Background: Multisystem inflammatory syndrome in children (MIS-C) occurs after infection with SARS-CoV-2 and its incidence is likely to depend on multiple factors, including the variant of the preceding SARS-CoV-2 infection and vaccine effectiveness. We aimed to estimate the incidence of MIS-C, and describe the clinical phenotype, following the delta variant of SARS-CoV-2 (B.1.617.2 and sublineages) according to vaccination status. We aimed to compare the incidence and clinical phenotype of MIS-C from our cohort during the pre-delta era. Methods: This prospective, population-based cohort study included patients aged 0–17 years hospitalised with MIS-C in Denmark, according to the US Centers for Disease Control and Prevention case definition, from Aug 1, 2021, to Feb 1, 2022, a period dominated by the delta variant. We identified MIS-C cases via a nationwide research collaboration involving real-time data collection from all 18 paediatric departments. Aggregated number of SARS-CoV-2 infections by vaccination status was obtained from the Danish COVID-19 surveillance registries. The incidence of MIS-C was calculated using the estimated number of infected individuals by vaccination status. We calculated the incidence of MIS-C per 1 000 000 vaccinated and unvaccinated person-years, and estimated vaccine effectiveness as 1–incidence rate ratio using Poisson regression. Incidence and phenotype of MIS-C were compared with MIS-C cases from the first year of the pandemic. This study is registered at ClinicalTrials.gov, NCT05186597. Findings: We identified 51 MIS-C cases among unvaccinated individuals and one in a fully vaccinated adolescent. The incidence of MIS-C was one in 3400 unvaccinated individuals (95% CI 2600–4600) with the delta variant and one in 9900 vaccinated individuals (95% CI 1800–390 000) with breakthrough infection. The estimated vaccine effectiveness against MIS-C after the delta variant was 94% (95% CI 55–99; p=0·0061) in individuals aged 5–17 years. The clinical phenotype during the delta wave was comparable to the pre-delta era. Interpretation: We found the incidence and phenotype of MIS-C in unvaccinated children during the delta wave to be similar to the incidence during the first year of the pandemic. We found vaccine effectiveness to be high against MIS-C, which we suggest was due to protection from infection and, possibly, a decreased incidence of MIS-C after breakthrough infection. Knowledge of the incidence of MIS-C after different SARS-CoV-2 variants and the effect of vaccination might contribute to the elucidation of the extent to which MIS-C is a vaccine-preventable disease. Funding: National Ministry of Higher Education and Science and Innovation Fund Denmark.