| dc.description.abstract | Despite current gains in malaria reduction, the disease continues to have devastating health and livelihood impacts especially in sub-Saharan Africa where over 80% of global malaria cases exist. Malaria is an environmentally sensitive disease linked to climatic and environmental factors. Changes in the aforementioned factors alter the distribution and abundance of malaria vectors and consequently malaria transmission risks. The aim of this study was to evaluate the influence of climatic and ecohydrological factors on malaria vectors and develop a framework for malaria risk reduction in Baringo County, Kenya. Remotely sensed climate and normalized difference vegetation index (NDVI) data were used in Mann-Kendall trend analysis over the period 2003-2016. Sampling of mosquito larvae was conducted monthly from December 2015 to December 2016. Using a generalized linear mixed model (GLMM), climate and ecohydrological variables (land cover, vegetation health, wetness index, slope and discharge) were used to model Anopheles gambiae s.l. distribution. Clinical malaria cases were used in malaria trend analysis over the 2005-2014 period, this being the duration over which hospital records overlapped with climate data. A triangulation of focus group discussion, key informant interviews and household survey was used to assess lay knowledge on climate and malaria risks and in developing a malaria reduction framework for communities in Baringo County. During the 2003-2016 period, a significant increase in mean annual minimum and maximum temperatures was observed in the lowland zone (p < 0.05). There was no change in total annual rainfall contrary to the local perception that rainfall had decreased. The March April May (MAM) rains season was reported as being unreliable and depressed. During the same study period, an increase in annual vegetation health was observed in the riverine (tau = 0.103; p = 0.047) and lowland zones (tau = 0.051; p > 0.05). A significant increase in Lake Baringo water levels occurred during the 2011-2014 period (tau = 0.569; p < 0.001). Discharge in River Arror decreased significantly (tau = -0.412; p < 0.001) amidst non-significant changes in rivers Perkerra and Kessup. NDVI had a one-month lagged response to rainfall and a zero-lag response to Tmin and Tmax. Higher Anopheles gambiae s.l. abundance occurred in the riverine zone compared to the lowland zone (W = 13; Z = -2.04; p < 0.05; r = 0.59). Rainfall, slope and vegetation health significantly influenced the distribution of Anopheles gambiae s.l. mosquito larvae at p = 0.00041, p = 0.012, and p = 0.038 respectively. Malaria trends increased significantly in the riverine zone during the 2005-2015 period (tau = 0.352; p < 0.001. There was a coexistence of misconceptions and correct medical knowledge on the causes and modes of malaria transmission. Night and early morning outdoor occupation increased exposure of residents to mosquito bite and malaria transmission. There was over reliance on mosquito nets for malaria control while supplementary strategies such as environmental management and use of chemical sprays were largely underutilized. In addition to mosquito nets currently in use, measures including unclogging irrigation canals, draining pools near homes, managing Prosopis bushes, fogging, use of screens, clean-up campaigns and enhanced malaria surveillance were proposed in the malaria risk reduction framework. Awareness and education should be conducted to demystify misconceptions and promote behaviour change. The time lags between peak malaria cases and the assessed factors can inform timely distribution of resources to deter malaria transmission. The county government should give prioritized consideration to the riverine zone with regard to malaria surveillance and resource distribution. Implementation of community-driven integrated approaches will result in universal utilization of malaria control measures and a substantial reduction in malaria prevalence in Baringo County. | en_US |
