Seroprevalence and SARS-CoV-2 invasion in general populations: A scoping review over the first year of the pandemic

Since the beginning of the COVID-19 pandemic, counting infected people has underestimated asymptomatic cases. This literature scoping review assessed the seroprevalence progression in general populations worldwide over the first year of the pandemic. Seroprevalence studies were searched in PubMed, Web of Science and medRxiv databases up to early April 2021. Inclusion criteria were a general population of all ages or blood donors as a proxy. All articles were screened for the title and abstract by two readers, and data were extracted from selected articles. Discrepancies were resolved with a third reader. From 139 articles (including 6 reviews), the seroprevalence estimated in 41 countries ranged from 0 to 69%, with a heterogenous increase over time and continents, unevenly distributed among countries (differences up to 69%) and sometimes among regions within a country (up to 10%). The seroprevalence of asymptomatic cases ranged from 0% to 31.5%. Seropositivity risk factors included low income, low education, low smoking frequency, deprived area residency, high number of children, densely populated centres, and presence of a case in a household. This review of seroprevalence studies over the first year of the pandemic documented the progression of this virus across the world in time and space and the risk factors that influenced its spread.


Introduction
The coronavirus disease outbreak, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was first reported in Wuhan in December 2019 and spread rapidly to other parts of the world. On April 30, 2021, COVID-19 accounted for 150,000,000 confirmed cases worldwide, more than 3,000,000 deaths and about 87,000,000 recoveries, representing the deadliest pandemic in decades [1]. To contain the spread of the virus, daily counts of laboratory-confirmed cases and deaths have been published in real time. The emergence of this new virus has resulted in an important early warning plan, and the WHO has tried to understand its modes of transmission, severity, characteristics and risk factors for infection. This alert plan was targeted to manage this epidemic, which became a pandemic barely 2 months after its appearance [2]. The plan includes refining case referrals, reinforcing surveillance, and defining the main epidemiological characteristics of COVID-19 to help understand the spread, severity, spectrum of the disease and impact on the community and to provide guidance on the application of countermeasures such as case isolation and contact tracing. Daily counts of confirmed COVID-19 cases and deaths alone provide incomplete information on the relative abundance of epidemiological compartments of a susceptible infected, recovered or deceased population. Therefore, the WHO has recommended repeatedly carrying out seroprevalence surveys in multiple geographical settings [2].
Several serological surveys of SARS-CoV-2 have been performed, and others are ongoing since the beginning of the COVID-19 pandemic, finding variable seroprevalence in different countries, sometimes even among regions of the same country [3,4].
The purpose of this literature scoping review was to summarise and map the results of the seroprevalence studies according to the time since the onset of the pandemic and geographical region and to identify risk factors.

Methods
We performed a literature review of seroprevalence studies conducted in different populations since the onset of the pandemic up to April 10, 2021, searching PubMed, Web of Science and medRxiv databases. We searched references of citations for reviews, which allowed for additional references to be added.

Eligibility
We included original articles and reviews written in English language that reported data for the general population living in a defined geographical area. We considered a general population as a population-based sample, a described general population sample, or the inclusion of men and women of unselected categories of all ages in the population, preferably (but not exclusively) obtained by a random sampling technique from a large population (survey or database). We also included blood donors as a proxy for the general population. We excluded studies of health care workers, people attending a clinic or a hospital, a professional branch (industry, factory, farmers, university) and a particular population (students, nursing home) as well as modelling studies.

Search strategy
The strategy consisted of searching PubMed and medRxiv with "(Covid OR SARS-CoV-2) AND seroprevalence" and Web of Science with "covid AND seroprevalence". Articles were selected by reading titles and abstracts. All article titles were first read by two readers (CM, AB), then by a third one (FG or TL), which allowed for discussion in case of discrepancies after abstract reading to obtain consensus. Full-article reading allowed for the final selection of relevant studies and data extraction.

Data extraction
Data extraction was conducted by three authors who used a standardized form. The data collected, when available, were location; selection criteria; type of population; sample size and method; seroprevalence; serological test and type of Ig antibody tested; time period of the Table 1

Discussion
Published studies of seroprevalence of SARS-CoV-2 in the general population over the year after the onset of the pandemic up to early April 2021 showed estimates ranging from 0% in Palestine in the West Bank between June and July 2020 to 69% in England between July and September 2020, with different dynamics across continents.
This review purposedly focused on the first year of the pandemic to better understand the spread and dynamics from the early stages of the pandemic and to summarize identified factors for SARS-CoV-2 penetration that could further serve as a reference for adapted measures to mitigate future epidemics. Such measures include barriers, social distancing, evolving vaccines according to the molecular and biological monitoring of viruses, and preventive or early treatments to avoid severity. Fig 2 shows that the blood donors were not fully representative of the general population. Their seroprevalence and respective peaks show variations of lower magnitude in general. Blood donors could have been an early resource for documenting the epidemic before their decrease in frequency, with population-based studies taking place later. Some continents have fewer donors than others, probably related to national heath care organisation.
The variations observed across studies, besides the true virus exposure and spread differences over space and time within countries, could be due to different sampling techniques and the use of different serological tests. Our results suggest that the number of symptomatic cases was lower than the number of actual cases, despite only few studies (n = 21) characterizing asymptomatic seropositive cases.
Most of the studies did not find or found minimal differences regarding sex. Age categories analysis did not yet reveal any conclusive results. Social disadvantage seemed to play a role, at least for the least well-off categories, but the impact of belonging to the most privileged categories, whatever the classification used, remains to be elucidated. Finally, the presence of a COVID-19 case in a household increasing the risk of the other members has been demonstrated consistently for developing antibodies.
This review had a longer study period than six previous reviews; it focused on general populations and blood donors; it covered a larger world area by including all continents as compared with Grant et al. [8], Rostami et al. [10], Chen et al. [6] and Levesque and Maybury [9], and it summarised a number of risk factors identified, mostly of a sociodemographic nature.
A limitation of this scoping review is the heterogeneity of samples with different age ranges, so seroprevalence data are not fully comparable. Potential biases in such seroprevalence observational studies also do not account for SARS-COV-2-related deaths. Also, we did not include grey literature because we did not know how to search such literature in the particular context of this pandemic with so much suspicion of non-peer-reviewed publications. A third limitation is that we did not assess the methodological quality of the studies reviewed. Finally, the results of seroprevalence studies may have been affected by the specificity and sensitivity of different serological methods used.
In conclusion, from this scoping review of seroprevalence studies over the first year of the COVID-19 pandemic, the seroprevalence of SARS-CoV-2 varied according to the study period, with lower seroprevalence at the beginning of the epidemic than between July and September 2020. This review documented the progression of this virus across the world in time and space and the risk factors that influenced its spread.