Back

Sporadic long-range contacts in out-of-venue settings dominate mass gathering events in Germany

Schulz, S.; Rincon Hidalgo, A.; Jarynowski, A. K.; Zambrano, M.; Suer, J.; Thampi, A.; Ferretti, L.; Phuong, H. T.; Xu, C.; Mikolajczyk, R.; Pastor, R.; Jaeger, V. K.; Karch, A.; Belik, V.

2026-08-21 infectious diseases
10.64898/2026.08.19.26359786 medRxiv
Show abstract

Mass gathering events (MGEs) play a critical role for infectious disease dynamics on a population level as they provide opportunities for superspreading; however, underlying mechanisms remain insufficiently understood. We analyzed nationwide GPS-based, individual-level location data from mobile phone users in Germany between April and August 2024 with 16m spatial precision. Potentially infectious contacts were inferred from close co-location and linked to contact settings using OpenStreetMap data. Various MGEs, including EURO 2024 matches, major concerts, festivals, and fairs were compared using a common contact metric. Non-football events generated substantially more contacts than football events. While overall national contact numbers remained stable, MGEs produced so-called "small-world" contacts which gather people from distant locations into close proximity and could strongly enhance infectious disease dynamics. Crucially, most high-risk contacts occurred within two hours before the event, not at the event itself, and concentrated in public transport, leisure, and event-adjacent areas. Our work provides the first systematic and comparative evaluation of contact exposure across various types of MGEs and contact settings. Event-type-specific dynamics, particularly indirect and mobility-driven contacts, critically shape infection risk. These insights can inform accurate transmission modeling, targeted intervention and event-management strategies.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.