Issues to be considered when planning sero-epidemiological studies in regions with a low incidence of SARS-CoV-2.

The pandemic of coronavirus disease has substantially affected the healthcare systems, economy and society. Research on the proportions of mild or asymptomatic cases and their roles in coronavirus transmission are ongoing worldwide. Sero-epidemiological investigation of SARS-CoV-2 allows inferences about the extent of infection and the cumulative incidence of infection in the population. However, there are some issues to be considered when planning sero-epidemiological studies in regions with a low incidence of SARS-CoV-2.

The first issue is study design. The World Health Organization (WHO) has recommended 3 design possibilities for implementation of sero-epidemiological investigation: (1) one-time cross-sectional study; (2) repeated cross-sectional study; and (3) longitudinal cohort study. Current evidence has suggested that IgM and IgG antibodies to SARS-CoV-2 can be developed 6-15 days after infection onset. Therefore, repeated cross-sectional and/or longitudinal cohort investigations are preferable for regions with a low incidence of SARS-CoV-2, as they are more likely to detect the seropositive mild or asymptomatic cases through the serial samplings approach. In contrast, one-time cross-sectional investigation is tailored for the regions and countries with initial or subsequent peaks of transmission of epidemic waves (ie, Italy, Spain, and Wuhan).3, 4, 5

The second issue is population selection. For regions with a low incidence of SARS-CoV-2, targeting specific population groups should be taken into account. There is still the question of which population to consider for detection: (1) seniors with underlying health conditions in nursing homes; (2) adults who have had close contact history with suspected or confirmed COVID-19 cases; or (3) people with suspected SARS-CoV-2 infection but with a negative reverse transcription polymerase chain reaction virus test results on their respiratory samples. The latter 2 targeted groups have been considered by an ongoing French sero-epidemiological study. Furthermore, the WHO protocol suggests that all attempts should include participants across a range of ages, in order to determine and compare age-specific sero-prevalence. Finally, instead of random sampling, convenience sampling from a selected population in the regions with a low incidence of SARS-CoV-2 has also been suggested, considering the better feasibilities for future follow-up as well as for tracking of long-term antibody dynamics.

The third issue is laboratory capacity and medical sources (ie, research funds). According to the available evidence, detection of total antibodies or IgG of SARS-CoV-2 should be prioritized, serological testing should be carried out using enzyme-linked immunosorbent assay, and the confirmation of the presence of neutralizing antibodies should ideally be performed. However, the laboratories in such regions with a low incidence of SARS-CoV-2 usually have limited capacities, that is lacking of biosafety level 3 laboratories and testing kits for SARS-CoV-2. Therefore, planners should take into account the need for well-trained specimen collectors, well-equipped facilities, and funding to cover the cost of shipping samples to an international reference laboratory for confirmation. Meanwhile, we should be aware of the false positive test results, especially in regions that have a low incidence of SARS-CoV-2. This will result in an overestimation of the proportion of infected population in the community, and could also be potentially misled about antibody status for the individuals.

The fourth issue is sample size calculation, which is critical when planning sero-epidemiological studies in regions with a low incidence of SARS-CoV-2. The WHO offers online statistical tools for sample size calculation. It is worthwhile to mention that household surveys, as a design feature, are likely to increase the required sample size of the study. Bendavid and colleagues have confirmed the positive effect of this approach. When testing 3,285 adults, they asked each adult to bring 1 child from the same household with them (889 children registered) in order to increase the study power.

Last but not least, outcome indicators should be considered. The most interesting outcome indicator is to measure the seroprevalence of antibodies of SARS-CoV-2 in the general population in affected regions. However, for regions with a low incidence, it is meaningful to estimate the proportion of asymptomatic or pre-symptomatic/subclinical infections in the population. The number of asymptomatic cases can help determine whether a region is facing the risk of a potential epidemic wave. Recently, Wuhan conducted a mass COVID-19 nucleic acid tests on about 10 million people, and only 300 asymptomatic cases were discovered. It demonstrated that asymptomatic cases may have limited transmissibility, but more studies are needed to confirm this fact.

In addition to the issues presented above, some other factors should also be considered when planning sero-epidemiological studies in regions with a low incidence of SARS-CoV-2, including socioeconomics status, geographical factors, cultures, compliance and responses, ethical considerations, and so forth.