Surveillance of COVID-19 after the pandemic. How do we do it?

Since the emergence of the omicron variant and the end of the sixth wave, a sharp decline in SARS-CoV-2^1,2 virulence, hospitalisation and mortality has begun to be observed, which may indicate that we are witnessing a turnaround in the pandemic. This has led to the development of an interesting debate on the virological and epidemiological surveillance of SARS-CoV-2 following an approach similar to that of influenza.

While wave 6 was not the best time for epidemiological monitoring and clinical management of SARS-CoV-2 similar to that of influenza, it was an opportunity to lay the groundwork for this new system for the 2022–2023 season and beyond. Much of the scientific community agrees that SARS-CoV-2 will continue to be with us, becoming just another infectious agent among the plethora of respiratory viruses that cause disease in humans.³ Some argue that it will be an endemic virus, i.e., it will circulate all year-round causing outbreaks or clusters of cases; others argue that circulation will be epidemic like influenza or the respiratory syncytial virus.

In either case, the current disease burden of SARS-CoV-2 makes it necessary to work now on a surveillance system that seeks to actively monitor this virus in a similar way as is already done with influenza. Therefore, the aim of this paper is to outline the steps necessary to carry out future surveillance of COVID-19 in a similar way to influenza, based on various documents from national and international bodies that have already made statements in this regard.

Influenza surveillance

Influenza is a notifiable disease (ND) that requires weekly reporting. Clinically diagnosed cases, or those with a laboratory diagnosis, should be reported at national level to allow for detailed monitoring of the disease and its evolution during epidemics.

Influenza surveillance has traditionally been carried out based on the clinical diagnosis of influenza-like illness (ILI), acute respiratory infection (ARI) and severe acute respiratory infection (SARI), through primary care (PC) and hospital care (HC) sentinel networks.⁴ By combining ILI and laboratory-confirmed data, the incidence rate of the disease can be obtained, either at local or national level, and represented by means of the classic weekly evolution graphs. On the other hand, both sentinel surveillance and, on some occasions, hospital surveillance, complement this monitoring by sending respiratory samples to reference laboratories, where virological analyses are carried out to detect and characterize different aspects of influenza viruses. These aspects include influenza A subtyping and the search for influenza B lineages, sequencing and other genetic characterization analyses, which are highly useful for updating seasonal influenza vaccines.

The main objectives of influenza surveillance, according to the CDC, are as follows:⁵

• •

  Determine where and when cases of influenza are occurring

• •

  Know what types/subtypes/lineages of influenza viruses are circulating

• •

  Detect changes in influenza viruses

• •

  Measure the impact of influenza on illness, hospitalization and death

All these objectives are summarised in the availability of useful information to carry out different public health actions to help mitigate the circulation of these viruses. To this end, it is important to know that the basic needs of any epidemiological surveillance system must be the integration of epidemiological information with health systems; to seek standardised systems for sharing information following a joint model; that this information can be shared electronically; and, finally, that the information obtained must facilitate the public health response to emerging health threats.⁶ Therefore, any modification made to existing protocols, or the creation of new ones must follow these characteristics.

Reasons and context for the transition of COVID-19 to an influenza-like surveillance model

The current epidemiological situation of COVID-19 in Spain suggests an evolution to an influenza-like surveillance model in the medium term. Mass testing has been the main strategy during the first 2 years of the pandemic. However, there is now a need to lay the groundwork for an orderly transition to more targeted and orderly analysis systems in order to have useful epidemiological information to assist in public health decision-making.

In the document “Strategy for surveillance and control of COVID-19 after the acute phase of the pandemic” of 23 March 2022,⁷ the Ministry of Health argues that given the high vaccination coverage against this disease in Spain, and given the number of natural infections, most of the population is protected against severe COVID-19. Therefore, it is necessary to focus efforts on the most vulnerable sectors of the population. These groups are, specifically, those over 65 years of age, immunocompromised people and those with underlying diseases.

