This study develops and evaluates a layered measurement model that extends conventional structural vulnerability through two explicitly represented layers: system buffering capacity (sCVI) and epidemiological modulation (RR). The resulting formulation preserves analytical separation between structural conditions, system mediation and realised outcomes rather than combining these components within a single descriptive index. The model was parameterised using longitudinal data from Kenya covering 1987–2023 and evaluated through empirical alignment, elasticity, dominance analysis, residual diagnostics, Monte Carlo uncertainty propagation and global sensitivity analysis.