A comparison of coronavirus disease 2019 and seasonal influenza surveillance in five European countries: France, Germany, Italy, Spain and the United Kingdom

Abstract Background In response to the coronavirus disease (COVID‐19) outbreak that unfolded across Europe in 2020, the World Health Organisation (WHO) called for repurposing existing influenza surveillance systems to monitor COVID‐19. This analysis aimed to compare descriptively the extent to which influenza surveillance systems were adapted and enhanced and how COVID‐19 surveillance could ultimately benefit or disrupt routine influenza surveillance. Methods We used a previously developed framework in France, Germany, Italy, Spain and the United Kingdom to describe COVID‐19 surveillance and its impact on influenza surveillance. The framework divides surveillance systems into seven subsystems and 20 comparable outcomes of interest and uses five evaluation criteria based on WHO guidance. Information on influenza and COVID‐19 surveillance systems were collected from publicly available resources shared by European and national public health agencies. Results Overall, non‐medically attended, virological, primary care and mortality surveillance were adapted in most countries to monitor COVID‐19, although community, outbreak and hospital surveillance were reinforced in all countries. Data granularity improved, with more detailed demographic and medical information recorded. A shift to systematic notification for cases and deaths enhanced both geographic and population representativeness, although the sampling strategy benefited from the roll out of widespread molecular testing. Data communication was greatly enhanced, contributing to improved public awareness. Conclusions Well‐established influenza surveillance systems are a key component of pandemic preparedness, and their upgrade allowed European countries to respond to the COVID‐19 pandemic. However, uncertainties remain on how both influenza and COVID‐19 surveillance can be jointly and durably implemented.


| INTRODUCTION
The first cases of coronavirus disease , caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), were reported in Europe in January 2020. 1 On 11 March 2020, the World Health Organisation (WHO) identified COVID-19 as a pandemic. 2 A multicomponent surveillance system is critical to monitor the epidemiology, virology and health consequences of this new respiratory pathogen but is compounded by the novelty of the virus, the definition of the symptomatology and the role of asymptomatic carriage. 3 The WHO called for a repurposing of influenza surveillance systems as an efficient, costeffective and sustainable approach to respond quickly to COVID-19 4,5 and the European Centre for Disease Prevention and Control (ECDC) subsequently issued guidance on surveillance strategies for COVID- 19. 6 Influenza surveillance systems are well established in Europe, reflecting the WHO Global Influenza Surveillance and Response System requirements 7 and supplemented by national tools or pan-European initiatives such as InfluenzaNet for non-medically attended participatory syndromic surveillance systems 8 and EuroMOMO for excess mortality modelling. 9 However, the objectives of routine seasonal influenza surveillance and COVID-19 surveillance are not exactly the same. For both influenza and COVID-19, surveillance enables an understanding of the epidemiology, virology and geographic spread of disease and measuring of disease burden, severity and impact of prevention measures. However, in the context of a pandemic, COVID-19 surveillance must also take into account disease containment using a test, track and trace approach and assess the overall disruption caused to healthcare systems, economies and communities. 10 Additionally, the higher transmission rate for SARS-CoV-2 means that its surveillance systems may need to be more reactive.
Although the enhancement and expansion of existing influenza systems for COVID-19 surveillance could benefit influenza surveillance, there could also be a diversion of resources from influenza surveillance towards COVID- 19. A solution could be to combine surveillance, without which a shift in favour of  could have an impact on the quality of influenza surveillance data. 11 The public health consequences could include an impact on the selection of seasonal influenza vaccine strains, less accurate burden of influenza disease estimates that could lead to reduced public awareness of disease severity, increased vaccine hesitancy, reduced vaccine coverage and reduced population-level protection if influenza viruses were to circulate as in previous years. Furthermore, WHO has warned that a shift from influenza to  surveillance may make it more difficult to avert future influenza pandemics. 12 The purpose of this research was to understand to what extent influenza surveillance systems in five western European countries were repurposed, strengthened or complemented by new components for COVID-19 surveillance and the implications for future influenza surveillance.

| Comparative framework
We applied a pre-existing framework of influenza surveillance systems in France, Germany, Italy, Spain and the United Kingdom with seven subsystems (Table 1) and five evaluation criteria (Table 2) to COVID-19 surveillance. 13 The seven subsystems-(1) non-medically attended surveillance, (2) virological surveillance, (3) community surveillance, (4) outbreak surveillance, (5) primary care surveillance, (6) hospital surveillance, and (7) mortality surveillance-encompass a list of 20 comparable outcomes following a scale of severity from non-medically attended suspected infection to lethal cases ( Table 1). The five evaluation criteria-granularity, timing, representativeness, data sampling, and communication-are associated with a list of subcriteria defined in WHO guidance ( Table 2). The framework was updated to include an extra outcome in the Community Surveillance subsystem (serum samples for population immunity estimates: Outcome 3.2 in Table 1) and two extra subcriteria in the Communication evaluation criterion (use of infographics/dashboard and interactive map:

| Data interpretation
The results of the comparative analysis were reviewed, discussed and adjusted with a panel of 28 experts (including the authors), incorporating epidemiologists, virologists, general practitioners, public health researchers and pharmaceutical industry medical experts from each country during a virtual event in September 2020.

| Surveillance subsystems
Existing influenza surveillance systems were to a large extent also used for COVID-19 surveillance in each country, but were expanded, and complemented with additional surveillance components. Table 3 (France, Germany, Italy, Spain) and  Beyond demographic information, importance was put on COVID-19 symptoms and complications. Contrary to influenza surveillance, most countries provided a breakdown between asymptomatic and symptomatic cases. The level of severity of the infection was also closely monitored, and data on hospitalisation or mechanical ventilation were often available. The outcome of a hospital admission, that is, death or recovery, was also part of the daily national statistics, and in Italy, cases were classified using a severity scale.

| Data timing
Surveillance frequency was adapted to improve responsiveness, with daily reporting of COVID-19 data compared with weekly reporting for influenza. COVID-19 surveillance also shifted from a seasonal to a year-round approach.

| Data representativeness
There was a significant shift from sentinel-based influenza surveillance in most countries, that relied on volunteer medical personnel in primary and secondary care, to a mandatory notification system, encompassing all medical settings as was already performed for influenza surveillance in Germany. Syndromic surveillance based on ARI and influenza-like (ILI) symptoms was also replaced by a systematic surveillance of possible COVID-19 cases, regardless of the presence of the symptoms, but taking into account travel history and personto-person contacts.
The systematic surveillance enforced through mandatory notification was intended to cover all medical settings, with more hospitals and laboratories involved in surveillance of COVID-19 than influenza.
Nursing/care homes also became better covered, and the aetiology of any outbreak was systematically investigated through testing. Lastly, surveillance of COVID-19 in closed settings, unlike influenza, was not limited to state-run institutions such as educational facilities or prisons but expanded to include private settings such as workplaces, restaurants and cafes.

| Sampling strategy
The massive and unprecedented use of RT-PCR testing led to a step change in determining the viral aetiology of suspected cases with or without symptoms, and laboratory-confirmed infection became the

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1111/irv.12941.

DATA AVAILABILITY STATEMENT
All data are available in public repositories, referenced throughout the article. Any further requests should be directed to the Corresponding Author.