Methodology

Five renowned European and international experts on testing policies and strategies joined an online panel facilitated by Policy Wisdom, LLC. Their academic backgrounds included microbiology, infectious diseases, virology, epidemiology, and health policy. Throughout the three months from August to October 2020, a multistage validation and feedback process was carried out. 
As the first step, we created a working document from a desk review of literature, policies, and epidemiological data. The search was guided by five pre-established topics:
Experts convened in a two-session meeting to inform and validate the contents of the working document and to discuss and propose a set of recommendations. The paper was then updated using this and subsequent written feedback. After reaching a consensus, the final document was approved by all panellists.

Results

Impact on population health

Morbidity and mortality rates have been especially elevated among the high-risk and vulnerable populations, while the pandemic has also worsened conditions for those who regularly need to access health facilities. Among the highest risk populations for the disease are (1) the elderly, particularly in long term care facilities (LTCF), (2) those with comorbidities, including communicable and non-communicable diseases, (3) vulnerable communities, which include those in outbreak-prone settings, and (4) healthcare workers. Europe has documented higher rates of infections among individuals over 70 years old and higher mortality rates in those above 65 [4,5]. Regional reports show figures of up to 88% mortality among those 65 years of age and above [16]. Furthermore, several countries including Italy, France, Germany, Sweden, Belgium, and the U.K., reported outbreaks in nursing homes, which accounted for a considerable fraction of their recorded fatalities [6].
Patients with comorbidities or underlying health-compromising conditions, such as diabetes, cancer, HIV, cardiovascular, chronic kidney, and chronic respiratory disease as well as other risk factors, such as obesity or smoking have also been shown to be disproportionately affected. Europe is the world region with the highest percentage of population that could be at risk due to underlying health conditions [7]. Additionally, those with chronic illnesses and other health-compromising conditions have had reduced access to healthcare services and medicines due to the burden on the health system imposed by the pandemic. Simultaneously, many patients may hesitate to visit health centres either for fear of infection, unclear communication on whether they can, and lack of understanding about which type of facility is safest to visit.
Moreover, socio-economic disparities may carry additional risks to certain sections of the population, particularly those of lower-income and/or marginalized communities [8]. Many low-income individuals live on daily wages and simply cannot forego employment and/or their economic livelihood depends on what are deemed as essential jobs. This situation exposes them to potential infection leading to an increased risk of outbreaks. Likewise, vulnerable or marginalized communities face extreme disparities that lead to increased difficulties in accessing healthcare (including laboratory testing) whether it be for economic reasons, literacy, transportation, language, cultural, or other barriers [9]. Among the most vulnerable populations are migrants (including seasonal workers that migrate between east and west Europe, and North Africa), correctional institution inmates, and the homeless [10,11].
 Secondary effects of the pandemic may exacerbate the clinical situation of those who are already suffering mental health conditions, such as depression, anxiety, and other similar disorders, as well as a surge in new mental health cases [17,18]. Stay-at-home measures, loss of employment as well as the stress associated with a pandemic situation and risk of infection have contributed to this effect [16]. Multiple sources identify healthcare workers among the most affected, including risk for suicide [11,17,19]. Lastly, preliminary reports have also suggested an increase in intimate partner violence in Europe during the outbreak. [20].

Broader Socio-Economic Impact

Although essential, imposing stringent non-pharmaceutical interventions has severe negative effects on country economies. According to the OECD, it is estimated that for every month that a country is in total lockdown, it loses up to approximately 2% of its GDP [13]. With most of the workforce under lockdown, unemployment rates have soared causing a domino effect in which almost all sectors of the economy have suffered [21,22]. International trade has slowed significantly and areas such as travel and tourism, as well as sports, food and entertainment have been especially affected. According to the World Travel & Tourism Council (WTTC), tourism generates 10,3% of global GDP, making it an important driver of the global economy. For countries such as Spain and Italy the contribution of tourism to their GDP is even higher than the global average, 14,3% and 13,0% of their GDP in 2019, respectively. It is predicted that 13 million jobs in the travel and tourism sector will be lost in Europe due to the SARS-CoV-2 pandemic, resulting in a tourism GDP loss of €633 billion [23].
The closure of schools and universities has also had great impact, limiting access to quality education, and sometimes resulting in students being unable to continue their studies. The disruption of classes at schools and universities has also placed an extra burden on society, causing concern for the short and long-term impact on the development of human capital and workforce [14,15]. Women and girls have been particularly affected by physical distancing measures, with a heavy loss in employment and lack of access to education. As the UN reported, decades of progress in women’s rights have been scaled back in a matter of months [24,25].

