Abstract
Background Our understanding of the global scale of SARS-CoV-2 infection remains incomplete: routine surveillance data underestimates infection, cannot infer on population immunity, there is a predominance of asymptomatic infections, and uneven access to diagnostics. We meta-analyzed standardized SARS-CoV-2 seroprevalence studies after two years in the pandemic to estimate the extent of population infection and remaining susceptibility.
Objectives/Methods We conducted a systematic review and meta-analysis, searching MEDLINE, Embase, Web of Science, preprints, and grey literature for SARS-CoV-2 seroprevalence published between 2020-01-01 and 2021-12-30. Eligible studies - those aligned with the WHO UNITY protocol - were extracted and critically appraised in duplicate. We meta-analyzed seroprevalence by country and month, pooling to estimate regional and global seroprevalence over time; compared seroprevalence from infection to confirmed cases to estimate under-ascertainment; meta-analyzed differences in seroprevalence between demographic subgroups; and identified national factors associated with seroprevalence using meta-regression. PROSPERO: CRD42020183634.
Results We identified 431 full texts reporting 803 distinct seroprevalence studies (43% LMIC), including 398 low/moderate risk of bias studies with national/sub-national scope in further analysis. By July 2021, global SARS-CoV-2 seroprevalence was 45.2%, 95% CI [40.7-49.8%]. Seroprevalence rose steeply in the first half of 2021 due to infection in some regions (e.g., 29.9% to 70.1% in Africa) and vaccination and infection in others (e.g., 5.6% to 94.9% in the Americas high-income countries), but remained low in others (e.g., 2.5% in the Western Pacific). In 2021 Q2, median seroprevalence to cumulative incidence ratios were 2.9:1 in HICs and 45.3:1 in LMICs. Children 0-9 years and adults 60+ were at lower risk of seropositivity than adults 20-29. In a multivariate model using pre-vaccination data, stringent public health and social measures were associated with lower seroprevalence.
Conclusions Global seroprevalence has risen considerably over time and with regional variation, however much of the global population remains susceptible to SARS-CoV-2 infection. Our estimates of infections based on seroprevalence far exceed reported COVID-19 cases. Quality and standardized seroprevalence studies are essential to inform COVID-19 response, particularly in resource-limited regions.
Introduction
The COVID-19 pandemic, caused by the SARS-CoV-2 virus, continues to severely impact population health and health care systems. The 394 million cases and 5.7 million deaths reported as of 7 February 2022(1) underestimate the global burden of this pandemic, particularly in low- and middle-income countries (LMICs) with limited capacity for contact tracing, diagnostic testing, and surveillance capacity.(2)
Seroprevalence studies estimate the prevalence of SARS-CoV-2 antibodies. These studies are crucial to understand the true extent of infection overall, by demographic group, and by geographic area, as well as to estimate case underascertainment. As anti-SARS-CoV-2 antibodies are highly predictive of immune protection,(3,4) seroprevalence studies are also indicative of population levels of protection, and therefore important to inform scenario modeling, public health planning, and national policies in response to the pandemic.
During 2021, many regions have experienced third and fourth waves of SARS-CoV-2 infection(1); concurrently, some countries have vaccinated most residents, while others remain unable to achieve high vaccine coverage due to challenges with supply and uptake.(5) A new wave of well conducted seroprevalence studies, including many in LMICs, provides robust estimates of seroprevalence in late 2020 and into 2021.(6–8) Synthesizing these studies is crucial to understand the shifting global dynamics and true extent of SARS-CoV-2 infection, humoral immunity, and population susceptibility. While previous global systematic reviews of seroprevalence have been conducted,(9–12) these have included only studies that sampled participants in 2020 and pooled seroprevalence across all time points. These meta-analyses also highlight the importance of improved standardization and study quality to enable more robust analysis.(9–11)
Estimates of seroprevalence can be difficult to compare systematically across different settings due to variations in design aspects including sampled populations, testing and analytical methods, timing in relation to waves of infection, and study quality and reporting. The World Health Organization’s UNITY Initiative aims to help produce harmonized and representative seroprevalence study results in accordance with global equity principles.(2) The UNITY population-based, age-stratified seroepidemiological investigation protocol (the SEROPREV protocol) (2) provides a standard study design and laboratory approach to general population seroprevalence studies. WHO UNITY and its partners have supported the implementation of SEROPREV by providing financial and technical resources, including a well-performing serologic assay. SEROPREV has been implemented in 74 countries globally and in 51 LMICs as of September 2021.(2) Synthesizing results aligned with the standard SEROPREV protocol improves study comparability, enabling further analysis of these comparable studies to answer key questions about the progress of the pandemic globally.
This systematic review and meta-analysis synthesized seroprevalence studies worldwide aligned with the WHO SEROPREV protocol, regardless of whether the study received support from WHO. Our objectives were to: (i) estimate changes in global and regional seroprevalence over time by WHO region and country income level; (ii) assess the level of undetected infection, by global and regional case ascertainment over time by calculating the ratio of seroprevalence to cumulative incidence of confirmed cases; and (iii) identify factors associated with seropositivity including demographic differences by 10-year age band and sex through meta-analysis, and study design and country-level differences through meta-regression.
