Modeling partial cross-protection for managing cassava mosaic disease: a vector–host framework and sensitivity analysis

Introduction Cross-protection—where prior infection by a mild strain reduces susceptibility to a severe strain—offers a promising but underexplored option for managing persistent vector-borne plant diseases such as cassava mosaic disease (CMD). Methods We developed a deterministic host–vector transmission model for CMD that incorporates partial cross-protection alongside roguing, replanting, harvesting, and vector control. We analyzed positivity and equilibria, derived threshold conditions for elimination, and assessed parameter influence using global sensitivity and uncertainty analyses. Results The analysis establishes positivity and global stability of the disease-free equilibrium and identifies conditions under which backward bifurcation occurs, implying that reducing the basic reproduction number below one may be insufficient for elimination. Sensitivity and uncertainty analyses indicate that mild-strain transmission parameters ( β 2 , β 4 ) and the roguing rate σ are the most influential drivers of CMD prevalence. Discussion Model outcomes suggest that improving protection effectiveness can shift cross-protection from a supplementary measure to a primary management strategy. Although motivated by cassava, the framework is adaptable to other vector-borne crop diseases and provides quantitative guidance for designing robust, integrated disease-management programs.