHF) is an important public health challenge in many areas, including the Ba Tri District, Ben Tre Province, Vietnam. Methods and Aim This study was conducted in 2015 using a retrospective secondary data analysis on monthly data of DF/DHF cases and climate conditions from 2004–2014 in Ba Tri District, which aimed to explore the relationship between DF/DHF and climate variables. Results During the period of 2004–2014, there were 5728 reported DF/DHF cases and five deaths. The disease occurred year round, with peaked from May to October and the highest number of cases occurred in June and July. There were strong correlations between monthly DF/DHF cases within that period with average rainfall (r = 0.70), humidity (r = 0.59), mosquito density (r = 0.82), and Breteau index (r = 0.81). A moderate association was observed between the monthly average number of DF/DHF cases and the average temperature (r = 0.37). The monthly DF/DHF cases were also moderately correlated with the Aedes mosquito density. Conclusions and Recommendations Local health authorities need to monitor DF/DHF cases at the beginning of epidemic period, starting from April and to apply timely disease prevention measures to avoid the spreading of the disease in the following months. More vector control efforts should be implemented in March and April, just before the rainy season, which can help to reduce the vector density and the epidemic risk. A larger scale study using national data and for a longer period of time should be undertaken to thoroughly describe the correlation between climate variability and DF/DHF cases as well as for modeling and building projection model for the disease in the coming years. This can play an important role for active prevention of DF/DHF in Vietnam under the impacts of climate change and weather variability.


Introduction
Dengue fever/dengue hemorrhagic fever (DF/DHF) is an infectious disease, transmitted by Aedes mosquito and has become a major public health concern in various countries worldwide.
Dengue epidemics were reported throughout the nineteenth and early twentieth centuries and the disease is found in both tropical and sub-tropical regions, predominantly in urban and semi-urban areas in Asian and Latin American countries [1]. Dengue viruses cause an estimated 50 million infections annually, among approximately 3.6 billion people at risk [2]. Dengue fever was first reported in Vietnam in Hanoi and Haiphong cities in 1959, and since then the disease has become endemic throughout the country. In the northern region, epidemics often occur between June to November, whereas in the southern region, cases have been documented year round [3]. During 1998, the largest DF/DHF epidemic in Vietnam since 1987 occurred, with official data recording 234,920 cases and 377 deaths in 56 out of 61 provinces throughout the country. Since then the disease has been an important public health challenge in the country [3].
Ben Tre province is in the Mekong River Delta, which is considered to be the most vulnerable region to climate change and sea level rise in Vietnam [4].  [5]. Previous studies have investigated the effect of climate change on DF/DHF transmission [6]. The risk of DF/DHF has been reported to be associated directly or indirectly with seasonal changes in climate [7,8]. Since water sources in Ba Tri District are quite salty, people usually use water containers to store rain water for drinking and other domestic purposes, which creates favorable condition for Aedes mosquito to breed.
Over the last few decades, there has been an increasing interest among scientists on the health impacts of climate change and weather variability. Numerous studies have been done and substantial results achieved mostly in developed countries, and not many quantitative evidences and assessment of the impacts at local levels were provided for the developing countries, including Vietnam. Some studies have shown that Aedes mosquitos are very sensitive to environmental conditions, especially temperature, rainfall, and humidity, while other studies did not find the correlation between rainfall and relative humidity with DF/DHF cases [6]. A study in Metro Manila, Philippines showed that DF/DHF was correlated with rainfall (p < 0.05) but not with temperature (p > 0.05) [9]. A multiple linear regression models were fitted to look for associations between changes in the incidence rate of DF/DHF and climate variability in the warm and humid region of Mexico and the results showed that the incidence rate or risk of infection was higher during El Niño events and in the warm and wet seasons [10].
In Vietnam, a study in the Central Highland showed that the risk of DF/DHF cases increased in rainy seasons when HI (house index), CI (container index) and BI (Breteau index) values of the vector density increased [11]. In 2011, Kien et al. published a book chapter, which showed that the change over time of incidences of malaria and the selected water-borne diseases in Vietnam were correlated with main climate variables (temperature and rainfall), and with inter-annual climate variability (the El Niño) [12]. Although being one of the most vulnerable countries to climate change, in Vietnam and in Ben Tre Province in particular, climate change and disaster issues have not been properly integrated into health policies, strategies and plans. This may partly be due to the weak linkages between policy development and scientific evidence and the lack of reliable quantitative and qualitative data. This study was conducted in 2015, which aimed to analyze retrospective secondary monthly data of DF/DHF cases and climate conditions from 2004-2014 in Ba Tri District, Ben Tre Province of Vietnam to explore the relationship between DF/DHF and climate variables.

