The present meta-analysis aimed to investigate fiber consumption and all-cause mortality, and cause-specific mortality. MEDLINE and web of science database were searched for cohort studies published from inception to August 2014. Studies were included if they provided a hazard ratio (HR) and corresponding 95% CI for mortality in relation to fiber consumption.We found that, compared with those who consumed lowest fiber, for individuals who ate highest fiber, mortality rate was lower by 23% (HR, 0.77; 95% CI, 0.72–0.81) for CVD, by 17% (HR, 0.83; 95% CI, 0.74–0.91) for cancer, by 23% (HR, 0.77; 95% CI, 0.73–0.81) for all-cause mortality. For each 10 g/day increase in fiber intake, the pooled HR was estimated to be 0.89 (95% CI, 0.86–0.93) for all-cause mortality, 0.80 (95% CI, 0.72–0.88) for CHD mortality, and 0.66 (95% CI, 0.40–0.92) for IHD mortality, 0.91 (95% CI, 0.88–0.94) for cancer. Dietary fiber and CVD mortality showed a strong dose–response relation. Apparently, fiber consumption is inversely associated with all-cause mortality and CVD, IHD, cancer mortality.