Hydro-climatic impact on cholera incidence in Dhaka under global warming

Abstract

ACKNOWLEDGEMENTS

About 5-year of my Ph.D. journey includes completing coursework, research design, writing journal article, receiving reviews or rejections, and finally with writing this dissertation. Within this period, a number of people supported, guided and encouraged me to complete the research as well as this dissertation. Therefore, a number of people should get deep gratitude from my heart. First, I like to express my gratitude, appreciation and a message of thanks to the Dissertation Supervisor Prof. Dr. A.K.M. Saiful Islamfor providing valuable guidance, supervision, mentoring as well as encouragement for carrying out this research. He always kept me focused on my research objectives.Next, I would like to express my gratitude to my co-supervisor, Dr. Peter Kjær Mackie Jensen, University of Copenhagen, Denmark, for his continuous assistance towards knowledge about epidemiology needed for the success of this work. I wish to thank all members of the Doctoral Committee and Board of examiners - Prof. Dr. Sujit Kumar Bala, IWFM, BUET; Prof. Dr. G.M. Tarekul Islam, IWFM, BUET; former Prof. Dr. M. MozzammelHoque, IWFM, BUET; Prof. Dr. Anowara Begum, Dept. of Microbiology, University of Dhaka; and Prof. Dr. Syed Hafizur Rahman, Dept. of Environmental Sciences, Jahangirnagar University for providing valuable comments, support and guidance throughout the course and research. I would like to thank my colleagues of Climate Modeling and Simulation lab, IWFM, BUET for the encouragement and for all the memories we have had in the last five years. This dissertation would not have been possible without the intellectual contribution of Dr. Mohan Kumar Das, Khaled Mohammed and Jamal Uddin Khan, a research associate in the lab. I am also very grateful for the C5 (Combating Cholera Caused by Climate Change in Bangladesh) project funded by DANIDA, Council of Development Research, Ministry of Foreign Affairs of Denmark for proving me all kinds of financial support for conducting this research.

Special thanks to the staff and management of IWFM, BUET for providing an opportunity to enhance and establishing this research. I am indebted to many individuals without whose assistance, this research could not have been completed in time. I want to thank everyone who has helped me along the way. Last but not the least, a very special thanks goes to my parents, my husband S.M. Tanvir Hassan, our son TazwarTanvir, and daughter Shabnam Sabah Mukti, without endless love, supports and encouragement it was hardly possible to complete this dissertation work at the end.

ABSTRACT Cholera, an acute diarrheal disease spread by lack of hygiene and contaminated water, is a major public health risk in many countries. Bangladesh experiences endemic cholera for more than 2,000 years, where cholera is triggered by environmental conditions influenced by climatic variables. This study employed to establish a correlation between cholera incidence and climatic variables, which would provide an opportunity to develop a cholera forecasting model in densely populated Dhaka megacity. The aim of this study was to predict the potential impact of future climate change on cholera using the consequences of high-end concentration scenario (RCP8.5), and further an adaptation guideline has been described through a systematic review on preparedness practices.

In this study, a time series analysis namely a seasonal-auto-regressive-integrated-moving-average (SARIMA) model as short-term forecasting was used considering the auto-regressive nature and the seasonal behavioral patterns of cholera. As both rainfall (r=0.43) and maximum temperature (r=0.56) have the strongest influence on the occurrence of cholera incidence, single-variable (SVMs) and multi-variable SARIMA models (MVMs) have been developed and tested for evaluating their relationship with cholera incidence for the period of 2000 to 2013. Low relationship was found with relative humidity (r=0.28). Using SVM for 1°C increase in maximum temperature at one-month lead time showed 7% increase of cholera incidence (p<0.001). However, MVM (AIC=15, BIC=36) showed better performance than SVM (AIC=21, BIC=39). An MVM using rainfall and monthly mean daily maximum temperature with 1-month lead time showed a better fit (RMSE=14.7, MAE=11) than the MVM with no lead time (RMSE=16.2, MAE=13.2) in forecasting.

In this study,another time series analysis namely Artificial Neural Network (ANN) model was used to compare the performance with SARIMA model and further, to assess the potential impact of climate change on cholera incidence as long-term prediction. Rainfall, maximum temperature and Brahmaputra River discharge (SWAT output) for 11 different climate projections were used as input data to calibrate (1986-2005) and validate (2006-2013) the cholera prediction model. Then, the calibrated model was used to simulate future impact of cholera for the perspective climate projections, and analyzed the simulated cholera to estimate the future change in different time slices. During the peak of the pre-monsoon season, cholera cases could potentially increase because of climate change by 18% - 80% for 2020-2039, by 24% - 119% for 2040-2059, by 11% - 208% for 2060-2079, and by 36% - 308% for 2080-2099.

Finally, a systematic review has been done on preparedness practices and response plan strategies of combating cholera for different countries of the world to recommend adaptation guideline for Bangladesh. The government of Bangladesh is eager to improve preparedness against waterborne diseases by ensuring safe drinking water all around the year including pre- and post-monsoon seasonal scarcity, which will be benefited by the findings of this study. Still, Bangladesh needs a ‘preparedness and response plan’ for its endemic cholera outbreaks every year during pre- and post-monsoon. This study proposed a preparedness and response plan as an adaptation guideline to combat cholera in Bangladesh.