LAUSR

Background Understanding the epidemiology of reinfections is crucial for SARS-CoV-2 control over a long period. Aim To evaluate the risk of SARS-CoV-2 reinfection by vaccination status, predominant variant and time after first infection. Methods We conducted a cohort study including all residents in the Reggio Emilia province on 31 December 2019, followed up until 28 February 2022 for SARS-CoV-2 first infection and reinfection after 90 days. Cox models were used to compare risk of first infection vs reinfection, adjusting for age, sex, vaccine doses and comorbidities. Results The cohort included 538,516 residents, 121,154 with first SARS-CoV-2 infections and 3,739 reinfections, most in the Omicron BA.1 period. In the pre-Omicron period, three doses of vaccine reduced risk of reinfection by 89% (95% CI: 87–90), prior infection reduced risk by 90% (95% CI: 88–91), while two doses and infection reduced risk by 98% (95% CI: 96–99). In the Omicron BA.1 period, protection estimates were 53% (95% CI: 52–55), 9% (95% CI: 4–14) and 76% (95% CI: 74–77). Before Omicron, protection from reinfection remained above 80% for up to 15 months; with Omicron BA.1, protection decreased from 71% (95% CI: 65–76) at 5 months to 21% (95% CI: 10–30) at 22 months from the first infection. Omicron BA.1 reinfections showed 48% (95% CI: 10–57) lower risk of severe disease than first infections. Conclusions Natural immunity acquired with previous variants showed low protection against Omicron BA.1. Combined vaccination and natural immunity seems to be more protective against reinfection than either alone. Vaccination of people with prior infection reduced the risk of severe disease.