Background: While the world's population ages, social isolation has continued to increase globally. However, no review exists on the prevalence of social isolation among community-dwelling older adults, and the global prevalence remains uncertain. This study aims to estimate the global prevalence of social isolation among community-dwelling older adults and to identify potential covariates including study characteristics (methodological diversity) or populations (clinical diversity) that contribute to the heterogeneity.

Methods: This review searched through seven search engines and databases. The meta-analysis was conducted using the metafor package in the R software. The random-effects model was used to calculate the prevalence rates. Cochran's Q statistics and I(2) statistics were used to assess the statistical heterogeneity of prevalence estimates. Studies were appraised using the Newcastle-Ottawa Scale and the Grading of Recommendations, Assessment, Development, and Evaluations criteria for the quality of individual articles and the certainty of the evidence, respectively.

Results: A total of 41 studies were selected from databases and reference lists. The pooled prevalence rate was 25% (95% CI: 21.0-30.0). The sample size was found to be a significant covariate of the prevalence estimate in the subgroup analysis.

Conclusions: We found 13 high-quality studies, but the overall quality of evidence very low. This study provides the prevalence of social isolation in community-dwelling older adults, identifying vulnerable groups for targeted intervention. Well-designed observational research with standard measures is recommended for future studies.