Executive Summary Insights Background Community Join in
Executive Summary of Round 1 - 2024/2025

Introduction

We used scenario modelling to provide mid- to long-term projections of influenza burden (using primary care consultation data) and COVID-19 burden (using hospital admission data) for individuals aged 65 years and older for winter and spring 2024-2025.

Unlike forecasts (presented on RespiCast), which provide short-term predictions based on trends in observed data, scenario modelling explores mid- to long-term projections that explore specified uncertainties defined in scenarios. The different scenarios captured uncertainties around influenza virus transmission, the waning of vaccine-induced immunity to COVID-19, as well as influenza and COVID-19 vaccination coverage. These scenario projections can support policymakers by providing insights on the potential impact of public health interventions to mitigate the burden of disease of influenza and COVID-19.

Modelling teams submitted their projections by early September 2024 with the projections of above-mentioned burden from September 2024 to May 2025.

Method

RespiCompass, the European respiratory diseases scenario modelling hub, is comprised of different modelling groups working collaboratively, with input from disease area experts working on respiratory virus surveillance. We applied an ensemble approach which combines projections from different models developed by separate groups, based on a shared set of model assumptions. This ensemble approach has been shown to consistently deliver more reliable projections [Howerton et al., 2023] when compared to individual model projections. Descriptions of the scenarios and shared assumptions used for influenza and COVID-19, and underlying data used to fit models can be found under Insights, the ensemble methodology is described in Background.

For influenza, burden was defined as an estimate of the number of primary care consultations for those aged 65 years and above presenting with influenza-like illness (ILI)[*] and testing positive for influenza virus infection. We call this indicator “ILI+”.

ILI+ = ILI x ‘sentinel test positivity[**] for influenza’

Sentinel test positivity was available for all countries apart from Malta, Iceland, Croatia, Romania, Latvia, and Finland, for which non-sentinel test positivity was used.

For COVID-19, burden was defined as the cumulative number of SARS-CoV-2 positive hospital admissions for those aged 65 years and above.

Results

  • Across the EU/EEA, influenza primary care consultation burden for those aged 65 years and above is expected to be 36% (95% UI[***]: -2% to 92%) higher in winter/spring 2024-2025 compared to 2023-2024. For context, the estimated burden for winter/spring 2024-2025 falls between numbers observed during the same period in 2017-2018 and 2018-2019.
  • A 10% increase in influenza vaccine coverage compared to the coverage in autumn/winter 2023-2024 was associated with an average 6% (95% UI: 6% to 7%) reduction in influenza burden.
  • COVID-19 hospitalisation burden for those aged 65 years and above in the EU/EEA is projected to be higher in winter/spring 2024-2025 compared to 2023-2024.
  • A 10% increase in COVID-19 vaccine coverage compared to the coverage in autumn/winter 2023-2024 was associated with a 7% (95% UI: 6% to 8%) reduction in COVID-19 hospitalisation burden.
  • There is a strong correlation between vaccine coverage and decreased burden for both influenza and COVID-19 in winter/spring 2024-2025, which holds across the EU/EEA. In countries with a higher absolute vaccine coverage, the relative reduction in burden was projected to be higher.

Limitations

While RespiCompass strives to model a comprehensive picture of COVID-19 and influenza burden in the EU/EEA, a main limitation is the availability of age-specific data. As a result, we focus on specific aspects of disease burden, the limitations of which include:

  • While the models consider the whole population, the burden reported here is specific to individuals aged 65 years and older for COVID-19 and influenza. Results do not capture the wider burden prevented by vaccination in other population groups, for example the burden of influenza in children; or vulnerable populations, such as pregnant women or those with underlying conditions under 65 years of age (for both influenza and COVID-19).
  • It is important to note that the burden of influenza primary care consultations (‘ILI+’) cannot be directly translated into hospitalisation rates. We focus on ILI+ because it was the influenza surveillance indicator with the highest coverage across EU/EEA countries, with comparatively limited coverage for influenza hospitalisation data.
  • Given that different burden indicators are used for influenza (‘ILI+’) and COVID-19 (hospitalisation), it is not possible to add the estimates for the two disease together for a combined burden estimate.
  • Due to differences between national surveillance systems, ensemble projections for COVID-19 hospitalisations or ILI+ should not be directly compared between countries.
  • Due to how ages were grouped in the available datasets, COVID-19 vaccine coverage scenarios were defined for the 60+ age group. This differs from influenza vaccine coverage scenarios, which were defined for the 65+ age group. ECDC modelling of COVID-19 vaccination impact [ECDC Technical report 2023] has shown that a five-year difference in cut off for vaccination scenarios can be considered small and is expected to have a relatively muted impact on hospital burden.
  • It is important to note that the scenario definitions do not consider an influenza vaccine mismatch nor a novel COVID-19 variant of concern.
  • The models used in this analysis are calibrated based on the available data. Data quality will inevitably affect the accuracy and reliability of the models.

Acknowledgements

We acknowledge the “European Scenario Modelling INfectious Diseases” (EU-SMIND) consortium (ISI foundation, Hasselt University, University of Antwerp, Safinea Ltd., Helmholtz Centre for Infection Research, and the London School of Hygiene & Tropical Medicine), whose technical expertise has been essential for the execution of this project. We further acknowledge the EU/EEA countries in the 'European Respiratory Virus Disease Network' that report data to TESSy (and therefore ERVISS).

Notes

[*] Influenza-like illness (ILI) refers to a group of symptoms linked to influenza disease, such as fever, cough, sore throat, and muscle aches (EU case definitions can be found on the ECDC Website). This is used by public health experts to track flu activity.

[**] In most EU/EEA member states, a sample of patients presenting with ILI at sentinel practitioners are tested for different viruses (influenza, RSV, SARS-CoV-2). The influenza test positivity is obtained by dividing the number of influenza positive tests by the total number of tests performed for influenza.

[***] Uncertainty intervals are computed based on the ensemble method (see Insights section).