Methods
Search strategy and study selection
We conducted a systematic review of seroprevalence studies (hereafter “studies”) published from 1 January 2020 to 29 October 2021. We designed a search strategy in MEDLINE, Embase, Web of Science, and Europe PMC using key terms such as SARS-COV-2, COVID-19, seroprevalence, and serology; We included published research articles, preprints, institutional reports, grey literature, and media reports (full strategy in Supplementary file S.3.1). We also contacted WHO UNITY study collaborators that had not yet made results available to the general public prior to our inclusion dates, to upload their aggregated, standardized results to the Zenodo research data repository(13). This systematic review and meta-analysis is registered with PROSPERO (CRD42020183634),(14) reported according to the MOOSE Checklist(15) (Supplementary file S1), and search and extraction conducted per the SeroTracker protocol.(16)
Studies were screened, data extracted, and critically appraised by two independent researchers. Conflicts were resolved by consensus. Inclusion and exclusion criteria were designed to align with the SEROPREV protocol (Supplementary File S2.2 and S2.3). We included cross-sectional or longitudinal cohort studies with the objective of estimating SARS-CoV-2 seroprevalence in the general population (see S.2.1 for list of included sampling frames). Both random or non-random (i.e., convenience, sequential, quota) sampling methods were included. Studies had to use serological assays with at least 90% sensitivity and 97% specificity as reported by the manufacturer or study authors (Supplementary file S2.1), unless conducted in vulnerable countries as defined in the Global Humanitarian Response Plan (HRP).(17) Multi-assay testing algorithms were included if the combined sensitivity and specificity met these performance thresholds.(18) We excluded studies sampling specific closed populations (such as prisons, care homes, or other single-institution populations), recruiting participants without a clear sampling frame approximating the target population or testing strategy, and studies that excluded people previously diagnosed with or vaccinated against COVID-19 after initial sampling.
Data extraction, synthesis, and analysis
From each study, we extracted seroprevalence estimates for the overall sample, and stratified by age, sex, vaccination status, and timing of specimen collection. We extracted information on study population, laboratory assay used, any corrections made in estimating seroprevalence (e.g., for population or assay performance), seroprevalence, and denominator. Standardized results uploaded to Zenodo by UNITY study collaborators additionally included information on the proportion of asymptomatic seropositive individuals.
We critically appraised all studies using a modified version of the Joanna Briggs Institute (JBI) checklist for prevalence studies.(19) To assess risk of bias, a decision rule assigned a rating of low, moderate, or high risk of bias based on the specific combination of JBI checklist ratings for that study.(20) This decision rule was developed based on guidance on estimating disease prevalence (21,22) and was validated against overall risk of bias assessments derived manually by two independent reviewers for previously collected seroprevalence studies in the SeroTracker database, showing good agreement with manual review (intraclass correlation 0.77, 95% CI 0.74-0.80; n = 2070 studies).(20)
We classified seroprevalence studies by geographical scope (local, sub-national, or national), sample frame, sampling method, and type of serological assay (Supplement S2.1, Table 5). Where multiple summary estimates were available per study, we prioritized estimates based on estimate adjustment, antibody isotypes measured, test type used, and antibody targets measured (full details: Supplement S3.1). We included multiple estimates per study when broken down by time frame in our analysis over time.
Countries were classified according to WHO region,(23) vulnerability via HRP status,(17) and World Bank income level. We stratified the European Region (EUR) and Region of the Americas (AMR) by high-income countries (HIC) and LMICs due to inter-country diversity.(24) We pooled HIC and LMIC together in the Eastern Mediterranean (EMR) and Western Pacific regions (WPR) due to the lower number of studies, and in the Africa (AFR) and South-East Asia regions (SEAR) (the only two HIC in these regions had no studies).
As all non-HRP studies used assays with at least 90% sensitivity and at least 97% specificity, we used seroprevalence estimates uncorrected for test characteristics. We anchored each estimate to the date halfway between sampling start and end (“sampling midpoint date”) to best reflect the time period of the study. To select the most representative and high quality studies for analysis, we used only sub-national or national studies rated low or moderate risk of bias to estimate seroprevalence in the general population over time and identify factors associated with seroprevalence (sub-dataset 1). We used only national studies rated low or moderate risk of bias to estimate case ascertainment (sub-dataset 2).
To estimate monthly regional and global weighted seroprevalence in the general population, we grouped studies in 12-week rolling windows and pooled effect sizes by random-effects meta-analysis.(25) Global estimates were weighted averages of regional estimates (full details: Supplement S3.2). We produced 95% confidence intervals for the mean seroprevalence estimate, reflecting uncertainty in the summary effect size.(26) To capture the trend in seroprevalence, we used nonparametric regression to fit a smooth function of time to the meta-analysis estimates (full details: Supplement S3.2).
We also estimated to what extent laboratory confirmed SARS-CoV-2 cases(27) underestimated the full extent of infections based on seroprevalence. For studies that sampled participants in 2021, we used national seroprevalence estimates and vaccination rates(28) to calculate seroprevalence attributable to infection only. In countries administering only vaccines using Spike (S) protein antigens (e.g., mRNA), we calculated the ascertainment ratio using only studies that detected anti-nucleocapsid (N) seroprevalence. In countries administering inactivated vaccines that may generate both anti-S and anti-N responses, we adjusted the reported seroprevalence using a standard formula.(29) We then produced regional and global estimates of seroprevalence as above, and computed the ratio to the corresponding cumulative incidence of confirmed SARS-CoV-2 cases. We stratified by HIC vs. LMIC in all regions.
Aggregated results shared by UNITY collaborators reported the proportion of seropositives that were symptomatic at some time point prior to sampling, summarized using the median and interquartile range, and tested for differences in distribution across age and sex groups using the Kruskal-Wallis H-test.