In this same document, the Ministry of Health justifies the transition of surveillance to a new model in which this decrease in severity makes it possible not to depend on mass screening using molecular methods or rapid tests. In this way, efforts can be directed towards ensuring quality care for the patient, while at the same time continuing to detect changes in the pandemic so that the necessary modifications can be made without overburdening the health system.^7,8

Transition to a comprehensive COVID-19 surveillance model

The transition to influenza-like surveillance of COVID-19 requires attention to 2 major aspects. On the one hand, surveillance should be based on clinical diagnosis, either through sentinel networks for mild or moderate cases, or through hospitals for severe cases. Secondly, it is essential that suspected samples are submitted to reference laboratories to reinforce the available information with different virological variables, including accurate information on variants of interest and concern (VOI and VOC), as well as characteristics of the circulating virus for future adaptation of COVID-19 vaccines. It is important that this information is shared electronically through databases such as the Global Initiative on Sharing All Influenza Data (GISAID).⁹

In the case of Spain, the Ministry of Health has determined that the surveillance of COVID-19 will be carried out mainly through the autonomous communities (ACs), which will have to report confirmed cases from primary and hospital care (private and public), as well as those detected at prevention services and laboratories with diagnostic capacity.⁷ COVID-19 therefore becomes a biweekly-reporting ND with special emphasis on continuous sequencing of samples to further characterise the virus in order to identify the potential emergence of new variants.

On an international level, the World Health Organisation (WHO) updated its Global Influenza Programme (GIP)¹⁰ on 31 January 2022 to address the combined and joint surveillance of influenza and SARS-CoV-2 through the 70-year-old reference laboratory logistics infrastructure, the Global Influenza Surveillance and Response System (GISRS). This surveillance network is one of the oldest in existence and specialises in monitoring influenza.¹¹ This document sets out the rationale for how National Influenza Centres (NICs), WHO Collaborating Centres (WHO CCs), H5 Reference Laboratories, National Influenza Laboratories and COVID-19 National Laboratories should conduct this joint surveillance.

The purpose of this coordinated programme is to have clinical, epidemiological and virological information on influenza viruses and SARS-CoV-2 in the shortest possible time, which allows public health decisions to be made, among other aspects. WHO, in this document, identifies the following objectives for sentinel surveillance, distinguishing between those that are “core” and those that are “optional”. The core objectives are those that all the above-mentioned types of laboratories should perform, and the optional objectives are ideal for supplementing information by those laboratories that have the necessary technology, capacity and expertise (Table 1).

Table 1.

WHO’s core and optional objectives for combined and integrated influenza and COVID-19 surveillance.

Core objectives

• Determine the start and end of influenza and SARS-CoV-2 activity and their epidemic thresholds

• Describe the seasonality of both viruses

• Establish moderate and severe disease activity levels, both from epidemics and future pandemics

• Describe the epidemiology of mild/moderate and severe cases of influenza and SARS-CoV-2 (ARI and SARI)

• Monitor locally circulating virus types/subtypes or lineages/sub-lineages and their relationship to global and regional patterns.

• Provide candidate viruses for influenza vaccines

Optional objectives

• Identify and monitor risk groups in terms of severity and mortality

• Analyse the relationship between the strains/variants with the severity of the disease

• Generate data to estimate the economic and disease burden of influenza to help decision-makers prioritize resources and plan public health interventions

• Provide a platform to evaluate vaccine effectiveness and possibly other interventions for both influenza and SARS-CoV-2

• Monitor susceptibility to antivirals

• Describe the genetic and antigenic characteristics of circulating influenza viruses and monitor circulation of SARS-CoV-2 variants of concern and variants of interest by PCR-based assays and/or sequence analysis

• Detect unusual and unexpected events such as outbreaks or epidemiologic clusters that may indicate a change in virus characteristics

• Define and characterize pathogens associated with ILI/ARI/SARI cases beyond influenza viruses and SARS-CoV-2

ARI: acute respiratory infection; SARI: severe acute respiratory infection; WHO: world health organization; PCR: polymerase chain reaction.