SARS-CoV-2 Serology Testing

Several types of serology tests exist and each one has a different use depending on the setting and application. Currently available serology tests, or immunoassays, to detect SARS-CoV-2 antibody responses are designed for the detection of antigens or antibodies. Tests to detect the two main isotypes of antibodies, immunoglobulin M (IgM) and immunoglobulin G (IgG), are those most used to determine subject immunity. IgM represents an early antibody response during the acute phase of infection and may indicate current or recent infection. On the other hand, IgG, which are usually produced at a later stage, may indicate that a patient has been exposed, but was/may also still be infected, with or without symptoms. Immunoglobulin A (IgA) antibodies can also be detected in mucosal secretions or in serum samples but are often associated with earlier production of antibodies, shortly before IgM, and not associated with longer term immunity. SARS-CoV-2 antibody production, however, seems to differ from the typical scenario in that IgM and IgG tend to appear at the same time, while other models depict IgA and IgM developing simultaneously, with IgA even outlasting IgM antibodies in some instances [26].
Testing methods and setting needed to take and process samples, may also influence access and turnaround time of results. Antibody tests can be carried out either by laboratory-based assays or with rapid diagnostic tests (RDT). While the processing of samples for laboratory-based assays are more centralized, laboratories can analyse a higher number of samples. RDTs usually take 15-30 minutes to complete and are easily conducted and processed in decentralized settings. In general, laboratory-based assays generate more accurate results as they can provide qualitative (whether antibodies are present) and quantitative (amount of antibodies present) data as opposed to RDTs that only provide qualitative results [9]. The choice will depend on the application setting and on the scope of testing. Convenience and easier access to patients at point-of-care (POC) could be a preferable option of RDTs, yet it may require additional confirmation from more reliable laboratory-based assays. 
Certain limitations of antibody tests have raised concerns, including their reliability or accuracy for detecting SARS-CoV-2 infection due to possible cross-reactivity [27]. Antibody test accuracy is evaluated by its specificity and sensitivity. While specificity measures a test’s ability to correctly generate a negative result (true negative rate), sensitivity measures how often a test correctly generates a positive result (true positive rate). In clinical settings no assays reach 100% accuracy in both specificity and sensitivity. Based on the prevalence of infection in a determined population, the specificity and sensitivity of tests are used to determine the positive predictive value (PPV) and negative predictive value (NPV). Predictive values permit assessment of the effectiveness of a specific testing strategy to determine the accuracy and reliability of serology tests at the individual and at population levels [9]. Choosing a test with higher specificity may be preferable in populations with low prevalence of the disease. Nevertheless, if not possible, the use of tests with lower specificity could be accompanied by orthogonal testing algorithms [28].
Additional concerns arise from the fact that some individuals may produce very low levels of antibodies, which may be missed by serological tests [29]. Furthermore, there is some indication that a classic long-term immunity may not exist for SARS-CoV-2 infection. In fact, some studies have documented reinfections [9,30], and others suggest neutralizing antibodies may wane after two to three months, especially for those who experience mild or asymptomatic cases [29,31]. These ambiguities have implications for determining herd immunity and understanding the proportion of the population that must be immune at the community level to cease high rates of transmission.

Benefits of SARS-CoV-2 Serology Testing 

Serology testing can help enrich epidemiological data through sero-surveillance to inform policymaking at different levels, especially when asymptomatic cases seem to be high. While some preliminary studies estimated the proportion of asymptomatic cases at 17.9%, others go as high as 40% [32,33]. Although serology tests cannot replace RNA tests in hospital settings, they may provide additional information to support decision making on patient clinical diagnosis [34]. Hence, conducting serology tests in the clinical setting, when the index of suspicion is high, but SARS-CoV-2 cannot be detected, represents an opportunity to determine if a patient had been infected, providing relevant data that can help complete datasets and enhance preparedness. Finally, implementation of systematic serology testing and sero-surveillance could also provide an alternative method to less accessible and affordable lab RNA testing [35]. To summarize, evidence suggests that serology testing can be used as one of the tools to control virus transmission, prevent community spread, and meet public health and community needs and demands. 
Serology testing might also be beneficial for research purposes, especially under scenarios where the virus is still new and less prone to mutations and genetic diversification. The use of serological tests to assist the development of effective treatments and/or vaccines is prescribed in this context [36], as so is research to identify the role of neutralizing antibodies [37,38].

Challenges of SARS-CoV-2 Serology Testing

Although WHO has acknowledged the importance of serology testing for epidemiological surveillance and research purposes, it provides limited guidance and support on implementation, and lacks emphasis on the value of other possible benefits. The existing WHO interim guidelines on the use of POC immunodiagnostic tests recommends molecular testing as a gold standard [37], and in the latest interim guideline indicates serology testing as complementary in specific cases [39].
At the regional level, the European Commission has acknowledged the overall utility of serology testing [40], releasing its guidelines on SARS-CoV-2 testing in April 2020 [41]. That same month, the ECDC published its Strategies for the Surveillance of COVID-19 [42]. However, neither of these documents set a clear plan for rolling out serology testing across the region. Meanwhile, some European countries have independently implemented serology testing strategies, while others are still lagging [9,43] (see Supplementary Material 1). A list of sero-epidemiological studies conducted in the region can also be found on the ECDC’s website [44]. The fragmented initiatives across the region might result in the loss of opportunities to gather critical information in an organized manner at a crucial time. Furthermore, in absence of comprehensive guidance, access to quality testing might be hindered due to insufficient supply and availability, and, in some cases, limited health systems capacity (effective infrastructure and a trained health workforce) [45].

Policy Recommendations 

The eight policy recommendations offered in this paper focus on two main themes: the first four recommendation address the role and value of serology testing to contain and understand the SARS-CoV-2 pandemic, and the last four address issues around health system strengthening. As shown in Fig. 1, serology testing might be a valuable resource and should be considered as part of a larger comprehensive pandemic preparedness and mitigation strategy. Currently serology testing is primarily recommended for sero-surveillance and research purposes; however, this study found the need to define targeted pathways and a framework for ultimate introduction of serology testing to complement the existing strategies along with new scientific and clinical data as it becomes available. Active promotion of well-designed projects should facilitate this goal. Moreover, evidence suggests - and this paper recommends - to prioritize by geographic hotspots and to consider serology testing as part of national containment strategies to jumpstart the economy for full control of epidemics.