To quantify population differences in SARS-CoV-2 seroprevalence, we identified studies with seroprevalence estimates for sex and age subgroups. We calculated the ratio in seroprevalence between groups within each study, comparing each age group to adults 20-29 and males to females. We then aggregated the ratios across studies using inverse variance-weighted random-effects meta-analysis. The amount of variation attributable to between-study heterogeneity vs. within-study variance was quantified using the I2 statistic.
To examine study and national factors affecting seroprevalence estimates, we constructed a Poisson generalized linear mixed-effects model.(30) Independent predictors were defined a priori as WHO region, income group, geographic scope, sample frame, pandemic timing, cumulative confirmed cases, and average public health and social measure (PHSM) stringency index.(31) To focus on factors associated with seroprevalence from infection, we excluded studies where over 5% of the national population was vaccinated two weeks before the sampling midpoint date. We compared the full model to all models dropping a single predictor at a time and selected the model with the lowest Akaike Information Criterion (AIC) (full details on the model and predictor definitions: Supplement S3.2).
Data was analyzed using R statistical software version 4.1.0.(32)
Results
We identified 73,348 titles and abstracts in our search (Figure 1). Of these, 4,221 full text articles were included in full text screening. 431 seroprevalence data sources containing studies aligned with the SEROPREV protocol were identified, 400 published and 31 aggregated results from collaborators, that contained a total of 803 unique seroprevalence studies included from 431 full-text sources (detailed references and information available in Supplement, S4.1-S4.3, S5).
In cases where sources contained multiple primary estimates of seroprevalence (i.e. in non-overlapping populations, separate methodological seroprevalence studies reported in the same article, etc), the source (full text) was split int multiple individual studies included in the analysis. For this reason, we report more unique seroprevalence studies than original full text articles included.
A total of 50% (97/194) of WHO Member States (MS) and four WHO Countries, areas and territories, across all 6 WHO regions, were represented among the seroprevalence studies included in the descriptive analysis (Figure S1). Twenty one of 47 MS were represented in AFR; 10 of 21 MS and one territory in EMR; 13 of 35 MS and one territory in AMR; 40 of 53 MS and two territories in EUR; 5 of 11 MS in SEAR; and 8 of 27 MS in WPR (Figure S1). Data from 59 of 135 LMICs and from 33 of 63 vulnerable HRP countries were included. A large proportion of studies included in the descriptive analysis were conducted in LMIC (43%) and in vulnerable HRP countries (21%). Of studies included in the meta-analysis and meta-regression, these proportions were 32% and 15%, respectively.
Among the 803 studies included in the descriptive analysis, 41% reported results at a local level, 36% at a national level, and 23% at a sub-national level. The most common sampling frame and method was households (52%) and probability sampling (55%), respectively. Among the testing strategies used to measure seroprevalence, most studies used ELISA (39%) or CLIA assays (33%) and few studies used a lateral flow immunoassay (10%) or multiple assay testing algorithm (8.6%). The majority of studies (82%) had no vaccination at the sampling midpoint date in the country of the study (Table 1).
Most (51%) studies were rated moderate risk of bias. A summary of overall risk of bias ratings and breakdown of each risk of bias indicator for all studies is available (Figure S2 and Table S8, respectively). In this meta-analysis of observational studies, most studies were not pre-registered and the seroprevalence estimated typically would not influence an attempt to publish.
Sub-national and national studies at low or moderate risk were included in the subsequent results.
We estimated weighted seroprevalence in a series of separate meta-analyses each month and found in July 2021, global seroprevalence from infection or vaccination (combined seroprevalence) was 45.2%, 95% CI [40.7-49.8%] - an 8.2 fold increase since the June 2020 estimate of 5.5% [3.0-9.6%]. In July 2021, global seroprevalence attributable to infection was 35.2% [33.3-37.1%]. (Supplementary file S.4.3, Table S10)
Regional analyses began in January 2020 and ended in February 2021-August 2021, depending on when seroprevalence studies in each region sampled participants. Combined seroprevalence in February 2021 was 2.5% [1.3-4.7%] in WPR (3.7x since June 2020), 43% [38-48%] in EMR (2.8x since June 2020). In April 2021, combined seroprevalence was 22% [16-29%] in AMR LMIC (9.3x since June 2020). In June 2021, combined seroprevalence was 43% [36-51%] in EUR LMIC, a 5.2x increase since July 2020. In July 2021, combined seroprevalence was 70% [63-76%] in AFR (18x since June 2020). Finally, in August 2021, combined seroprevalence was 75% [63-84%] in SEAR (8.1x since June 2020), 72% [55-84%] in EUR HIC (16x since June 2020), and 95% [94-96%] in AMR (HIC) (35x since June 2020). (Figure 2, middle panel and Supplementary file S.4.3, Table S9). In the meta-analyses with at least 2 studies, 93% (283/304) showed considerable heterogeneity from 75% to 100%.(26)
Snapshots of seroprevalence to confirmed case ratios, based on estimated weighted seroprevalence using national studies, are shown in Table 2. Globally, the median ratio was 30:1 in 2020 Q3 and 17:1 in 2021 Q2. In 2020 Q3, the median ratio ranged from 3.5:1 in AMR (HIC) to 83:1 in AFR. In 2021 Q2, this ranged from 1.2:1 in AMR (HIC) to 183:1 in AFR (Table 2).
