No description available.

Thesis (PhD)--Stellenbosch University, 2013. / ENGLISH ABSTRACT: Although malaria remains a major public health concern, especially in sub-Saharan Africa, little information exists on the physiological tolerances of malaria vectors. Here, I aimed to provide a comprehensive set of physiological tolerances for Anopheles arabiensis and An. funestus, by investigating thermal tolerance traits of adults, larvae and pupae; desiccation resistance of adults and development rate-temperature relationships for both vectors. Critical thermal limit (CT) and desiccation data showed significant effects of increasing adult age on reducing tolerance to temperature or dry conditions. Females of both species were more tolerant of high or low temperatures in CT experiments and were more desiccation tolerant than males in desiccation trials. Anopheles funestus was more desiccation tolerant than An. arabiensis, despite the common misconception that An. arabiensis is the more arid-adapted of the two species. Comparisons between thermal tolerance traits of adult laboratory and wild strain progeny of both species indicated a high degree of similarity between critical thermal limits in wild and laboratory strains, suggesting that the use of laboratory populations of both mosquito strains can provide an accurate estimate of wild population responses to thermal change. Lethal temperature estimates for both vectors indicated a higher tolerance to high temperature in An. arabiensis larvae and pupae when compared with An. funestus, and a greater tolerance of high or low temperatures in adult females when compared with adult males. Species differences between the vectors were further highlighted in development rate-temperature experiments. Under fluctuating and constant temperatures, An. arabiensis developed significantly faster than An. funestus and had higher survival to the adult stage. Under fluctuating temperatures, An. arabiensis developed faster or no different to constant temperatures, while survival under fluctuating temperatures was also comparable to constant temperature estimates. This faster development rate of this species is likely a consequence of the puddle-breeding nature of An. arabiensis and the need to develop to adulthood before evaporation of breeding sites. Anopheles funestus on the other hand, showed reduced survival and development under fluctuating temperatures when compared with constant temperatures, probably as a result of the more thermally stable breeding sites usually used by this species. Distribution data of these species, combined with developmental parameters in a process-based distribution model, suggests that both species will show range changes in response to climate change. Areas where these species were previously only present on a seasonal basis might become more suitable for vector population establishment and persistence, while areas on the northern margins of current distributions will become less favourable, leading to an overall southerly shift in habitat suitability for both species. Increases in temperature and changes in rainfall patterns as predicted to occur with climate change are likely to impact the distribution of both malaria vectors. Combining the physiological tolerance data collected in this thesis in a future, planned mechanistic distribution model, will provide an accurate indication of potential range shifts of these vectors and hence, provide an indication of areas that may be at increased risk of malaria. / AFRIKAANSE OPSOMMING: Alhoewel malaria „n groot publieke gesondheidskwelling bly, veral in sub-Sahara Afrika, bestaan min inligting rakende die fisiologiese toleransies van malaria vektore. Hier het ek gepoog om 'n omvattende reeks van fisiologiese toleransies te voorsien vir Anopheles arabiensis en An. funestus, deur termiese verdraagsaamheidseienskappe, uitdrogingsweerstand en ontwikkelingstempo-temperatuur verhoudings vir beide vektore te ondersoek. Kritiese termiese limiet (CT) en uitdroging data het beduidende uitwerkings getoon van toenemende ouderdom op die vermindering van verdraagsaamheid teenoor temperatuur of droë toestande. Wyfies van beide spesies was meer verdraagsaam vir hoë of lae temperature in CT eksperimente en was meer verdraagsaam teenoor uitdroging as mannetjies in die uitdrogingsproewe. Anopheles funestus was meer verdraagsaam teenoor uitdroging as An. arabiensis, ten spyte van die algemene wanopvatting dat An. arabiensis die meer ariede aangepaste van die twee spesies is. Vergelykings tussen termiese verdraagsaamheidseienskappe van laboratorium-en wilde stamlyn nageslagte van beide spesies het 'n hoë mate van ooreenkoms tussen kritieke termiese limiete in wilde en laboratorium stamlyne aangedui, wat voorstel dat die gebruik van laboratorium bevolkings van beide muskiet stamlyne 'n akkurate skatting kan gee van wilde bevolkingsreaksies tot termiese verandering. Fatale temperatuur beramings vir beide vektore het „n hoër toleransie getoon by hoë temperature in An. arabiensis larwes en papies wanneer dit vergelyk word met An. funestus, en 'n groter verdraagsaamheid van hoë of lae temperature in wyfies, wanneer vergelyk word met mannetjies. Spesies verskille tussen die vektore is verder uitgelig in die ontwikkelingstempo-temperatuur eksperimente. Onder wisselende en konstante temperature ontwikkel An. arabiensis aansienlik vinniger as An. funestus en het hoër oorlewing tot die volwasse stadium getoon. Onder wisselende temperature ontwikkel An. arabiensis vinniger of met geen verskil van konstante temperature nie, terwyl oorlewing onder wisselende temperature ook vergelykbaar was met konstante temperatuur beramings. Die vinniger tempo van hierdie spesie is waarskynlik 'n gevolg van die poel-broeiende aard van An. arabiensis en die behoefte om tot volwassenheid te ontwikkel voor die verdamping van broeiplekke. Anopheles funestus aan die ander kant, het verminderde oorlewing en ontwikkeling onder wisselende temperature gewys wanneer dit vergelyk word met konstante temperature, waarskynlik as gevolg van die meer termies stabiele broeiplekke wat gewoonlik gebruik word deur hierdie spesie. Verspreidingsdata van hierdie spesies, gekombineer met ontwikkelings-parameters in 'n proses-gebaseerde verspreidingsmodel, dui daarop dat beide spesies reeks veranderinge sal wys in reaksie tot klimaatsverandering. Gebiede waar hierdie spesies voorheen slegs teenwoordig was op 'n seisoenale basis, mag dalk meer geskik word vir vektor bevolkingsvestiging en volharding, terwyl areas op die noordelike grense van die huidige verspreidings minder gunstig sal word, wat sal lei tot algehele suidelike verskuiwing in die habitat geskiktheid vir beide spesies. Toenames in temperatuur en veranderinge in reënvalpatrone, soos voorspel word om voor te kom met verandering van die klimaat, sal waarskynlik die verspreiding van malaria vektore beïnvloed. Deur die fisiologiese toleransie data, versamel in hierdie tesis, te kombineer met 'n toekoms, beplande meganistiese verspreidingsmodel, sal dit 'n akkurate aanduiding gee van die potensiële verspreidingsverskuiwings van hierdie vektore en dus 'n aanduiding gee van gebiede wat onder verhoogde risiko van malaria sal wees.

