Differential COVID-19 case ascertainment by age and vaccination status in Victoria, Australia: a serosurveillance and record linkage study

Abstract

Objectives: To compare serological evidence of prior severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection with linked coronavirus disease 2019 (COVID-19) case notification data in Victoria, Australia, and to determine in vitro SARS-CoV-2 neutralisation activity based on prior infection and vaccination history. Design, setting, participants: Four cross-sectional serological surveys were conducted between 30 June and 31 October 2022 (a period of Omicron BA.4/BA.5 dominance) using 1,974 residual serum samples obtained from the Victorian Infectious Diseases Reference Laboratory. Serological results were linked to COVID-19 case notification and vaccination data. Surrogate virus neutralisation testing was performed to obtain in vitro inhibition estimates by anti-nucleocapsid serostatus and COVID-19 vaccination history. Main outcome measures: Adjusted anti-SARS-CoV-2 spike and nucleocapsid seropositivity by sex, age and region of residence; adjusted proportion of cases notified by anti-nucleocapsid serostatus, age and number of COVID-19 vaccination doses received; adjusted percentage in vitro inhibition against wildtype and Omicron BA.4/BA.5 SARS-CoV-2 variants by anti-nucleocapsid serostatus and COVID-19 vaccination history. Results: The prevalence of anti-SARS-CoV-2 nucleocapsid antibodies was inversely proportional to age. In October 2022, prevalence was 84% (95% confidence interval [95% CI]: 75–93%) among 18–29-year-olds, compared to 39% (95% CI: 27–52%) among ≥ 80-year-olds. In most age groups, approximately 40% of COVID-19 cases appear to have been notified via existing surveillance mechanisms. Case notification was highest among individuals older than 80 years and people who had received COVID-19 vaccine booster doses. In vitro neutralisation of Omicron BA.4/BA.5 sub-variants was highest for individuals with evidence of both prior infection and booster vaccination. Conclusions: Under-notification of SARS-CoV-2 infections in the Victorian population is not uniform across age and vaccination strata. Seroprevalence data that give insights into case notification behaviour provide additional context for the interpretation of existing COVID-19 surveillance information.