NA = national studies not available. Seroprevalence studies that sampled participants in 2021 were adjusted for antibody target and vaccination rate to calculate seroprevalence attributable to infection (full details: Methods and Supplement S3.2). *There are no high income countries in the WHO South-East Asia region; the two high-income countries in the WHO Africa region, Mauritius and Seychelles, both have no seroprevalence studies and were hence not included in this analysis.
Asymptomatic seroprevalence in the by age and sex subgroups for studies reporting subgroups on symptoms are shown in Supplementary Figure S3. Median asymptomatic prevalence was similar across age groups (Kruskal-Wallis (KW) H-test p = 0.13). Median asymptomatic prevalence in males was 61.1% compared to 55.7% in females (KW H-test p = 0.40).
Within studies, compared to the reference category of 20-29 years old, seroprevalence was significantly lower for children 0-9 years (prevalence ratio [PR] 0.77, 95% CI [0.71-0.84]) and adults 60+ years (0.88 [0.80-0.98]). There were no differences between other age groups nor between males and females. (Full results: Figure 3)
Top panel: Box plots show median and IQR of reported seroprevalence. Middle panel: We produced weighted point estimates of seroprevalence by meta-analyzing studies in 12-week rolling windows. To capture the trend i seroprevalence in each WHO region and globally, we fit a flexible, smooth function of time using to the point estimates using non-parametric regression (full details: Supplement S.4.2). Bottom panel, left axis: Shaded areas represent the relative frequency of major variants of concern (VOC) circulating, based on weekly counts of hCoV-19 genomes submitted to GISAID we have aggregated by month. Weeks with fewer than 10 total submissions in a given country were excluded from the analysis. Bottom panel, right axis: New confirmed cases per 100,00 people, smoothed using local regression (LOESS).
Left panel: Meta-analysis results. We calculated the ratio in prevalence between subgroups within each study then aggregated the ratios across studies using inverse variance-weighted random-effects meta-analysis. Heterogeneity was quantified using the I2 statistic. Each row represents a separate meta-analysis. Right panel: Regression results. We fit a log-Poisson generalized linear mixed-effects model, excluding studies where over 5% of the geographic population had received at least one dose of any vaccine. We performed model comparison and selected the model with the lowest AIC criterion (Appendix 6). Sampling frame was dropped from the final model. Public health and social measures (PHSM) data was taken from the London School of Hygiene and Tropical Medicine global dataset. The PHSM index scale ranged from 0 (least stringent) to 10 (most stringent) (see Supplement S4). k = 305; □2(se) = 0.79 (0.89). The marginal R2, or variation between studies explained only by fixed effects, was 60.9%. Multivariable analysis included additional controls for WHO region, transmission phase, and age group not shown in figure.
In the multivariate analysis (pre-vaccination), the final model included all a priori independent predictors except sampling frame (model comparison and diagnostics: S.4.3, Table S11). Sub-national studies reported higher seroprevalence estimates compared to national studies (PR 1.36, 95% CI [1.10-1.68]). Compared to HIC, higher seroprevalence estimates were reported by low income (PR 7.44 [3.67-15.08], lower-middle income (PR 8.47 [5.17-13.87]), and upper middle-income countries (PR 4.46 [3.20-6.22]). Higher cumulative incidence of reported cases was associated with higher seroprevalence (PR 1.28 [1.16-1.42]), while more stringent PHSM measures up to the sampling midpoint date, continuous from 0 to 10, were associated with lower seroprevalence (PR 0.88 [0.78-0.99]).
Discussion
We synthesized data from over 800 seroprevalence studies worldwide (43% from LMICs) published up to December 2021, providing global and regional estimates of SARS-CoV-2 seroprevalence over time with substantial representation of regions with limited available seroprevalence data. We estimate that approximately 45.2% of the global population has antibodies against SARS-CoV-2 in July 2021 (35.2% when excluding vaccination). Global seroprevalence has risen considerably over time, from 5.5% a year before, in June 2020.
Our findings provide evidence of regional and temporal variation in the estimated seroprevalence, up to 70-75% in SEAR and AFR in July-August 2021 and 95% in AMR HIC in August 2021. In WPR, there was a paucity of high-quality population-based studies in 2021 and estimated seroprevalence was as low as 2.5% in February 2021, though it has likely increased since. Regional variation is driven by differences in the extent of SARS-CoV-2 infection and vaccination. This is exemplified by our monthly timeline of seroprevalence by region, 2020-2021, which provides estimates of evolving temporal changes of the global pandemic. We observe increases in seroprevalence following the emergence of variants in regions with available data (e.g., 6% (July 2020) to 41% (April 2021) in AFR following the Beta variant and 12% (February 2021) to 75% (August 2021) in SEAR following the Delta variant), demonstrating the substantial number of infections caused by more transmissible variants. In HIC regions, the increases in seroprevalence are also driven by increased vaccine coverage in 2021 (e.g., 6% (January 2021) to 95% (August 2021) in AMR HIC and 7% (January 2021) to 72% (August 2021) in EUR HIC). Our results add global representation and principled estimation of changes in seroprevalence over time as compared to previous evidence syntheses (8–10). These estimates are similar to estimates of true infections by global epidemiological models. For example, our global estimate of seroprevalence attributable to infection (3.2%) is similar to the Institute of Health Metrics and Evaluation cumulative infection incidence estimate of 37.4% on 31 July 2021.(34) Our analysis provides an orthogonal estimate based solely on seroprevalence data, using a method that has the added value of being easily interpretable and with fewer assumed parameters.
