The spatial epidemiology of leprosy in Kenya

Abstract

Leprosy elimination is defined as a prevalence rate below 1 case per 10,000 population. This rate was attained in Kenya at the national level in 1989. However, there remain leprosy endemic counties where late diagnosis and physical disability persist. There is therefore need to define the populations at risk to enable the national program to plan leprosy control activities considering the County specific needs, as well as evidence based prioritization of resources within the respective counties. This is a retrospective ecological correlational study that aims to describe the spatial epidemiology of leprosy in Kenya for the period 2012 through to 2015. The study utilized secondary data sources; leprosy case based data was extracted from the National Leprosy Control Program database, whereas GIS and demographic data were obtained from KNBS. Exploratory analysis showed that there is active leprosy transmission in Kenya with children less than 15 years accounting for 7.5% of the total cases. Most notified cases were males (62%) whereas females accounted for the remaining 38%. 88% of all notified cases had MB leprosy and 52.9% had grade 1 or grade 2 disability. Two separate spatial Poisson Conditional Autoregressive (CAR) models were fitted; for all leprosy cases and for only new cases. The risk of leprosy incidence for all cases increased by about 5% for every 1 year increase in age, whereas a one percent increase in the proportion of MB cases increased the leprosy risk among new cases by approximately 4%. When all cases were considered, counties with the highest risks of leprosy include Kwale (RR of 15), Kilifi (8.9) and Homabay (4.1), whereas Turkana had the lowest relative risk of 0.005. A similar trend was observed with the new cases, with Kwale leading at a relative risk of 16, kilifi (8.6) and Homabay (3.7). Turkana remained the county with the lowest risk for new case incidence at a RR value of 0.003. The study concluded that leprosy incidence exhibits geographical variation and there is therefore need to institute local control measures.