Source: WHO.¹⁰

In short, future surveillance for COVID-19 consists of protocols similar to those already in place for influenza, but in this case for SARS-CoV-2. This change aims to eliminate mass testing and move to a surveillance system focused on clinical diagnosis, supported by virological diagnosis only in those cases requiring clearer identification. In fact, the Spanish Ministry of Health has defined the specific cases in which virological identification should be performed through molecular methods:⁷

• •

  People who meet vulnerability criteria: immunocompromised, elderly and pregnant people

• •

  People living in vulnerable areas

• •

  Lower respiratory symptoms requiring hospital admission

• •

  Individuals who in the last 14 days have travelled to regions where variants of interest or concern are circulating but are not circulating in our country on a community-wide basis

For this system to work properly, one of the main actions to be undertaken is to clearly define the terms ILI/ARI/SARI for COVID-19 disease, as in the case of influenza these definitions are several years old. This seems essential because a poor case definition can lead to low sensitivity and thus failure to detect cases correctly in sentinel and hospital networks. Conversely, if sensitivity is too high, warning signals of the onset of the epidemic may be issued without this being true, resulting in the mobilisation of unnecessary resources that will negatively affect patient care and the surveillance system.¹²

In the case of COVID-19, the symptoms and assessments used to define these 3 parameters have changed slightly over the last year, as the symptoms have become more clearly understood since then. In this document,¹⁰ and in the absence of a consensus of pathognomonic symptoms of COVID-19 disease that clearly define a clinical case, WHO limits the search for ILI/ARI/SARI to the classical definition of influenza. Table 2 shows the case definitions for ILI, ARI, and SARI, and the sensitivity and specificity for influenza and COVID-19. However, in our opinion, in the specific case of ILI, the low sensitivity for identifying COVID-19 (20%–55%) could represent a limitation for the detection of cases and could underestimate the real incidence of this new disease.

Table 2.

Case definitions for influenza-like illness, acute respiratory infection, and severe acute respiratory infection and their sensitivity and specificity for influenza and COVID-19.

Case definitions	Sensitivity and specificity for the diagnosis of influenza	Sensitivity and specificity for the diagnosis of COVID-19
Influenza-like illness (ILI)
• Onset of symptoms in the last 10 days, and	Sensitivity: 85%–95%	Sensitivity: 20%–55%
• Fever ≥38 °C, and	Specificity: 45%–55%	Specificity: 38%–90%
• Cough
Acute respiratory infection (ARI)
• At least cough, sore throat, shortness of breath, runny nose, with or without fever, and	Sensitivity: 94%	Sensitivity: 60%–96%
• Clinician’s judgement that the illness is due to an infection	Specificity: 27%	Specificity: 10%–45%
Severe acute respiratory infection (SARI)
• Severity (hospitalization), and	Sensitivity: 45%–70%	Sensitivity: 33%–62%
• Acute process (onset of symptoms in the last 10 days), and	Specificity: 45%–70%	Specificity: 31%–77%
• Fever (≥38 °C), and
• Cough

Source: modified from WHO¹⁰ and Kulkarni et al.¹⁹

Upgrading the surveillance of COVID-19 to a system more akin to other viral respiratory diseases should not affect the surveillance of other viruses, as it appears to have happened with influenza during the pandemic.¹³ Therefore, one of the clear objectives of this GIP update is that the new coordinated surveillance does not have a negative impact on what has already been done before. WHO argues that it is better to have data from few sites, but with sufficient safeguards (quality data and timely reporting), than to have a lot of data, but of low quality. This should start with the work of existing laboratories, and gradually build up the network with new facilities to help cover areas from where data are not being reported. Another aspect to be taken into account is that entities participating in such combined networks should make the same efforts to monitor influenza and COVID-19 at clinical, epidemiological and laboratory levels. This will ensure that there is no under-representation of one of the two viruses, resulting in suboptimal data that would not serve the purpose of improving public health protocols and decisions.

In order to carry out a coordinated surveillance for both viruses, the reference laboratories should preferably obtain samples from sentinel networks (either from PC or HC). However, this is sometimes not possible, or does not allow receiving an optimal number of samples in order to conduct adequate surveillance, which is around 100–150 samples/week. In these cases, the GIP suggests expanding surveillance to include a subgroup of non-sentinel PC and HC samples that meet the ILI/ARI/SARI definitions, or from COVID-19 laboratories whose patients meet these definitions. It is important that when reporting virological results, the sentinel or non-sentinel origin of the results is specified.

