Although a growing number of original epidemiological studies imply a link between ambient pollution exposure and mortality risk, the fndings associated with carbon monoxide (CO) exposure are inconsistent. Thus, we conducted a systematic review and meta-analysis of epidemiological studies to evaluate the correlations between ambient CO and non-accidental, cardiovascular, and respiratory mortality in China. Eight databases were searched from inception to 15 May 2021. A randomefect model was used to calculate the pooled relative risks (RRs) and 95% confdence intervals (CIs). Subgroup analyses as well as sensitivity analyses were performed. The I square value (I 2 ) was used to assess heterogeneity among diferent studies. The assessment of publication bias on included studies was examined by funnel plot and Egger’s test. The infuence of a potential publication bias on fndings was explored by using the trim-and-fll procedure. Ultimately, a total of 19 studies were included in our analysis. The pooled relative risk for each 1 mg/m3 increase of ambient carbon monoxide was 1.0220 (95%CI: 1.0102–1.0339) for non-accidental mortality, 1.0304 (95%CI:1.0154–1.0457) for cardiovascular mortality, and 1.0318 (95%CI:1.0132–1.0506) for respiratory mortality. None of subgroup analyses could explain the source of heterogeneity. Exclusion of any single study did not materially alter the pooled efect estimates. Although it was suggestive of publication bias, fndings were generally similar with principal fndings when we explored the infuence of a potential publication bias using the trim-and-fll method. Our meta-analysis demonstrated that exposure to ambient CO was positive with risk of deaths from all non-accidental causes, total cardiovascular, and respiratory diseases. Based on these fndings, tougher intervention policies and initiatives to reduce the health efects of CO exposure should be established.