Background The association of the meal timing of dietary total antioxidant capacity (DAC) with mortality is unclear. We aimed to investigate the association between the meal timing of DAC and all-cause, cardiovascular disease (CVD), and cancer mortality in general adult populations.Methods A total of 56,066 adults who participated in the US National Health and Nutrition Examination Survey (NHANES) from 1999 to 2018 were recruited for this study. Dietary intake (quantity and timing) was evaluated by nonconsecutive 24-h dietary recalls. The main exposure variables were the DAC across three meals (total, breakfast, lunch, and dinner; without coffee) and the difference between dinner and breakfast DAC (Δ=dinner-breakfast; without coffee). The outcomes were all-cause, CVD, and cancer mortality. The adjusted hazard ratios [aHRs] and 95% confidence intervals [CI] were imputed by Cox proportional hazards regression.Results Among the 56,066 participants, there were 8566 deaths from any cause, including 2196 from CVD and 1984 from cancer causes. Compared to participants in the lowest quintiles of the total DAC, those in the highest quintiles had 34% and 27% decreased risks of all-cause and CVD mortality, respectively (all-cause mortality: aHRs 0.66 [95% CI 0.57–0.76]; CVD mortality: aHRs 0.73 [95% CI 0.57–0.94]). More importantly, participants in the highest quintiles of the dinner DAC, but not those in that of breakfast or lunch, had a 24% decrease in all-cause mortality (aHRs 0.76 [95% CI 0.67–0.87]) compared with those in the lowest quintiles. Inverse associations were further confirmed for Δ DAC (aHRs 0.84 [95% CI 0.74–0.96]). Above associations did not change when including DAC from snacks or tea. Mediation analysis showed that the total associations of total, dinner or Δ DACs with reduced all-cause mortality were 24%, 13% and 6%, respectively, mediated by serum CRP. Additionally, all-cause mortality was decreased by 7% in models.