Our results provide evidence of considerable case under-ascertainment, indicating that many cases of SARS-CoV-2, including subclinical cases, are not captured by surveillance systems which in many countries are based on testing of symptomatic patients or simply access to testing. There was wide variation in under-ascertainment (as estimated through seroprevalence-case ratios) in all regions, income groups and over time, with higher ratios consistently observed in LMICs compared to HICs. Our ratios of seroprevalence to reported cases in late 2020 were comparable to other studies for AMR, EUR, and SEAR.(9–12) Our estimates of seroprevalence to cumulative incidence ratios for AFR, WPR, and EMR are novel, with no other analyses we found having systematically estimated ascertainment through seroprevalence in these regions; moreover, estimates of true infections from epidemiological models suggest the high levels of under-ascertainment suggested by this study are plausible.(34)
We also provide more granular evidence of significant variation in infection by age by 10-year band. Children aged <10 years, but not children aged 10-19, were less likely to be seropositive compared to adults aged 20-29 years; similarly, adults aged >60 years, but not those aged 30-39, 40-49, or 50-59, were less likely to be seropositive than adults 20-29. These findings add nuance and granularity to differences in seroprevalence by age observed by other studies.(10) Lower seroprevalence in adults 60+ could be explained by immunosenescence that can lead to quicker seroreversion,(35) higher mortality and hence a lower proportion of individuals with evidence of past infection, gaps in vaccine access, or more cautious behaviour resulting in fewer infections in this age group. There are several possible explanations for lower seroprevalence in children: milder infections, which are generally associated with lower antibody titers;(36) school closures; and ineligibility for vaccination.
Our multivariate model suggests higher seroprevalence estimates were reported by low and lower-middle income countries compared to high-income countries, with the highest seroprevalence in lower-middle income countries (pre-vaccination). Potential explanations for this result are multifaceted and include weaker health system functionality and performance, lower capacity to isolate, and less stringent use of and ability to effectively implement PHSM. This is also consistent with findings by Rostami et al.(11) Our results also suggest that an increase in overall PHSM stringency was associated with lower seroprevalence. This and other work has shown that the use of PHSM was associated with reduced SARS-CoV-2 infections, especially when implemented early and limiting population mobility.(37–39)
Our regional and global meta-analysis estimates are timely, robust and geographically diverse with estimates from all WHO regions. The laboratory and epidemiological standardization enabled by the SEROPREV protocol, as well as the inclusion of only studies assessed to have low or moderate risk of bias using a validated risk of bias tool (20), enabled analysis of high-quality and comparable data. In line with the equity principles of the UNITY initiative, our dataset had global coverage, including a broad range of LMICs (one third of studies included in our meta-analysis of seroprevalence over time, n=127) and vulnerable HRP countries (15% of included studies). UNITY study collaborators shared timely evidence by uploading their aggregated and standardized early results to an open data repository, enabling geographic coverage and reducing publication bias.
A few limitations should be described. First, although we conducted meta-regression to explore heterogeneity of the included studies, there remained some residual heterogeneity that could not be explained quantitatively — likely driven by differences in disease transmission in the different countries and time points that serosurveys were conducted. Second, we did not account for waning of population immunity, so the present work likely underestimates the extent of past infection and case ascertainment. Thirdly, seroprevalence studies are cumulative, meaning that results reflect all COVID-19 countermeasures implemented up to the time of participant sampling and, thus, we cannot isolate the contributions of particular PHSM. Fourthly, while we screened study eligibility based on high assay performance criteria, different serological assays may yield varying results which should be taken into account when interpreting seroprevalence data. Finally, at certain points in time, our meta-analysis estimates were driven by studies from specific countries — either very populous countries (i.e. SEAR: India, AMR HIC: USA, AMR LMIC: Brazil, WPR: China), or countries in regions with scarce data during the time in question (e.g. EMR: 2 countries in early 2021). We also could not produce global estimates for late 2021 due to the delays between when studies conducted their sampling (we extracted from the ‘sampling midpoint’), and when these results were later published or released within our search dates.
Population-based seroprevalence studies primarily give a reliable estimate of the exposure to infection. In cases where antibodies can be measured quantitatively, such as for SARS-CoV-2, they can also provide correlates with protection against infection(4). While antibodies persist in most infected individuals for up to year (with early evidence pointing at up to 18 months),(40–43) the reinfection risk with the d immune-escaping Omicron variant, is reported to be much higher than in previous VOCs in both vaccinated and previously infected individuals, indicating that the presence of antibodies is less indicative of a level of protection against infection. However, seroprevalence estimates remain indicative of protection against severe disease, as cellular immunity is unlikely to be disrupted even with an immune escaping VOCs.
Seroprevalence studies have been invaluable throughout the COVID-19 pandemic to understand the true extent and dynamics over time of SARS-CoV-2 infection and, to some extent, immunity. Seroprevalence has increased considerably in past months, due to infection in some regions and vaccination in others. As our understanding of SARS-CoV-2 develops, the role of seroprevalence studies may change including the adaptation of study objectives and methodology to the situation. Currently, our global estimates of infections based on seroprevalence far exceed reported cases captured by surveillance systems. As we enter the third year of the COVID-19 pandemic, continued implementation of a global system for continuous, multi-pathogen, and standardized serosurveillance (44,45) is a crucial next step to monitor the COVID-19 pandemic and contribute to readiness for other emerging respiratory pathogens. The pandemic persists in large because of inequitable access to countermeasures tools such as vaccines; emphasizing the importance of equitable vaccine deployment globally, the strengthening of health systems and of tailored PHSM to mitigate disease transmission until high population protection is achieved. Globally standardized and quality seroprevalence data continue to be essential to inform health policy decision-making around COVID-19 control measures, particularly in capacity-limited regions with low testing capacity and vaccination rates.
