National Influenza Surveillance Systems in Five European Countries: A Qualitative Comparative Framework Based on WHO Guidance – BMC Public Health

Using the qualitative framework described, detailed data for the comparison criteria are presented for each surveillance subsystem (community surveillance without medical assistance, virological surveillance, community surveillance, epidemic surveillance, primary care surveillance, hospital surveillance and surveillance mortality) in the supplementary file. A synthetic overview of these data for each country is presented in the table 4.

Table 4 Overview of the comparative framework of influenza surveillance systems

Presentation of the comparative framework

France and the United Kingdom (England and Wales) had the widest reach across the 7 surveillance systems, monitoring 16 out of 19 outcomes, thanks to the breadth of outcomes covered by hospital surveillance. All five countries included in the analysis met basic WHO requirements, with well-established laboratory networks and sentinel systems for primary care and hospital surveillance.

For virological surveillance (surveillance subsystem 2: Table 4), each country had one or more WHO National Influenza Centers (NICs) to perform strain typing and subtyping, sequence the whole genome, and assess antiviral drug resistance of samples from influenza provided by sentinel programs.

Primary Care Surveillance (Surveillance Subsystem 5: Table 4) was also well developed, with all countries having sentinel systems of general practitioners (including paediatricians) reporting and testing patients with symptoms of ARI and/or ILI. No data were collected on excessive GP visits or absenteeism associated with influenza, except for Germany where statistical modeling was in place for primary care monitoring.

Hospital Monitoring (Monitoring Subsystem 6: Table 4) has also been implemented in all countries, but with differences in scope. In France, England and Wales, surveillance systems routinely monitored emergency department visits for ILI. The German surveillance included statistical modeling of influenza-associated hospitalizations, but did not cover emergency room visits or intensive care unit (ICU) admissions. In Spain, hospital surveillance included all outcomes except emergency room visits and statistical modeling for influenza-related hospitalizations. In Italy, hospital surveillance focused on the most severe outcomes, reporting only data on ICU admissions and no published data on other outcomes of this subsystem.

All countries had community surveillance and unmedicated mortality surveillance systems in place (although in Spain online survey data has not been published since 2016) (surveillance subsystems 1 and 7: Table 4). These were coordinated by the European initiatives InfluenzaNet and EuroMOMO respectively. It should be noted that Italy and Spain modeled excess all-cause mortality using the EuroMOMO model, while France, England and two German states (Berlin and Hesse) produced excess mortality estimates attributable to influenza using the FluMOMO model. However, the Robert Koch Institute uses its own model to estimate excess mortality estimates from influenza.

There were significant differences between the five countries for community surveillance and outbreak surveillance (surveillance subsystems 3 and 4: Table 4). Laboratory-confirmed or laboratory-confirmed cases have been reported from sentinel laboratories in France, Spain and the United Kingdom and suspected or laboratory-confirmed influenza outbreaks in some closed settings (e.g. schools or nursing homes) have were published on a weekly basis. Germany has required notification of all laboratory-confirmed and community-reported cases and outbreaks in closed settings. In Italy, notifiable data was not collected (laboratory confirmed reported cases) or published by health authorities (closed homes).

No country has collected data on complications and events after hospitalisation, ie the course of care.

Performance on the 5 comparison criteria

The main conclusions for the qualitative comparison criteria (granularity, timing, representativeness, sampling strategy and communication [as defined in Table 3]) for each monitoring subsystem are described below. Further details are included in the supplementary file.

Data granularity was below WHO standards in most countries, especially for stratification by risk condition (defined in Table 3), which was available in France and Spain for some results, but not in the UK, Germany and Italy (see additional file for details). Except in the UK and parts of Spain, lack of information on flu vaccination status was also common; this was particularly evident in France, Italy and Germany due to the lack of both a vaccination registry and standardized vaccination software tools in the practices of general practitioners and paediatricians.

The timing of data was well aligned between countries, with monitoring generally starting in week 40 (end of September) and ending in week 20 (mid-May) the following year and most results provided on a weekly basis.

