The aim of this systematic review and meta-analysis was to investigate the pooled prevalence of cognitive frailty among community-dwelling older adults and provide evidence-based support for policy-makers planning health and social care policies. A systematic review and meta-analysis. PubMed, Web of Science, Embase and the Cochrane Library were systematically searched from their inception to December 10, 2020. Descriptive studies (cross-sectional studies or population-based longitudinal studies) and cohort studies were available. Participants were community-dwelling older adults aged 60 years and above. Two researchers independently screened the literature, extracted the data and evaluated the quality of the included studies. All statistical analyses were conducted using Stata 15.0. We screened 2815 records, among which 24 studies met the inclusion criteria and were included in the review. The pooled prevalence of cognitive frailty was 9% (95% CI: 8%-11%, I2 = 99.3%). The results of the subgroup analysis showed that the pooled prevalence of cognitive frailty was 11% (95% CI: 9%-14%) in men and 15% (95% CI: 11%-19%) in women. The pooled prevalence of cognitive frailty based on the descriptive studies and cohort studies was 7% (95% CI: 5%-9%) and 17% (95% CI: 11%-22%), respectively. The pooled estimates of cognitive frailty prevalence were 6% (95% CI: 4%-8%) from 2012 to 2017 and 11% (95% CI: 9%-14%) from 2018 to 2020. This systematic review analyzed the available literature and revealed that the pooled prevalence of cognitive frailty among community-dwelling older adults was 9%. The stratified analysis showed that the prevalence of cognitive frailty was higher in older women. In addition, the prevalence has increased in recent years, which has important implications for adapting health and social care systems. Copyright © 2021. Published by Elsevier Ltd.