Data Availability
Standardized results uploaded to Zenodo by UNITY study collaborators supporting the findings of this study are available at this community and DOIs for each included Zenodo study are cited in the Supplementary File (S5) https://zenodo.org/communities/unity-sero-2021?page=1&size=20 (13). Detailed information on each study is available in Supplement S4. Other relevant data are available in a data repository (doi:10.5281/zenodo.5773152) and/or available from the Zenodo community upon reasonable request.
Supplementary material
Supplementary material file is attached.
Contributors
Conceptualization: IB, LS, AN, MVK, RKA, NB.
Data curation: HCL, AnV, LA, JO, TA, LL, AiV, MW, XM, ZL, NB, HW, CC, MYL, ML, MS, GRD, NI, CZ, SP, HPR, TY, KCN, DK, SAA, ND, CD, NAD, EL, RKI, ASB, ELB, AS, JC.
Formal analysis: HW, RKA, DB, JP, MPC.
Funding acquisition: IB, MVK, RKA, NB, TY.
Investigation: IB, LS, HCL, AN, MV, BC, AnV, LA, AR, JO,TA, PW, LL, AiV, PM, KV, EB, SM, OMK, LS, CZ, EC, KKS, OAN, SEQ, JDN, SO, DB, PCM, MS, AFM, TMW, SN, BC, ED, TSD, SAA, LF, AC, TK, RS, AZM, SAS, MO, DAM, RKI, TS, VT, ML, AR, ITT, ASB, NT, SI, JCK, ASB, EAI, RAA, SB, TGH, PJL, HM, AC.
Methodology: HW, RKA, DB, NB, ML, HQ, CPY, TW, ND, CD, TE.
Project administration: IB, MVK, MW, NB, RKA, TY, PM, HQ, OMK, EC, OAN, IYR, EAA, SO, AB, MS, PC, SAA, ZN, AR, GSF, RKI, TS, FI, AR, ECB, ASB, BLH. Resources: LS, BC, HCL, AnV, LA, AR, JO, TA, LL, AiV.
Supervision: IB, AnV, LA, AR, JO, TA, PRW, LL, RP, MVK, DB, MPC, JP, RKA, NB, TY.
Writing – original draft: IB, LS, HCL, AN, MW, HW, RKA.
Writing – review & editing: all authors.
All authors debated, discussed, edited, and approved the final manuscript. All authors had full access to the full data in the study and accepted responsibility to submit for publication.
Declaration of interests
Competing interests
WHO had a role in study design, data collection, data analysis, data interpretation, and writing of the report. No other funders had any such role. The corresponding authors had full access to all the data in the study and had final responsibility for the decision to submit for publication. Authors were not precluded from accessing data in the study, and they accept responsibility to submit for publication.
Potential other competing interests of named co-authors include: RKA, MW, HW, ZL, XM, CC, MYL, DB, JP, MPC, ML, MS, GRD, NI, CZ, SP, HPR, TY, KCN, DK, SAA, ND, CD, NAD, EL, RKI, ASB, ELB, AS, JC and NB report grants from Canada s COVID-19 Immunity Task Force through the Public Health Agency of Canada, and the Canadian Medical Association Joule Innovation Fund. RKA, MW, HW, ZL, CC, MYL, NB also report grants from the World Health Organization and the Robert Koch Institute. RKA reports personal fees from the Public Health Agency of Canada and the Bill and Melinda Gates Foundation Strategic Investment Fund, as well as equity in Alethea Medical (Outside the submitted work). MPC reports grants from McGill Interdisciplinary Initiative in Infection and Immunity and Canadian Institute of Health Research, and personal fees from GEn1E Lifesciences (Outside the submitted work), nplex biosciences (Outside the submitted work), Kanvas biosciences (Outside the submitted work). JP reports grants from MedImmune (Outside the submitted work) and Sanofi-Pasteur (Outside the submitted work), grants and personal fees from Merck (Outside the submitted work) and AbbVie (Outside the submitted work), and personal fees from AstraZeneca (Outside the submitted work). DB reports grants from the World Health Organization, Canadian Institutes of Health Research, Natural Sciences and Engineering Council of Canada (Outside the submitted work), Institute national d excellence en sante et service sociaux (Outside the submitted work), and personal fees from McGill University Health Centre (Outside the submitted work) and Public Health Agency of Canada (Outside the submitted work).
Data availability
Standardized results uploaded to Zenodo by UNITY study collaborators supporting the findings of this study are available at this community and DOIs for each included Zenodo study are cited in the Supplementary File (S5) https://zenodo.org/communities/unity-sero-2021?page=1&size=20 (13).
Detailed information on each study is available in Supplement S4. Other relevant data are available in a data repository (doi:10.5281/zenodo.5773152) and/or available from the Zenodo community upon reasonable request.
