Contacts in context: large-scale setting-specific social mixing matrices from the BBC Pandemic project

Petra Klepac; Adam J Kucharski; Andrew JK Conlan; Stephen Kissler; Maria Tang; Hannah Fry; Julia R Gog

The measured age-specific social mixing matrices also varied considerably between different types and settings. Reported contacts at home in the population contact matrix followed a strong age-assortative pattern (strong diagonal band), with inter-generational mixing shown by the offdiagonal bands in contacts (Fig 3A-B) , which is especially pronounced in physical contacts at home. Contacts at work showed less age-assortativity than contact at home (Fig 3C-D) , and were predominantly non-physical ( Fig 3D) . Within school-aged groups, more contacts were reported on average at school than in other settings (Fig 3E-F ), but for a very narrow age-band. Overall contacts in other settings (i.e. not home, work or school) were age assortative for younger groups, but less assortative for older groups, with an off diagonal peak in contact intensity between older participants and other adults (Fig 3G-H) . Physical contacts in other settings were less common, but also exhibited the transition from age-assortativity to less structured mixing in older age groups ( Fig 3G) .