The addictive protection against SARS-CoV-2 reinfection conferred by vaccination, as compared to natural immunity alone, remains to be quantified. We thus carried out a meta-analysis to summarize the existing evidence on the association between SARS-CoV-2 vaccination and the risk of reinfection and disease. We searched MedLine, Scopus and preprint repositories up to July 31, 2022, to retrieve cohort or case-control studies comparing the risk of SARS-CoV-2 reinfection or severe/critical COVID-19 among vaccinated vs. unvaccinated subjects, recovered from a primary episode. Data were combined using a generic inverse-variance approach. Eighteen studies, enrolling 18,132,192 individuals, were included. As compared to the unvaccinated, vaccinated subjects showed a significantly lower likelihood of reinfection (summary Odds Ratio—OR: 0.47; 95% CI: 0.42–0.54). Notably, the results did not change up to 12 months of follow-up, by number of vaccine doses, in studies that adjusted for potential confounders, adopting different reinfection definitions, and with different predominant strains. Once reinfected, vaccinated subjects were also significantly less likely to develop a severe disease (OR: 0.45; 95% CI: 0.38–0.54). Although further studies on the long-term persistence of protection, under the challenge of the new circulating variants, are clearly needed, the present meta-analysis provides solid evidence of a stronger protection of hybrid vs. natural immunity, which may persist during Omicron waves and up to 12 months.