Predictive analysis across spatial scales links zoonotic malaria to deforestation

Area of Science:

• Environmental Epidemiology
• Disease Ecology
• Spatial Analysis

Background:

• Vector-borne and zoonotic diseases present complex control challenges, particularly with landscape changes.
• Human incidence of zoonotic malaria, Plasmodium knowlesi, is associated with deforestation, but the underlying mechanisms remain unclear.

Purpose of the Study:

• To identify key spatial scales linking zoonotic malaria (Plasmodium knowlesi) occurrence with environmental change.
• To refine predictions of disease occurrence and inform spatially-targeted interventions.

Main Methods:

• A novel method combining machine learning and statistics was applied to predict disease occurrence.
• Predictive models of Plasmodium knowlesi occurrence at the household level were fitted using satellite imagery, case-control, and cross-sectional survey data.
• Sixteen environmental variables were analyzed across 11 spatial scales simultaneously.

Main Results:

• A strong peak in predictive influence was observed for the proportion of cleared land within 1 km of households.
• Environmental factors like aspect, slope, and canopy regrowth were significant at smaller spatial scales (0.5–2 km).
• Fragmentation of deforested areas showed the strongest influence on Plasmodium knowlesi occurrence probability at larger scales (4–5 km).

Conclusions:

• The study identified specific spatial scales critical for understanding the relationship between land use change and Plasmodium knowlesi.
• These findings narrow down plausible mechanisms connecting environmental changes to zoonotic malaria.
• Results enable improved disease occurrence predictions and the design of effective, spatially-targeted interventions.