South African Population Immunity and Severe Covid-19 with Omicron Variant

Background We conducted a seroepidemiological survey from October 22 to December 9, 2021, in Gauteng Province, South Africa, to determine SARS-CoV-2 immunoglobulin G (IgG) seroprevalence primarily before the fourth wave of coronavirus disease 2019 (Covid-19), in which the B.1.1.529 (Omicron) variant was dominant. We evaluated epidemiological trends in case rates and rates of severe disease through to January 12, 2022, in Gauteng. Methods We contacted households from a previous seroepidemiological survey conducted from November 2020 to January 2021, plus an additional 10% of households using the same sampling framework. Dry blood spot samples were tested for anti-spike and anti-nucleocapsid protein IgG using quantitative assays on the Luminex platform. Daily case, hospital admission, and reported death data, and weekly excess deaths, were plotted over time. Results Samples were obtained from 7010 individuals, of whom 1319 (18.8%) had received a Covid-19 vaccine. Overall seroprevalence ranged from 56.2% (95% confidence interval [CI], 52.6 to 59.7) in children aged <12 years to 79.7% (95% CI, 77.6 to 81.5) in individuals aged >50 years. Seropositivity was more likely in vaccinated (93.1%) vs unvaccinated (68.4%) individuals. Epidemiological data showed SARS-CoV-2 infection rates increased and subsequently declined more rapidly than in previous waves. Infection rates were decoupled from Covid-19 hospitalizations, recorded deaths, and excess deaths relative to the previous three waves. Conclusions Widespread underlying SARS-CoV-2 seropositivity was observed in Gauteng Province before the Omicron-dominant wave. Epidemiological data showed a decoupling of hospitalization and death rates from infection rate during Omicron circulation.

: Demographics of sampling area in serosurvey 14

Study design and sample size
This survey included households that had been sampled during the first seroprevalence survey, which was conducted done from November 4, 2020, to January 22, 2021, 1 and included an additional 10% households per cluster to accommodate for non-participation. The sample size calculation was based on the Africa Centres for Disease Control and Prevention generic protocol for a population-based, age-and gender-stratified seroprevalence survey study for SARS-CoV-2 2 and on the World Health Organization population-based age-stratified seroepidemiological investigation protocol for COVID-19 virus infection. 3 Assuming a seroprevalence of 10%, response rate of 0.75, intra-cluster correlation (ICC) of 0.33, with a precision of 0.1, α of 0.05, and design effect of 3.31, the resultant required overall survey sample size was 6025 (61 to 1948 by sub-district) individuals. Design effect of 3.31 was based on the observation that seroprevalence rates vary by geographic area within the same region. 2 A conservative ICC of 0.33 was used considering clustering nature of respiratory diseases.
Three-stage sampling was employed; first, the GeoTeraImage (GTI)-2019 dataset (https://geoterraimage.com/), which comprises more than 16,000 "small areas" used for demarcating census areas was stratified by housing type. Systematic random cluster selection without replacement and probability proportional to estimated size (PPES) was then used to select clusters. Finally, a random sample of 9 households was selected from each cluster. The number of clusters, and hence the number of households selected per sub-district, was proportional to the sub-district population size. The average household size was assumed to be 4 individuals based on census data with a target of 9 households per cluster. All individuals residing in sampled households, irrespective of age, were eligible.

Data collection
Electronic data collection was done using android tablets with real time synchronisation of data into a

Sample collection and processing
Dried blood spots (DBS) were obtained by finger-prick with a single-use lancet needle, with 3-5 DBS collected on filter cards (Ahistrom Munksjo, Germany, catalogue number 8.460.0013.A). The DBS were dried for 3 hours at room temperature, packed into plastic pouches with silica-gel sachets, transported to the laboratory, and stored at -20°C until analysis. 1 Elution of antibodies from DBS card specimens was performed as previously described. 4 Briefly, one spot was cut from the filter card using a 6 mm hole punch and added to 600 μl of assay buffer. The spot was kept in a shaker at 2-8°C overnight for elution and the following day was centrifuged at 2000 g for 10 min before analysis.

