Investigation of systems ecologic models of human African Trypanosomiasis epidemic: A case of Kenya

Abstract

An investigation of the use of systems ecologic models in developing a general theory of the human African trypanosomiasis (HAT) or human sleeping sickness epidemics has been carried out. The study problem has successfully been decomposed into three levels of analysis, namely: regional (mega-scale), ecosystem (macro-scale) and biotopes (micro-scale). Scale integration issues are addressed and linked to specific features of the Lambwe Valley ecosystem, the core area of this study, situated in the Lake Victoria Basin, Kenya. Theory expounded in this study has been devoted to a synthesis of diffusion models of the HAT infection vector i.e. tsetse-fly (particularly G. pallidipes) population dispersal; the geometry that describes mega-, macro, and micro-scale dynamics influencing the epidemiology of human sleeping sickness; and the systems analysis modelling of the aforementioned aspects. In essence,the principle objective of this study was to develop a systems ecologic theory on the geographic spread of the HAT infection in space and time by mathematical modelling. Major theoretical and applied contributions are made. First, is the articulation of the methodology of systems ecology to the study problem by making use of a myriad of ideas and techniques. Second, is the development of a new concept, termed 'parasitemia force' at a mega-scale (regional dynamics), which is then linked with environmental attributes delineating the study site. Third, using the Lotka-Volterra model, the Ruma National Park located in the study site is characterization as a 'predator' or 'infective cell' and the surrounding human settlements as a 'prey' or 'impact zone' , at the macro-scale analysis ( ecosystem dynamics). The theory is then extended with a component of diffusion theory thereby explaining the significance of the concept of 'parasitaemia force' in elucidating the wave-like spread of the HAT epidemics. Next, the biology and ecology of the salient biotopes, namely the trypanosome parasite (T. b. rhodesiencei, the reservoir agents (both domestic and wildlife animals), the vector (G. pallidipes) and the definitive human hosts have been appreciated. Consequently, the theoretical ecology of the intra- and inter-biotopes interactions has been explained. In retrospect, parasite susceptibility in the reservoir agents, the vector and human hosts is examined by use of differential equations and probability distribution functions (pdf). Finally, an exposition of the geometry of the geographic spread of the HAT infection built on susceptible (X), infected (Y) and immune or recovered (Z) individuals, has been accomplished and the steady state equilibrium determined. The ensuing difficulties in solution determination authenticates the underlying complexity of the disease. A flow model on the vector (G. pallidipesi life-cycle has been formulated. Also, the salient variables, parameters, and functional relationships governing vector dynamics on a temporaland spatial dimensions in the study area have been determined, explained and tabulated. Next, a part of the theoretical study is devoted to derivation of a mathematical model depicting the complex transmission system of the HAT infection. Also, systems ecologic models that muster intra- and inter-biotopes interactions which influence and determine the threshold levels of the HAT epidemics have been developed. A summary of methods of estimation of the apt parameters is included. An implementation agenda for computer simulation of these models (whichis beyond the scope of the present study) is proposed. The applied components of the study include a demonstration on the use of satellite data processing techniques to identify tsetse fly habitats. Such a geomatics methodology utilizes the advantage of multistage sampling in generating information on land-use patterns, vegetation status, e.t.c., as they influence habitat preference characteristics that mitigating tsetse fly infestation. The approach being cost-effective, permitts the use of multi-level data on epidemiologic factors of the HAT epidemics. Thus the contribution made should prove resourceful in the formulation of effective vector population control programmes. The epidemiology of the HAT infection has been synthesized further. The analysis of case records for the period 1959-1990 reveals that the most vulnerable age category is 10-34 years while the least is below 10 and above 60 years. In general more males contract the infection compared to females. The influencing factors are discussed. Subsequently, a map depicting the HAT epidemic surface has been developed and presented. Hence, the spatial extent of the problem in the study site has been established. Further, parasitological surveys were carried out to demonstrate their significance to the understanding of the disease prevalence. Following evidence covered in these empirical analyses, the probability distribution of the HAT infection risk capacity is derived and articulated to the elucidation of those factors influencing human settlement in an epidemic environment. From a practical point of view, the development of systems ecologic and epidemiologic theory on the HAT epidemics has been found to be beneficial, particularly its potental in the design of effective control of the disease vector population. Several recommendation are made for policy makers as well as for researchers. For instance, in the latter case, it is suggested in one of the recommendations that, it would be useful to further extend the theory developed in this study to use variable probabilities (i.e. stochasticity) of biotopes' interactions via the climatic variables and link the result to the design of a Trypanosomiasis Information System (TIS) that integrates information at all scales of problem analysis. Thus, it is proposed that a real time expert system (ES) based on knowledge engineering (i.e. artificial intelligence tools) on the geographic spread of the human African trypanosomiasis (HAT) infection in affected tropical ecosystems can be developed. Certainly, this approach should help toward the realization of effective and efficient epidemic control or eventual eradication of the disease, especially so, within the study site. Implications of natural resources and public health policies are also discussed.