Abstract:
Malaria is still the leading cause of morbidity and mortality especially among children below five years of age in Kenya. Although current reports indicate declining prevalence of malaria in some parts of Africa, some reports paint a grim and opposite picture in some of the areas. In Kenya, some of the regions such as Bungoma County continue to experience persistently high prevalence of malaria all year round despite the scale up of control measures. This can be attributed to various plausible reasons among them local spatial variation in malaria risk (heterogeneity). Heterogeneity in malaria risk and transmission has been previously documented even on a very small scale. Spatial analysis of self reported fevers in the HDSS indicates clustering of fevers in some villages. The main objective for this study was therefore to determine whether there were actual differences in transmission of malaria in villages with higher than average fevers (fever hotspots) and villages with lower than average fevers as well as the risk factors by measuring malaria transmission indices for a period of one year. The information can be used for designing strategies for targeting malaria control measures to the local situation with the final aim of elimination of malaria in this area. Methodology: This was a prospective closed cohort study. The study was conducted in the Webuye Health and Demographic Surveillance Site (HDSS) located in Bungoma East Sub-County, a region which has had persistent and perennial malaria burden. Six villages (two in the fever coldspot and four in the fever hotspot) were selected for fixed entomological surveillance. One household was randomly selected in each of the villages to set up a window exit trap (WET) while two other households within the same village were selected for monthly mosquito monitoring using Pyrethrum Spray Catches (PSC). Parasitological surveillance was done for all household members in the same households where mosquito surveillance was set up as well as their immediate neighbours for a period of one year at quarterly intervals. A total of 400 participants in 72 households were followed up longitudinally and tested for malaria quarterly for the entire period. The person-month xi incidence rate of malaria was computed for one year. Risk factors for malaria in the fever hotspots and coldspots were computed using multi-level mixed effects modelling. A t-test was used to compare vector densities in the fever coldspots and fever hotspots as well as the incidence of malaria. ANOVA was used to test if there were significant differences in malaria incidence among the villages. Generalised estimating equation (GEE) was used to model factors associated with asymtomatic status. Linear regression was used to show the correlation between the vector densities and the incidence of malaria. Results: Although there was no statistically significant difference in the incidence of malaria infections between the fever hotspots and fever coldspots, those living in fever hotspots had almost one and half times increased risk of infection compared to those in the fever coldspots. There was marked and significant hetereogeneity in the incidence of malaria among the villages. Entomological risk factors such as increased larval sites and mosquito densities were mainly responsible for the observed differences in the incidence of malaria in both the fever hotspots and coldspots. Almost half (46.3%) of all the malaria infections were asymptomatic indicating a high prevelence of asymptomatic infections within the region. Malaria infections during the dry season (January) were less likely to be asymptomatic (A.O.R: 0.26, C.I: -2.289 - 0.400). Conclusions and Recommendations There is significant heterogeneity in the incidence of malaria among the villages correlating with entomological risk factors. There is need to target interventions based on the presenting local context in-order to increase their effectiveness. The high number of asymptomatic cases indicates the need to set up active malaria surveillance inorder to capture the asymptomatic individuals and treat them so as to reduce the parasite reservoir. Targeting the asymptomactic reservoir will reduce malaria infections further and therefore contribute towards the goal of elimination. Key words Malaria incidence, prevalence, hotspots, vector behavior, rural western Kenya.