BACKGROUND: The number of individuals with vision impairment worldwide is increasing because of an ageing population. We aimed to systematically identify studies describing the association between vision impairment and mortality, and to assess the association between vision impairment and all-cause mortality. METHODS: For this systematic review and meta-analysis, we searched MEDLINE (Ovid), Embase, and Global Health database on Feb 1, 2020, for studies published in English between database inception and Feb 1, 2020. We included prospective and retrospective cohort studies that measured the association between vision impairment and all-cause mortality in people aged 40 years or older who were followed up for 1 year or more. In a protocol amendment, we also included randomised controlled trials that met the same criteria as for cohort studies, in which the association between visual impairment and mortality was independent of the study intervention. Studies that did not report age-adjusted mortality data, or that focused only on populations with specific health conditions were excluded. Two reviewers independently assessed study eligibility, extracted the data, and assessed risk of bias. We graded the overall certainty of the evidence using the Grading of Recommendations, Assessment, Development and Evaluations framework. We did a random-effects meta-analysis to calculate pooled maximally adjusted hazard ratios (HRs) for all-cause mortality for individuals with a visual acuity of <6/12 versus those with ≥6/12; <6/18 versus those with ≥6/18; <6/60 versus those with ≥6/18; and <6/60 versus those with ≥6/60. FINDINGS: Our searches identified 3845 articles, of which 28 studies, representing 30 cohorts (446 088 participants) from 12 countries, were included in the systematic review. The meta-analysis included 17 studies, representing 18 cohorts (47 998 participants). There was variability in the methods used to assess and report vision impairment. Pooled HRs for all-cause mortality were 1·29 (95% CI 1·20-1·39) for visual acuity <6/12 versus ≥6/12, with low heterogeneity between studies (n=15; τ2=0·01, I2=31·46%); 1·43 (1·22-1·68) for visual acuity <6/18 versus ≥6/18, with low heterogeneity between studies (n=2; τ2=0·0, I2=0·0%); 1·89 (1·45-2·47) for visual acuity <6/60 versus ≥6/18 (n=1); and 1·02 (0·79-1·32) for visual acuity <6/60 versus ≥6/60 (n=2; τ2=0·02, I2=25·04%). Three studies received an assessment of low risk of bias across all six domains, and six studies had a high risk of bias in one or more domains. Effect sizes were greater for studies that used best-corrected visual acuity compared with those that used presenting visual acuity as the vision assessment method (p=0·0055), but the effect sizes did not vary in terms of risk of bias, study design, or participant-level factors (ie, age). We judged the evidence to be of moderate certainty. INTERPRETATION: The hazard for all-cause mortality was higher in people with vision impairment compared with those that had normal vision or mild vision impairment, and the magnitude of this effect increased with more severe vision impairment. These findings have implications for promoting healthy longevity and achieving the Sustainable Development Goals. FUNDING: Wellcome Trust, Commonwealth Scholarship Commission, National Institutes of Health, Research to Prevent Blindness, the Queen Elizabeth Diamond Jubilee Trust, Moorfields Eye Charity, National Institute for Health Research, Moorfields Biomedical Research Centre, Sightsavers, the Fred Hollows Foundation, the Seva Foundation, the British Council for the Prevention of Blindness, and Christian Blind Mission.