Burundi is a small country located in East-central Africa, between 2°20 and 4°27 of latitude South and between 28°50 and 30°55 of longitude East; the altitude varies between 775 metres and 2670 metres. Bounded on the north by Rwanda, in south-east by Tanzania and in west by the Democratic Republic of Congo, Burundi covers an area of 27,834 km2, of which 2,634 km2 are occupied by Tanganyika lake. With a population estimated at about 8.4 million and a life expectancy of 48 years, Burundi remains one of the ten poorest countries in the world[1] . Approximately 80% of Burundi's population live in poverty [2]. Fifteen years of civil war since 1993 has worsened the situation. In terms of habitat, it remains essentially rural, with 91.6% of the population living in rural area. The urban population is estimated at 8.4% with an annual growth rate of 5.7%. The Burundi population is young: 46.1% are under 15 years of age, while people aged 60 and above represent only 5.4% [3]. With an average density of 266 inhabitants per km2, a population growth rate of 3.44% and a total fertility rate of 6 children per woman, Burundi is one of Africa's most densely populated countries [3]. The country is structured in 17 provinces. Burundi has generally a tropical highland climate, with a significant daily temperature variation in many areas [4]. Temperature also varies considerably from one region to another, basically as a result of differences in altitude. The central plateau is cool, with temperature averaging 20°C. The area near Lake Tanganyika is warmer, with temperature averaging 23°C; the highest mountain areas are cooler, with temperature averaging 16°C. Rain is irregular in Burundi, falling most heavily in the Northwest [4]. Dry season varies in length, and there are sometimes longer periods of drought, specially in the North. Most parts of Burundi receive between 130 and 160 cm of rainfall per year [4]. The epidemiological profile can be summarized as follows. The health system suffers from a shortage of qualified personnel with 1 doctor per 34,750 inhabitants and 1 nurse for 3,500 inhabitants. 17.4% of patients do not have access to health care, while 81.5% of patients are forced to go into debt or sell property to pay the health costs [3]. There is a big disparity between the capital Bujumbura and the remainder of the country as 80% of doctors and more than 50% of nurses are engaged in Bujumbura where only 1/20 of the population is living[5]. Malaria is the major public health problem in Burundi and therefore places a heavy burden on the health system. According to the ministry of public health statistics, each year between 2 and 2,5 million people are affected by malaria[6], it is responsible for up to 60% of all outpatient visits and up to 50% of deaths in health facilities among children under five years of age. Almost the entire population of Burundi lives in areas at risk of malaria [1]. Respiratory infections and diarrheal diseases are also among the most frequent causes of mortality and are mainly due to poor environmental conditions, such as precarious housing, overcrowding, poor sanitation,
limited access to drinking water, etc. Preventing and fighting against epidemic-prone diseases remain a major priority of Public Health in Burundi [3]. More details on the study area, mainly the geography, climate, population, history, health situation, malaria epidemiology namely its history and evolution from 1940, are well described in the chapter 2 of the thesis by Protopopoff [7]. The national malaria control strategy (2008-2012) has the ambitious goal of early detection and containment of 100% of malaria epidemics in the country by 2012 [1]. The approach that was proposed to be used was to carry out meteorologic monitoring to predict conditions that could initiate a sudden increase in malaria transmission [1]. However, few studies have been undertaken to understand the association between malaria and climatic variables in Burundi. Our thesis attempts precisely to fill that gap. The following questions are addressed in this thesis.
The thesis is organized as follows.
In chapter 1
we present and discuss the epidemiology of malaria. We describe transmission cycle and clinical manifestation of the disease, discuss the effect of climatic factors on malaria and present the burden of malaria with more emphasis on its burden on Africa. We also provide some fuel of the worse situation of malaria in Burundi.
The chapter 2
provides an extensive exploratory analysis of the data used in our thesis. We perform both non-spatial and spatial exploratory analysis.
The chapter 3 investigates the effects of climate on malaria in the area of Burundi. We propose a mathematical regression model to assess which climatic variables significantly influence malaria incidence in Burundi.
In chapter 4,
we analyze the effect of increased temperature on malaria in Burundi. We propose a Bayesian generalised additive model to assess whether the (predicted/ forecast ) increase in temperature will result in increasing malaria transmission in Burundi.
In chapter 5
we propose a Bayesian Geoadditive modeling of malaria in Burundi to assess the existence of spatial patterns of malaria which are explained by factors other than climate in Burundi.
In chapter 6
we assess which forecasting method allows for a better forecasting of malaria cases in Bujumbura when taking into account association between climatic factors and the disease.
In chapter 7
we summarize our findings by presenting some answers to the questions addressed in the introduction. We describe what is learned and how it is useful to malaria control in Burundi.
References
Chapter 8.
General Conclusions
In this thesis statistical analysis have been conducted to assess
Monthly data on malaria cases and climatic variables namely rainfall, temperature and humidity were collected in EPISTAT and IGEBU. An extensive exploratory analysis of these data is conducted in chapter 2. Detailed discussion on the findings has been given previously in chapter 4-7. Here we only present a summary of the main contributions and recommendations for future research.
The main findings of the studies carried out in this thesis can be summarized as follows.
Potential contributions of these findings are
Future researches in the area are worth to be mentioned here. As already mentioned above, there is a need to incorporate factors other than climate in the modeling of malaria. Factors such as malnutrition, housing conditions, level of education should be taken into account to better understand the distribution of malaria in Burundi. Moreover,
in our study the modeling was mainly based the generalized linear or general additive regression models. Some other modeling methods such as the proper generalized Markov random field (PGMRF) will be tried out.
Our study was subject to a number of limitations that are worth to be acknowledged. The first limitation is the way malaria cases were reported. This may have changed in the 12-years period. Our assumption of uniform data collection may not hold. The second limitation is the absence of microbiological confirmation of malaria cases in most of the health centers. This may have overestimated the number of malaria cases as in many health centers fever was the only criterion. A third limitation is that only one meteorological station was available per province. More monitoring stations are needed to account for the variability of the climatic variables on small distances. A fourth limitation is the assumption that the record of the climatic variables for the study period (12 years) has remained uniform with the same precision. This assumption may not hold since in
12- years time the measurements instrument may have had some malfunction mainly due to the lack of maintenance. The fifth limitation arises from the used data themselves. Our study used monthly data, but in epidemiology studies weekly data would be of great advantage, especially in terms of early warning. Unfortunately in countries like Burundi it's very difficult, even impossible to obtain weekly report of malaria cases.
