Targeting remaining pockets of malaria transmission in Kenya to hasten progress towards national elimination goals: an assessment of prevalence and risk factors in children from the Lake endemic region

Background With an overall decline of malaria incidence, elimination of malaria is gradually becoming the next target for many of countries affected by the disease. In Kenya the national malaria control strategy is aiming to reach pre-elimination for most parts of the country. However, considerable heterogeneity in prevalence of the disease within the country and especially the remaining high prevalent region of the Lake endemic region is likely to slow progress towards this target. To achieve a sustained control and an eventual elimination, a clear understanding of drivers of ongoing malaria transmission in remaining hotspots is needed. Methods Data from the 2015 Malaria Indicator Survey (MIS) were analysed for prevalence of malaria parasitaemia in children (6 months to 14 years) of different countries within the highly endemic Lake region. Univariate and multivariate logistic regression analysis were preformed to explore associations between selected risk factors and being parasitaemic. A predictive model was built for the association between malaria and the risk factors with the aim of identifying heterogeneities of the disease at the lower administrative levels. Results Overall, 604/2253 (27%, 95% CI 21.8–32.2) children were parasitaemic. The highest prevalence was observed in Busia County (37%) and lowest in Bungoma County (18%). Multivariate logistic regression analysis showed that the 10–14 years age group (OR = 3.0, 95% CI 2.3–4.1), households in the poorest socio-economic class (OR = 2.1, 95% CI 1.3–3.3), farming (OR = 1.4, 95% CI 1.2–2.5) and residence in Busia (OR = 4.6, 95% CI 2.1–8.2), Kakamega (OR = 2.6, 95% CI 1.3–5.4), and Migori counties (OR = 4.6 95% CI 2.1–10.3) were associated with higher risk of parasitaemia. Having slept under a long-lasting insecticide-treated bed net (LLIN) was associated with a lower risk (OR = 0.7, 95% CI 0.6–0.9). No association were found between malaria infection and the gender of the child, the household head, and the education status of the household head. Discussion and conclusion Detailed analysis of malaria prevalence data in a hotspot area can identify new threats and avail opportunities for directing intervention. In the Lake endemic region of Kenya, interventions should be focused more on counties with the highest prevalence, and should target older children as well as children from the lower socio-economic strata. Precisely targeting interventions in remaining hotspots and high-risk populations will likely make impact and accelerate progress towards pre-elimination targets.