No description available.

Ce travail est construit autour de deux problématiques de santé relatives aux deux plus grandes pandémies qui sévissent en Afrique sub-saharienne : i) l'hétérogénéité rencontrée dans la répartition spatiale et temporelle des vecteurs de paludisme ; ii) la variabilité dans l'observance au traitement antirétroviral par des personnes vivant avec le virus de l'immunodéficience humaine. Sur le plan méthodologique, ces deux problèmes se rapportent à la prise en compte de l'hétérogénéité dans la modélisation de données issues de mesures répétées ; ils nécessitent en outre le développement d'outils statistiques permettant de distinguer à partir des données, des sous-groupes (de localités, d'individus. . .) homogènes indispensables pour rendre plus efficientes les mesures de santé souvent déployer par les praticiens dans le cadre de la lutte contre le paludisme ou le VIH/SIDA. Les modèles de mélanges finis, grâce à leur flexibilité, sont des outils capables de fournir non seulement de bonnes estimations en présence d'une grande hétérogénéité dans les observations mais aussi une bonne partition des unités statistiques. Nous les distinguons, parmi d'autres méthodes, comme étant adaptés aux problématiques du présent travail. Deux applications de ces modèles aux données issues de capture de moustiques ont permis de modéliser la répartition spatiale et temporelle de vecteurs de paludisme et de dégager une méthode simple d'évaluation d'impact de mesures de lutte anti vectorielle. Nous introduisons la notion de _trajectoires de variances_ dans une troisième application portant sur des données d'observance aux traitements antirétroviraux par des personnes vivant avec le virus de l'immunodéficience humaine / This work focuses on two health issues relating to two major pandemics in sub- Saharan Africa : i) the heterogeneity encountered in the spatial and temporal distribution of malaria vectors ; ii) the variability in adherence to antiretroviral treatment by people living with the human immunodeficiency virus. Methodologically, these two problems are related to the consideration of the heterogeneity in the modeling of data from repeated measurements. They also require the development of statistical tools to distinguish from the data, homogeneous clusters of localities, individuals. . . that are needed to make more efficient health measures often deployed by practitioners in the fight against malaria and HIV/AIDS. The finite mixture models, due to their flexibility, are statistical tools that not only provide good estimates in the presence of heterogeneity in the observations but also a good classification of statistical units. We show that they are able to deal with the problematics of our study. The spatial and temporal distributions of malaria vectors are modeled through two different applications of finite mixture models and a simple tool to evaluate the impact of vector control methods is generated. We introduce a ”variance trajectories” method in a third application of finite mixture models to data on adherence to antiretroviral therapy by people living with human immunodeficiency virus