Space-time sensitive modelling of subclinical malaria prevalence at the village level in low burden areas of Myanmar using random forest models

Authors: Poe Poe Aung, Zaw Win Thein, Milena Pavlova, Wim Groot, Regien Biesma, Thura Htay, Zin Min Tun, Aye Kyawt Paing, Kay Thwe Han, Nay Yi Yi Lin, Zaw Lin, Varada Shevade, Tatiana Loboda

In low-burden areas, village-level socio-demographic, climatic and environmental factors are not sufficient to reliably predict subclinical malaria rates. Incorporating temporal predictors in areas with higher malaria endemicity will likely improve model performance.

Malaria elimination in Myanmar has slowed in recent years. Very low parasitaemia is acting as a silent reservoir for the continued transmission of asymptomatic infections, complicating efforts to accurately estimate malaria risk and the true burden of infection. Understanding malaria epidemiology is important for developing targeted strategies in resource-limited settings. However, there is limited research about the impact of urbanisation, climate change, environmental factors and human characteristics in low transmission areas. This study examined the temporal variation in subclinical malaria at the village level in low burden areas and quantified the impact of socio-demographic, climatic and environmental factors.

This poster was presented at the 73rd Annual Meeting of the American Society of Tropical Medicine & Hygiene (ASTMH).