Yamba, Edmund Ilimoan (2016). Improvement and validation of dynamical malaria models in Africa. PhD thesis, Universität zu Köln.

Edmund_Yamba_thesis.pdf - Accepted Version

Download (1MB)


Overcoming the serious public health burden of malaria in Africa especially the sub-Saharan Africa requires a detailed understanding of malaria epidemiology in the region. To contribute to this effect, this work embarked on several research steps. The first part of this study investigated the impact of climatic and environmental factors on seasonal malaria transmission in Africa. Monthly Entomological Inoculation Rate (EIRm) data initially gathered from different malaria locations across the region via a literature review was utilized for this purpose. The results revealed that rainfall was the primary climatic determinant of malaria seasonality at markedly seasonal rainfall areas such as Sahel and Eritrea. But its impact at bimodal rainfall distributed and more humid zones was more complex. Temperature was not a limiting factor of malaria seasonality in Africa except for East Africa where it can delay the impact of rainfall. The seasonal peaking characteristics of malaria were mostly unimodal. At zones characterised by bimodal rainfall distributions, the peaks were frequently associated with the first rainfall maximum of the year. Seasonal malaria intensity anti-correlated with elevation and population density. Though seasonal malaria transmission is driven mainly by An. gambiae, An funestus and An. arabiensis, the vectors had competing and complex individual impact on seasonality. The findings of this section have important implications for the disease control especially the spatial and temporal target of malaria interventions and resource allocation. Besides, it provides information regarding future malaria modelling efforts and the validation and evaluation of existing weather-driven malaria models. The second part of the study validated seasonal malaria transmission in Africa simulated by two malaria models using the observed EIRm data. The models include the 2010 version of the Liverpool Malaria Model (LMM2010) and the VECtor-borne disease community model of the international centre for theoretical physics,TRIeste (VECTRI). The goal was to determine the accuracy of both models in simulating seasonal malaria transmission in Africa. The validation revealed that LMM2010 and VECTRI error ranges were generally within or about the same as the standard deviation of the observed EIRm data though larger errors were detected for Guinea and some individual monthly minor differences. Both models also agreed with observations that the seasonal peaking behaviour of malaria was predominantly unimodal. However, transmission peaks in the models tend to be delayed by one month in the Sahel and Eritrea area. Both models further agreed with observed values of a seasonality index that the seasonal malaria transmission contrast is closely linked with the latitudinal variation of climatic covariates such as rainfall in Africa. VECTRI revealed a stronger ability in capturing the levels of malaria endemicity in East Africa than LMM2010. The hydrology model in VECTRI poorly captures seasonal malaria transmission at permanent water body locations. Though both models had loopholes, inferences from the validation conclude that they could realistically reproduce the seasonal evolution of the disease in Africa as a function of climate and environment. The findings, therefore, provides the basis for further review and refinement of the models by their developers to stage them as best fundamental tools for seasonal malaria prediction. In the third part of this work, a formulated simple model of immunity to malaria and incorporated into VECTRI was evaluated. Also, the section performed a one-at-atime sensitivity study of VECTRI parameter settings to its output variability. The results revealed that the immunity model enabled VECTRI to simulate different levels of malaria for Africa by reducing transmission rates at increased exposure of humans to malaria. The simple immunity model also substantially improved the seasonal malaria simulations of VECTRI by reducing its output error. The one-at-a-time sensitivity analysis performed on VECTRI parameter settings revealed parameters showing the strongest variation of the model output. The most sensitive parameter settings consisted of survival probabilities (i.e. adult vector and larval survival), threshold temperatures (i.e. minimum temperature for larval survival and that for the sporogonic cycle), degree days (i.e. larvae growth), and hydrological components (i.e. total evaporation and infiltration losses). The new immunity model represents a helpful tool for future malaria modelling effort, and its refinement for consideration in VECTRI is necessary. Parameters contributing most to VECTRI output variability require additional research to strengthen knowledge base to reduce VECTRI output uncertainty.

Item Type: Thesis (PhD thesis)
CreatorsEmailORCIDORCID Put Code
Yamba, Edmund Ilimoaneyilimoan48@gmail.comUNSPECIFIEDUNSPECIFIED
URN: urn:nbn:de:hbz:38-74323
Date: 25 October 2016
Language: English
Faculty: Faculty of Mathematics and Natural Sciences
Divisions: Faculty of Mathematics and Natural Sciences > Department of Geosciences > Institute for Geophysics and Meteorology
Subjects: Natural sciences and mathematics
Life sciences
Uncontrolled Keywords:
Entomological Inoculation Rate (EIR)English
Human Biting Rate (HBR)English
CircumSporozoite Protein Rate (CSPR)English
Date of oral exam: 25 October 2016
NameAcademic Title
Crewell, SusanneProf. Dr.
Fink, Andreas H.Prof. Dr.
Tompkins, Adrian P.Prof. Dr.
Tezkan, BülentProf. Dr.
Ermert, VolkerDr.
Funders: Katholischer Akademischer Ausländer-Dienst (KAAD), Universität zu Köln (extra funding from my professor)
Refereed: Yes
URI: http://kups.ub.uni-koeln.de/id/eprint/7432


Downloads per month over past year


Actions (login required)

View Item View Item