The logistics of sampling, shipping, and biosecurity are also important issues during combined surveillance. The samples of choice for influenza have always been throat and nasopharyngeal swabs.⁴ For COVID-19 this will be the same, as it is the most cost-effective sample. It is also important to note that the transport medium used is essential, since the requirements for viral culture and isolation, both for influenza and SARS-CoV-2, require the use of viral transport media without inactivators. These samples must be sent refrigerated (+4 °C) or frozen at −70 °C, and always following the international transport regulations that regulate this type of samples as a category B infectious substance, code UN3373.¹⁴

However, the handling of these samples in the laboratory requires special caution given the current regulations on SARS-CoV-2. For influenza virus culture, it is especially important to ensure through molecular methods that SARS-CoV-2 is not present in the sample. Although this virus does not seem capable of growing in the classical cell lines used for influenza (MDCK) or in embryonated eggs,¹⁵ for safety reasons, the culture of SARS-CoV-2 should only be carried out in adequate BSL-III biocontainment facilities according to international regulations.^14,16

Virological surveillance in the laboratory should be done by molecular methods, using specific PCRs to detect and distinguish influenza and SARS-CoV-2 viruses, and if possible, differentiate them from other respiratory viruses causing ARI. This is not a change from what was previously done with influenza, but WHO does emphasise the use of multiplex PCR methods to distinguish influenza and SARS-CoV-2 at the same time (as far as possible) to make co-ordinated surveillance more efficient, reduce the time to obtain results and improve laboratory throughput. There are currently many commercially available multiplex/molecular-type reagents that are capable of detecting up to 19–21 respiratory pathogens in the same sample and analysis. For its part, WHO, through the International Reagent Resource (IRR), provides influenza reference laboratories with free multiplex-type reagents capable of detecting influenza and SARS-CoV-2 in a single test.¹⁷ It remains vital to identify influenza A subtypes (H1 and H3) and influenza B lineages (Victoria and Yamagata), in order to provide as much virological and epidemiological information as possible, and to detect possible animal influenza cases from those influenza A viruses that laboratories are not capable of subtyping.

One of the most complex points in this new coordinated protocol is the sending of samples from the NICs to the WHO-CCs, in the context of the development of the annual influenza vaccine.¹⁸ For the time being, only those influenza-positive samples that have tested negative for SARS-CoV-2 should be submitted either to the sending laboratory or to another reference laboratory. If they cannot test them, the WHO-CCs request information on whether they have been previously tested or not, in order not to expose their employees to the culture of a possible SARS-CoV-2. Since the sample submission period from the NICs to the WHO-CCs is so short due to the logistics of influenza vaccine manufacturing, it is important to properly coordinate and structure the timing in the laboratory so that all information is available before samples are shared. It is likely that SARS-CoV-2 positive samples will need to be shared in the future if the continued emergence of VOI and VOC poses a challenge for COVID-19 vaccines.

Genetic characterisation is important for influenza and will be equally important for SARS-CoV-2. The GISAID platform has facilitated access to genetic sequences of influenza viruses and has been essential for this same purpose during the COVID-19 pandemic, but with SARS-CoV-2. The objective of uploading these sequences to GISAID is twofold. On the one hand, it is necessary to have data that help to represent which variants circulate in different geographical areas; and, on the other hand, to know the representativeness in percentage of these variants, in order to know their temporal dynamics. To achieve these goals, the WHO recommends that laboratories with sequencing capacity analyse about 10% of the influenza and SARS-CoV-2 positive samples received from sentinel networks by performing haemagglutinin (HA) gene sequencing for influenza and spicule (S) sequencing for SARS-CoV-2, or if available, whole genome sequencing. Samples with a high viral load should be prioritised to obtain the greatest cost-effectiveness from this process.

Conclusions

In summary, the essential steps for surveillance of COVID-19 in a similar way as for influenza viruses are to adapt sentinel surveillance systems to actively look for SARS-CoV-2 along with influenza through clinical diagnosis, and subsequent confirmation by laboratory testing in those cases described as being at increased vulnerability. It is also necessary to provide reference laboratories with the necessary diagnostic, analytical and biosafety capacity to work with this new virus. The challenge of monitoring both COVID-19 and influenza at the same time is possible with the necessary means and the commitment of professionals and managers, as the infrastructure of the NICs and the GISRS, as well as other national laboratory networks, has extensive experience and professionals with sufficient knowledge on how to adapt what was already being done to a new challenge, COVID-19 surveillance and diagnostic capacity building. All these actions have a clear objective: to provide useful information in order to facilitate the public health response and improve clinical care.

Funding

This paper has not received any funding.

Conflict of interests

The authors declare that they have no conflict of interest.