The Study Area
Located in Indo-China peninsula of the Southeast Asia or the Greater Mekong Sub-region, Vietnam is a densely populated country and has been recognized as one of the most vulnerable countries in the world to the potential impacts of climate change and rising sea levels. Ba Tri District is located in Ben Tre Province, which is in the South of Vietnam having a tropical climate, with regular tropical typhoons, storms, and extensive flooding. The area is strongly influenced by Southwest monsoon, leading to two distinct seasons, being dry and rainy or wet seasons.
climate variables included series of monthly mean temperatures, monthly maximum and minimum mean temperatures ( o C), rainfall (millimeter), relative humidity (percentage) that was used as primary prioritized climate indicators and were collected at meteology station at Ba Tri District. The data were reported as average monthly data for each year during the studied period 2004-2014. The data set was provided by the Ben Tre Provincial Center for Hydro-Meteorological Forecasting.

Variables and Data Analysis
Variables analyzed in this article included: (1)  Square and Correlation (r value) and multivariate regression models to explore the correlation between DF/DHF cases, Aedes mosquito density, and climate variables.

Results
The results showed that the average annual temperature at Ba Tri District was 27.  Figure 1 shows that when the temperature was slightly reduced in the rainy months of June and July, the number of DF/DHF cases started to increase. Additionally, here was approximately a two month lag between the peak temperature reported in May and the peak of DF/DHF occurrences in July. Data during the period cases (r = 0.37). The average monthly rainfall recorded in Ba Tri District during the period 2004-2014 was 132 mm, with the highest monthly rainfall being 296 mm in July, while there was little or no rain in January and February. The number of DF/DHF cases started to increase from April when rainfall level increased, and continued to increase to achieve the pick level in July before it gradually declined in the following months. The data within the studied period showed a strong correlation between monthly DH/DHF cases and average monthly rainfall (r = 0.70, Figure 2).

Figure 2. The correlation between DF/DHF cases and average monthly rainfall (mm) at Ba Tri District, 2004-2014.
The average relative humidity in Ba Tri District during the period 2004-2014 was 83.1% (SD = 1.99), with the highest relative humidity level being 86% (in August and September), and the lowest being 80% (in January and February). Figure 3 shows that DF/DHF cases increased in correlation with the relative humidity from April to July, but then decreased in the following months while the relative humidity continued to increase until October. There was a strong positive correlation between the total monthly DF/DHF cases and relative average humidity (r = 0.59).

Discussion
The results of data analysis on vector density  and at provincial or national scales would provide more useful data for the analysis.

Conclusions and Recommendations
During the period from 2004 to 2014, there were strong positive correlations between the average montly DF/DHF cases with average monthly rainfall (r = 0.70), relative humidity (r = 0.59), Aedes aegypti density (r = 0.82), and BI (r = 0.81). There were moderate positive correlations between the average montly DF/DHF cases with average temperature (r = 0.37). The Ba Tri District Department of Preventive Medicine and commune health centers in Ba Tri District should monitor vector density, early diagnose DF/DHF cases at the beginning of rainy seasons starting from April, and apply active vector control measures to reduce the Aedes density and to prevent disease from developing and spreading in the following months. In addition, a larger study using data for all provinces and cities throughout Vietnam for a longer period (e.g. started from 1979 when data on DF/DHF and climate variables became available to 2015) should be undertaken to thoroughly describe the correlation between climate variability and DF/DHF cases as well as for modeling and to build projection model for the disease in the coming years based on climate data. This can play important role for active prevention of DF/DHF in Vietnam under the impacts of climate change and weather variability.