Inconsistent findings have been reported on the association between high-sensitivity C-reactive protein (hs-CRP) and mortality risk. The objective of this meta-analysis was to investigate the association of elevated baseline hs-CRP levels with all-cause, cardiovascular, and cancer mortality risk in the general population. PubMed and Embase were systematically searched for studies published from inception to October 2016. Prospective observational studies were eligible if they reported the effects of elevated baseline hs-CRP levels on cancer-related, cardiovascular or all-cause mortality in the general population. The pooled adjusted risk ratio (RR) with 95% confidence interval (CI) comparing the highest to the lowest category of hs-CRP levels was used as association measures. A total of 83,995 participants from 14 studies were identified. When comparing the highest to the lowest category of hs-CRP levels, the pooled RR was 1.25 (95% CI 1.13-1.38) for cancer-related mortality, 2.03 (95% CI 1.65-2.50) for cardiovascular mortality, and 1.75 (1.55-1.98) for all-cause mortality, respectively. Subgroup analysis showed that the effect of elevated hs-CRP levels on cancer-related mortality was observed in men (RR 1.26; 95% CI 1.11-1.43) but not in women (RR 1.03; 95% CI 0.83-1.27). Elevated hs-CRP levels can independently predict risk of all-cause, cardiovascular mortality in the general population. However, the gender differences in the predictive role of hs-CRP on cancer mortality should to be further investigated.