Serology analysis
SARS-CoV-2 full-length spike and nucleocapsid (N) protein immunoglobulin G (IgG) were measured by quantitative assay on the Luminex platform. The expression plasmid encoding for the SARS-CoV-2 fulllength spike were obtained from Florian Krammer, Mount Sinai, USA. The recombinant spike protein was expressed as described previously, 5 and N protein was obtained from BioTech, Africa (Cat no# BA25-P, South Africa). Proteins were coupled to magnetic microsphere beads (Bio-Rad, USA) using a two-step carbodiimide reaction. 6 Samples were analyzed in single-plex, and each plate included two in- respectively. Bland-Altman assessment showed good agreement between two sampling methods. 4 Serum samples collected prior to 2020 (n = 31) and serum samples obtained from randomly selected (n = 15) participants who were SARS-CoV-2-positive on PCR test and who had serial blood sampling before and after symptom onset (including individuals with mild-to-moderate illness and asymptomatic infections) were used for the analysis of assay specificity and sensitivity, and for calculation of the threshold for IgG seropositivity. Based on IgG titers from pre-Covid-19 era samples, baseline and postinfection samples of participants who were SARS-CoV-2-positive on PCR test, 32 BAU/mL and 15 BAU/mL were selected as the thresholds indicative of seropositivity for full-length spike and N protein IgG, respectively. The sensitivity of the assay to detect seropositivity at the selected thresholds was 100% (15/15) and 85% (12/14) for samples taken >14 days following the SARS-CoV-2-positive PCR test for full-length spike and N protein IgG, respectively.

Covid-19 data sources
Daily case, hospital admission, and reported death data were sourced from the DATCOV database, which is hosted by South Africa's National Institute for Communicable Diseases, 7,8 as described previously. 9 The system was developed during the course of the first wave of Covid-19, with gradual onboarding of facilities. Hence, data from DATCOV could underestimate hospitalized cases in the first wave relative to subsequent waves. The hospitalized cases include individuals with Covid-19 as well as coincidental infections identified as part of routine testing for SARS-CoV-2 of individuals admitted to the facilities to assist in triaging of patients in the hospital.
DATCOV was initially implemented in eight sentinel public hospital sites but within six months was Data on excess weekly mortality were sourced from the weekly report by the Burden of Disease Research Unit of the South African Medical Research Council. 11 Within the glossary to each report, the following statements are provided regarding the definition and calculation of excess mortality rates: • "Excess deaths: There is no universal definition of, or understanding of what is meant by, "excess mortality". It is a term used in epidemiology and public health that refers to the number of deaths that are occurring above what we would normally expect. The WHO uses the term to describe "Mortality above what would be expected based on the non-crisis mortality rate in the population of interest. Excess mortality is thus mortality that is attributable to the crisis conditions. It can be expressed as a rate (the difference between observed and non-crisis mortality rates), or as a total number of excess deaths." • "Excess natural deaths associated with COVID-19: Generally, the number of excess deaths per week is calculated as the number of all-cause deaths in that week less the number that might be assumed to have occurred had there not been the epidemic (i.e. the counterfactual number), provided that the counterfactual is lower. However, this approach has generally only been applied to countries where deaths have been tracking the counterfactual before the onset of

Statistical analyses
We estimated anti-SARS-CoV-2 seroprevalence as the proportion of individuals testing positive for either spike or N protein IgG and assessed variability by age, gender, vaccination status, and sub-district of residence. Factors associated with seropositivity were determined using generalized linear models with log link to estimate risk ratios. These were unadjusted, univariable analyses for each risk factor.
Confidence intervals have not been adjusted for multiplicity and should not be used for inference.
Comorbidities included any of self-reported hypertension, diabetes, asthma, HIV positive status, cancer, tuberculosis, stroke, or lung, liver, kidney, or heart disease. Stata (version 16.1) was used for data analyses, with sampling weights calculated based on the sampling frame. Data analyses techniques used clustering adjusted for at the household level.
For epidemiological analyses, daily rates were smoothed using locally weighted scatterplot smoothing (Lowess) in STATA version 16.1 using a bandwidth of 0.06, which closely matches the 7-day moving average but which provides a smoother graph line, allowing the minimum values at the beginning and end of each Covid-19 wave to be easily identified. These minima are taken as cut-points for each wave, with no overlap and no data omitted from analysis. Raw counts were obtained at these cut-points for cumulative numbers for each wave. Weekly excess death rates were unsmoothed. Data on excess mortality were unavailable by age group and sex.    Figure S4: Covid-19 daily case rates, weekly hospital admission rates, weekly excess death rates, and daily reported death rates over the time period of the pandemic in the five districts of Gauteng Province, South Africa, as of January 12, 2022 All data are from the National Institute for Communicable Diseases daily databases except for weekly excess deaths. Excess mortality from natural causes was defined per and sourced from the South African Medical Research Council; the excess mortality data are reported through to January 8, 2021. 11 Stratified excess mortality data were unavailable for Sedibeng and West Rand. 11 The solid vertical black line represents the start of the fourth, Omicron-dominant wave on November 15, 2022. Changes in testing rates, particularly the lower rates during Wave 1 due to constraints in laboratory capacity and prioritization of testing for hospitalized individuals, prevent direct comparisons, especially in terms of case numbers during the first wave in relation to the subsequent waves. Cases include asymptomatic and symptomatic individuals. Cumulative reported cases were sourced from the National Department of Health. 12 Hospitalization data are from DATCOV, hosted by the National Institute for Communicable Disease, 7 as described previously. 9 The system was developed during the course of the first wave, with gradual onboarding of facilities; hence, these data could underestimate hospitalized cases in the first wave relative to subsequent waves. The hospitalized cases include individuals with Covid-19, as well as coincidental infections identified as part of routine testing for SARS-CoV-2 of individuals admitted to the facilities to assist in triaging of patients in the hospital. Cumulative reported deaths were sourced from the National Department of Health. 12 As absolute rates differ between districts, different Y-axis scales have been used for each individual district in order to provide clarity and aid visual interpretation of the trends in each district.  Covid-19 incidence (cases per 100,000 persons) increases with age; the lowest incidence is in those aged <5years (862.4), and incidence increases linearly with increase in age, with the highest incidence in those aged 50-54 years (11,160.6). Covid-19 incidence in those aged >60 years is lower than in those aged 20-59 years Race or ethnic group

