Background <p>The positive relationship between higher delay discounting, an indicator of increased impulsivity, and reduced engagement in diabetes care has been investigated. However, the association between delay discounting and diabetes onset, likely linked through unhealthy behaviors, has not been well investigated. Additionally, although depression has been linked to an increased risk of diabetes and greater delay discounting, studies examining associations among all three factors are scarce. The present study aimed to determine the association between depressive symptoms and the onset of diabetes, with delay discounting as a mediator of this relationship.</p> Methods <p>Using data from a three-phase online prospective survey of a community sample, cross-sectional and longitudinal mediation analyses were conducted to examine diabetes prevalence from Phase 1 and incidence from Phases 2 and 3 as the outcomes, with depressive symptoms at Phase 1 as the independent variable and delay discounting at Phase 1 as the mediator.</p> Results <p>Delay discounting was positively associated both with diabetes prevalence (coefficient = 0.170; 95% confidence interval [CI] = 0.066 to 0.278; <i>P</i> = 0.002) and incidence (coefficient = 0.306; 95% CI = 0.098 to 0.540; <i>P</i> = 0.006). Furthermore, through delay discounting, depressive symptoms were indirectly associated with diabetes prevalence (indirect coefficient = 0.091; 95% bootstrap CI = 0.034 to 0.149) and incidence (indirect coefficient = 0.138; 95% bootstrap CI = 0.037 to 0.256), respectively.</p> Conclusions <p>Delay discounting may increase the risk of diabetes onset by mediating the positive association between depressive symptoms and diabetes onset.</p>