Unity Study Collaborator Group
**UNITY Study Collaborator Group: Ximena Aguilera19, Sheikh Al-Shoteri20, Eman A Aly8, Mauricio Apablaza21, Rosemary A Audu22, Amal Barakat8,23, Abdulla S Bin-Ghouth24, Enyew Birru25, Dejan Bokonjic26, Shelly Bolotin27,28, Henry K Bosa29,30,31, Emily L Boucher32, Elma Catovic-Baralija33, Alexei Ceban34, Annie Chauma-Mwale35, Judy Chen36, Battogtokh Chimeddorj37, Pui Shan Chung10, Cheryl Cohen38, Tienhan S Dabakuyo-Yonli39, Gabriel R Deveaux32, Boly Diop40, Titus H Divala41, Emily K Dokubo42, Irene O Donkor43, Claire Donnici32, Nathan Duarte44, Natalie A Duarte45, Timothy G Evans7, 14, Lee Fairlie46, Ousmane Faye47, Gudrun S Freidl11, Claudia González19, Tiffany G Harris48,49, Belinda L Herring3, Sopon Iamsirithaworn50, Gloria Icaza51, Rhoda Ila52, Natasha Ilincic45, Elsie A Ilori53, Francis Y Inbanathan9, Vicki Indenbaum54, John Kaldor55, Dayoung Kim32, Olatunji M Kolawole56, Jambo C Kondwani57,58, Tatiana Kuchuk59, Pritesh J Lalwani60, Moses Laman61, Evelyn Lavu61‡, Juliana Leite12, Michael Liu62, Emma Loeschnik63, Kristine Macartney64, Dorothy A Machalek55,65, Sheila Makiala-Mandanda66,67, Henri-Pierre Mallet68, Alexandre Manirakiza69, Pilly Mapira52, Pinyi N Mawien70, Puneet Misra71, Sanjin Musa72,73, Portia C Mutevedzi74,75, Osama A Najjar76, Sutthichai Nakphook50, Kim C Noel16, Zuridin Nurmatov77, Maria Ome-Kaius61, Eric M Osoro78, Krishna P Paudel79, Sara Perlman-Arrow7, Sharif E Qaddomi8,80, Hude Quan81, Alissar Rady82, Hannah P Rahim83, Muriel Ramírez-Santana84, Izzat Y Rayyan8,80, Angel Rodriguez12, Karampreet Sachathep85, Mitchell Segal13, Anabel Selemon32, Tahmina Shirin86, Kristen A Stafford87,88, Laura Steinhardt42, Vanessa Tran27,89, Isidore T Traore90,91, Pablo Vial92, Tri Yunis M Wahyono93, Tyler Williamson2,81, Cedric P Yansouni94,95, Caseng Zhang96, Chong Zhuo Lin97
19Centro de Epidemiología y Políticas de Salud, Facultad de Medicina Clínica Alemana, Universidad del Desarrollo, Chile
20Aden University, Yemen
21Facultad de Gobierno, Universidad del Desarrollo, Chile
22Nigerian Institute of Medical Research, Nigeria
23Infectious Hazard Preparedness, WHO Health Emergencies Programme
24Hadhramout University, Al Mukalla, Yemen
25Ethiopian Public Health Institute, Addis Ababa, Ethiopia
26University of East Sarajevo Faculty of Medicine Foča, Bosnia and Herzegovina (Republic of Srpska)
27Public Health Ontario, Toronto, Ontario, Canada
28Dalla Lana School of Public Health, University of Toronto, Ontario, Canada
29Ministry of Health, Uganda
30Uganda Virus Research Institute, Entebbe, Uganda
31Uganda Peoples Defense Forces, Kampala, Uganda
32Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
33Department for Blood Transmissible Disease Testing, Institute of Transfusion Medicine of the Federation of Bosnia and Herzegovina, Sarajevo, Bosnia and Herzegovina (Federation)
34World Health Organization, Country Office in the Republic of Moldova
35Public Health Institute of Malawi, Ministry of Health, Malawi
36Faculty of Medicine and Health Sciences, McGill University, Montreal, Quebec, Canada
37Department of Microbiology and Infection Prevention Control, School of Biomedicine, Mongolian National University of Medical Sciences, Ulaanbaatar, Mongolia
38Center for Respiratory Disease and Meningitis, National Institute for Communicable Diseases, Johanessburg, South Africa
39Epidemiology and Quality of Life Research Unit, INSERM U1231, Georges François Leclerc Centre – UNICANCER, Dijon, France 40Surveillance Division, Prevention Directorate, Ministry of Health and Social Action, Dakar, Senegal
41Kamuzu University of Health Sciences, University in Blantyre, Malawi 42U.S. Centers for Disease Control and Prevention, Atlanta, United States
43Noguchi Memorial Institute for Medical Research, Accra, Ghana
44Faculty of Engineering, McGill University, Quebec, Canada
45University of Toronto, Toronto, Ontario, Canada
46WITS Reproductive Health and HIV Institute, Faculty of Health Sciences, University of the Witwatersrand, South Africa
47Virology Department, Institut Pasteur de Dakar, Dakar, Senegal
48The International Center for AIDS Care and Treatment Programs (ICAP), Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, United States
49Department of Epidemiology, Mailman School of Public Health, Columbia University
50Department of Disease Control, Ministry of Public Health, Thailand
51Instituto de Matemáticas, Universidad de Talca, Chile
52School of Medicine and Health Sciences, University of Papua New Guinea
53Nigeria Centre for Disease Control
54Central Virology Laboratory and Sheba Medical Center, Ministry of Health, Tel-Hashomer, Israel
55The Kirby Institute, University of New South Wales, Kensington, Australia
56Department of Microbiology, Faculty of Life Sciences, University of Ilorin, Ilorin, Nigeria
57Malawi-Liverpool-Wellcome Clinical Research Programme
58Liverpool School of Tropical Medicine
59Research and Production Center “Preventive Medicine” of the Ministry of Health of the Kyrgyz Republic, Kyrgyzstan
60Instituto Leônidas e Maria Deane (ILMD), Fiocruz Amazônia, Manaus, Amazonas, Brazil
61Papua New Guinea Institute of Medical Research, Goroka, Papua New Guinea
62Harvard Medical School, Boston, Massachusetts
63Department of Epidemiology and Biostatistics, Schulich School of Medicine and Dentistry, Western University, Ontario, Canada
