Prevalence and Levels of Anti-SARS-CoV-2 Antibodies in the Eswatini Population and Subsequent Severity of the Fourth COVID-19 Epidemic Wave

Background: Seroepidemiology studies are useful for quantifying the magnitude of past infections and estimating the extent of population-based immunity to inform risk mitigation strategies for the future. We report on the only national population-based survey of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) immunoglobulin G (IgG) seroprevalence in Eswatini. Methods: The survey was undertaken from 31 August to 30 September 2021, following three earlier waves of coronavirus disease (COVID-19), and preceded the onset of the fourth wave, which was dominated by the Omicron variant of concern. We also report on epidemiological trends of recorded COVID-19 cases and hospitalizations before and after the fourth COVID-19 wave through to March 2022. We evaluated the immunoglobulin G (IgG) seropositivity based on either anti-nucleocapsid (N) or anti-spike (S) antigens. Results: Of 4564 individuals, 58.5% were female, 36.0% were aged 18-50 years, and 863 (18.9%) of adults who were older than 18 years had received at least a single dose of COVID-19 vaccine. Overall, 2769 (60.7%) were seropositive with heterogeneity across sub-regions (53.7%; 95% CI:49.2-58.1 to 68.6%; 95% CI:64.5-72.4), with the highest rates occurring in sub-regions of the Manzini region. Seropositivity was higher in vaccinated individuals (84.5%; 95% CI: 81.9-86.7) compared to unvaccinated individuals (55.1%; 95% CI:53.5-56.7). Amongst unvaccinated individuals, seropositivity was highest in 18-50-year-olds (59.5%;95% CI: 56.9-62.1). Seropositivity was associated with female gender, previous positive SARS-CoV-2 NAAT status and being vaccinated, non-smoking, and being formally employed. We estimated as of 15 September 2021 that there had been 639,475 SARS-CoV-2 infections (95% CI; 620,824-658,003) in Eswatini, which was 25.5-fold greater than the 25,048 COVID-19 cases that had been recorded by then. The national case fatality rate (CFR) based on recorded cases was 4.8%, being 25-fold greater than the infection fatality rate (0.19; 95% CI: 0.18-0.19) based on recorded deaths and extrapolating the force of infection from seroprevalence. Nationally and across all four regions, we report the decoupling of COVID-19 cases from hospitalisations and deaths, observed as early as during the third wave, which was dominated by the Delta variant compared with earlier waves. Conclusions: We identified that 60.7% of people in Eswatini had been infected by SARS-CoV-2 at least once and before the onset of the Omicron wave in mid-November 2021. Despite a modest uptake of COVID-19 vaccines, the evolution of population immunity from infection has likely contributed to the decoupling of infection and severe COVID-19 in Eswatini.