The representativeness of the data should be affected by important differences in the proportion of the population covered by sentinel surveillance by general practitioners and paediatricians. Territorial and population coverage also varies for hospital influenza surveillance, with 60-70% of the French population and about 50% of the Spanish population but only 6% of the German population covered by national surveillance systems. flu hospital. Data was not available for Italy and the UK at the time of the study.

The sampling strategy was broadly aligned across countries, although there were nuances in terms of the syndromic criteria (ARI, ILI, or severe ARI) used in hospital surveillance and the swab strategy applied by sentinel physicians and pediatricians. However, the use of the reverse transcriptase polymerase chain reaction (RT-PCR) test was the main standard for virological analysis coordinated by the national influenza reference centers in each country.

Data reporting was inconsistent across the five countries. In the UK, France, Germany and Spain, consolidated weekly and annual reports were provided in a timely, easily accessible and manageable manner, with the exception of national excess mortality estimates in Germany which were only available one year late. In contrast, in Italy, influenza surveillance data has been published in several formats, in a range of reports and via various web pages. The five countries shared weekly surveillance data, including outside the usual epidemic period, on influenza circulation and virology with WHO and ECDC.

Country specific observations


In general, a wide range of surveillance subsystems have been used, particularly for virological, epidemic, primary and secondary surveillance. Several tools were used for community surveillance, but the use of modeling for influenza-associated events was limited except for estimating mortality. There was a high level of data granularity, particularly for risk conditions, and data representativeness was good due to large sentinel networks in hospital surveillance. This hospital monitoring included alert thresholds regarding the number of patients admitted to the emergency department that triggered different responses based on hospital-specific strategies. Information on post-hospitalization events (eg impact on frailty/autonomy) and the effect of influenza on the economy and health systems was limited.


Overall, a large number of surveillance outcomes were covered (including excessive consultations and absenteeism), but emergency room visits were not included in the surveillance and there was very little information available for intensive care admissions. Estimates of excess mortality did not include a disaggregation by age, and the European FluMOMO model [26] was only used in two Länder (Berlin and Hesse).

There was a lack of granularity for risk conditions and vaccination status and limited information on complications and treatment.

The accuracy of data in hospital surveillance was affected by an extensive syndromic surveillance program that included influenza as well as other acute lower respiratory infections, and by excessive hospitalization estimates based on data shared by primary care general practitioners, likely leading to undercounting of cases. Reporting some monitoring data in graphical rather than tabular form precludes their use in further research.


There was no systematic reporting of influenza cases in the community, while results related to ARI/ILI outbreaks in closed settings, emergency room visits and hospital admissions were collected but not publicly broadcast. There was a lack of granularity by age and condition at risk, the latter not being collected by a form used for sentinel surveillance of general practitioners, and information on the severity of influenza among cases admitted to ICUs was limited.


Virological, outbreak, primary care, hospital and mortality surveillance are all covered, while online survey data on community ILI surveillance without medical assistance were conducted for the previous 8 influenza seasons [25, 27]but not published since 2016. ILI cases reported by all GPs through compulsory notification Enfermedades de Declaración Obligatoria (EDO) were not included in the annual report because data on the incidence of SG of sentinel physicians were preferred. For hospital surveillance, there was no information on emergency department visits due to influenza. Despite the standardized methods of flu surveillance in all regions coordinated by the Spanish Institute of Public Health[[[[28], regional disparities were observed in terms of data granularity and proportion of the population covered by sentinel networks, and Galicia, Murcia and Aragon did not provide ILI data because they did not have no sentinel surveillance networks in place at the time of the study. the research[[[[29].


There was a wide variety of monitoring tools used to cover all sub-systems and a high level of data granularity, except for data on at-risk conditions which were not made public except in the research papers. However, as for the other countries, monitoring did not extend to post-hospital follow-up of discharged patients, ie the course of care.

#National #Influenza #Surveillance #Systems #European #Countries #Qualitative #Comparative #Framework #Based #Guidance #BMC #Public #Health

Leave a Comment

Your email address will not be published. Required fields are marked *