64National Centre for Immunisation Research and Surveillance, Sydney, Australia
65Centre for Women’s Infectious Diseases, The Royal Women’s Hospital, Victoria, Australia
66Institut National de Recherche Biomédicale, Kinshasa, Dominican Republic of the Congo
67Université de Kinshasa, Democratic Republic of Congo
68Agence de Régulation de l’Action Sanitaire et Sociale de Polynésie française
69Institut Pasteur of Bangui, Central African Republic
70Preventive Health Services, Ministry of Health, Juba, South Sudan
71All India Institute of Medical Sciences, New Delhi, India
72Department of Epidemiology, Institute for Public Health of the Federation of Bosnia and Herzegovina, Bosnia and Herzegovina
73Sarajevo Medical School, University Sarajevo School of Science and Technology, Bosnia and Herzegovina
74South African Medical Research Council Vaccines and Infectious Diseases Analytics Research Unit, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
75School of Pathology, Faculty of Health Sciences, University of Witwatersrand, South Africa
76Palestinian National Ministry of Health, Occupied Palestine Territory
77Scientific and Production Association for Preventive Medicine, Ministry of Health of the Kyrgyz Republic, Kyrgyzstan
78Washington State University, Global Health Kenya, Nairobi, Kenya
79Epidemiology and Disease Control Division, Ministry of Health and Population, Nepal
80The Palestinian National Institute for Public Health, Ramallah, Occupied Palestine Territory
81Department of Community Health Sciences, University of Calgary, Calgary, Alberta, Canada
82World Health Organization, Country Office Beirut, Lebanon
83Boston Consulting Group
84Public Health Department, Facultad de Medicina, Universidad Católica del Norte, Coquimbo, Chile
85The International Center for AIDS Care and Treatment Programs (ICAP), Department of Population and Family Health at Mailman School of Public Health, Columbia University, New York, United States
86Institute of Epidemiology, Disease Control and Research (IEDCR), Bangladesh
87Center for International Health, Education, and Biosecurity, University of Maryland School of Medicine, Baltimore, United States
88Division of Epidemiology and Prevention, Institute of Human Virology, University of Maryland School of Medicine
89Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Canada
90Programme de Recherche sur les maladies infectieuses, Centre MURAZ, Bobo-Dioulasso, Burkina Faso
91Institut Supérieur des Sciences de la Santé, Université Nazi Boni, Bobo-Dioulasso, Burkina Faso
92Instituto de Ciencias e Innovación en Medicina, Facultad de Medicina Clínica Alemana, Universidad del Desarrollo, Chile
93Department of Epidemiology, Faculty of Public Health, University of Indonesia, Depok, Indonesia
94J.D. MacLean Centre for Tropical Diseases, McGill University Health Centre, Quebec, Canada
95Divisions of Infectious Diseases and Medical Microbiology, McGill University Health Centre, Quebec, Canada
96Faculty of Health Sciences, McMaster University, Ontario, Canada
97Institute for Public Health, National Institutes of Health, Ministry of Health, Malaysia
‡Deceased
Acknowledgements
This work was supported by WHO through funding from the WHO Solidarity Response Fund and the German Federal Ministry of Health COVID-19 Research and Development. IB, LS, AnV, LA, AR, JO, TA, PW, LL, AiV, RP, MVK are employed by WHO, and HCL, AN, MV, BC are WHO consultants; the authors alone are responsible for the views expressed in this publication and they do not necessarily represent the decisions, policy, or views of WHO.
SeroTracker (led by RKA, including MW, HW, ZL, XM, TY, CC, MYL, JP, MPC, DB, ML, MS, GRD, NI, CZ, SP, HPR, TY, KCN, DK, SAA, ND, CD, NAD, EL, RKI, ASB, ELB, AS, JC). is grateful for support from WHO, Canada s COVID-19 Immunity Task Force through the Public Health Agency of Canada, the Robert Koch Institute, and the Canadian Medical Association Joule Innovation Fund. RKA additionally thanks the Rhodes Trust for its support.
We thank colleagues at partner organizations including WHO/Dubai Logistics Team; Myrna Charles, Kathleen Gallagher, Amen Ben Hamida, Christopher Murrill, Toni Whistler, Venkatachalam Udhayakumar (US Centers for Disease Control and Prevention); Eeva Broberg, Erika Duffell, Maria Keramarou and Pasi Penttinen (European Centre for Disease Prevention and Control), Vincent Richard (Institut Pasteur and the Institut Pasteur International Network); WHO regional offices (Jacob Barnor, Alina Guseinova, Jason M Mwenda, Dmitriy Pereyaslov, Harimahefa Razafimandimby, and WHO HQ (Michael Ryan).
We would especially like to thank all WHO UNITY Study Collaborator Group, who are also named co-authors of this paper and listed in the section below. These and other collaborators are recognized on a dedicated webpage on the UNITY and WHO website,(46) in all the countries who embarked in this global response effort to COVID-19, as well as all individuals who supported, conducted or participated in each of the studies supported.
Footnotes
** Full group author list is listed in the acknowledgements section
Extended search, initially was from Jan 202 to Oct 2021, and now covering the full first two years of the pandemic: Jan 2020-Dec 2021, pre-omicron variant
References
