Non-random distribution of Plasmodium Species infections and associated clinical features in children in the lake Victoria region, Kenya, 2012–2018

Abstract Background While Plasmodium falciparum (Pf) stands out as the most lethal malaria parasite species in humans, the impact of other species should not be dismissed. Moreover, there is a notable lack of understanding of mixed-species infections and their clinical implications. Methods We conducted eight school-based cross-sectional malariometric surveys in the Lake Victoria region of western Kenya between January–February 2012 and September–October 2018. In each survey, a minimum of 100 children aged 3 to 15 years were randomly chosen from a school in Ungoye village on the mainland and as well as from each school selected in every catchment area on Mfangano island. Plasmodium infection was determined by microscopy and nested polymerase chain reaction (PCR). The multiple-kind lottery (MKL) model calculated the expected distribution of Plasmodium infections in the population and compared it to observed values using a chi-squared test (χ2). Results The Plasmodium prevalence was 25.9% (2521/9724) by microscopy and 51.1% (4969/9724) by PCR. Among all infections detected by PCR, Pf, P. malariae (Pm), and P. ovale (Po) mono-infections were 58.6%, 3.1%, and 1.8%, respectively. Pf/Pm, Pf/Po, Pm/Po, and Pf/Pm/Po co-infections were 23.5%, 4.3%, 0.1%, and 8.6%, respectively. MKL modelling revealed non-random distributions, with frequencies of Pf/Pm and Pf/Pm/Po co-infections being significantly higher than expected (χ2 = 3385.60, p