The global prevalence of diabetes is increasing, partly due to lifestyle factors such as high sugar consumption. While mathematical models have often been used to study how diabetes develops, the direct effect of eating sugary foods on progression between disease stages has not been well studied. This study presents a compartmental ordinary differential equation model that characterizes transitions among susceptible, prediabetic, and diabetic populations, with sugar consumption intensity incorporated as a key driver of disease progression. The model is studied using equilibrium analysis, nullcline structure, and phase-plane dynamics within the prediabetes to diabetes state space. Analytical results identify threshold conditions that determine whether the system exhibits prediabetes-dominant or diabetes-dominant equilibria. Increased sugar intake shifts the system toward persistent diabetes. Phase-plane simulations indicate that increased sugar consumption qualitatively modifies long-term system dynamics, leading to higher diabetes prevalence. The results underscore the significant influence of dietary behaviour on diabetes progression and establish a mathematical framework for evaluating the long-term effects of sugar-reduction interventions.