Influenza activity in Europe during eight seasons (1999–2007): an evaluation of the indicators used to measure activity and an assessment of the timing, length and course of peak activity (spread) across Europe

BACKGROUND: The European Influenza Surveillance Scheme (EISS) has collected clinical and virological data on influenza since 1996 in an increasing number of countries. The EISS dataset was used to characterise important epidemiological features of influenza activity in Europe during eight winters (1999–2007). The following questions were addressed: 1) are the sentinel clinical reports a good measure of influenza activity? 2) how long is a typical influenza season in Europe? 3) is there a west-east and/or south-north course of peak activity ('spread') of influenza in Europe? METHODS: Influenza activity was measured by collecting data from sentinel general practitioners (GPs) and reports by national reference laboratories. The sentinel reports were first evaluated by comparing them to the laboratory reports and were then used to assess the timing and spread of influenza activity across Europe during eight seasons. RESULTS: We found a good match between the clinical sentinel data and laboratory reports of influenza collected by sentinel physicians (overall match of 72% for +/- 1 week difference). We also found a moderate to good match between the clinical sentinel data and laboratory reports of influenza from non-sentinel sources (overall match of 60% for +/- 1 week). There were no statistically significant differences between countries using ILI (influenza-like illness) or ARI (acute respiratory disease) as case definition. When looking at the peak-weeks of clinical activity, the average length of an influenza season in Europe was 15.6 weeks (median 15 weeks; range 12–19 weeks). Plotting the peak weeks of clinical influenza activity reported by sentinel GPs against the longitude or latitude of each country indicated that there was a west-east spread of peak activity (spread) of influenza across Europe in four winters (2001–2002, 2002–2003, 2003–2004 and 2004–2005) and a south-north spread in three winters (2001–2002, 2004–2005 and 2006–2007). CONCLUSION: We found that: 1) the clinical data reported by sentinel physicians is a valid indicator of influenza activity; 2) the length of influenza activity across the whole of Europe was surprisingly long, ranging from 12–19 weeks; 3) in 4 out of the 8 seasons, there was a west-east spread of influenza, in 3 seasons a south-north spread; not associated with type of dominant virus in those seasons.