Johannesburg District
There is no evidence that Covid-19 disproportionately affects people of a specific race or ethnicity. In South Africa, 81% of the population is Black African. Geography Covid-19 incidence varies by geographic area. 1,13,14 Gauteng Province has the highest incidence rate in South Africa, accounting for 33% of the total cases in the country. Across the world, Covid-19 incidence and severity vary widely associated with underlying population structures; countries with older populations have more Covid-19 cases.

Socio-economic status
In two seroprevalence surveys conducted in Gauteng Province, inclusive of this one, there was lower seroprevalence in informal settlements compared to formal stand-alone houses and higher seroprevalence in blocks of flats/high-rise buildings. In South Africa and Gauteng Province, females constitute 51.2% and 49.9% of the population, respectively. Biological gender at birth was reported by the participants and captured as male, female, or declined to respond. This survey comprised 58% females, and thus slightly overrepresented females. Covid-19 incidence is higher in older populations and Africa generally has younger populations compared to Europe and America. 73.6% of this survey population were aged 15-59 years comparable to 67.7% in Gauteng Province and South Africa. Our survey sample was representative of the underlying population residential types, an indicator for socio-economic status. Seroprevalence estimates by age and gender from this survey are consistent with Covid-19 epidemiological data in South Africa. 13

Participants in survey
Anti  *All data are from the National Institute for Communicable Diseases daily databases. The Omicron-dominant fourth case wave is at its tail-end but has not yet fully subsided. Totals, incidence, and proportions of cases are anticipated to continue to increase somewhat over the next few weeks until the wave has fully subsided. †Changes in testing rates, particularly the lower rates during Wave 1 due to constraints in laboratory capacity and prioritization of testing for hospitalized individuals, prevent direct comparisons, especially in terms of case numbers during the first wave in relation to the subsequent waves. Cases include asymptomatic and symptomatic individuals. Cumulative reported cases were sourced from the National Department of Health. 12 ‡ Sex was not recorded for 7895 Covid-19 cases, who are excluded from analyses by sex. *All data are from DATCOV, hosted by the National Institute for Communicable Disease, 7 as described previously. 9 The system was developed during the course of the first wave, with gradual onboarding of facilities; hence, these data could underestimate hospitalized cases in the first wave relative to subsequent waves. The hospitalized cases include individuals with Covid-19, as well as coincidental infections identified as part of routine testing for SARS-CoV-2 of individuals admitted to the facilities to assist in triaging of patients in the hospital. The Omicron-dominant fourth case wave is at its tail-end but has not yet fully subsided. Totals, incidence, and proportions of hospitalizations are anticipated to continue to increase somewhat over the next few weeks until the wave has fully subsided. †Changes in testing rates, particularly the lower rates during Wave 1 due to constraints in laboratory capacity and prioritization of testing for hospitalized individuals, prevent direct comparisons, especially in terms of case numbers during the first wave in relation to the subsequent waves. ‡ Sex was not recorded for 162 hospitalizations, who are excluded from analyses by sex. *All data are from the National Department of Health. 12 The Omicron-dominant fourth case wave is at its tail-end but has not yet fully subsided. Totals, incidence, and proportions of deaths are anticipated to continue to increase somewhat over the next few weeks until the waves has fully subsided. † Sex was not recorded for 22 recorded deaths, who are excluded from analyses by sex.