Characterization and forecast of global influenza (sub)type dynamics

The (sub)type composition of seasonal influenza waves varies in space and time. (Sub)types tend to have different impacts on population groups; therefore, understanding the drivers of their co-circulation and anticipating their composition is important for epidemic preparedness. FluNet provides data on influenza specimens by (sub)type for more than 150 countries. However, due to surveillance variations across countries, global analyses usually focus on (sub)type compositions, a kind of data difficult to treat with advanced statistical methods. We used Compositional Data Analysis to circumvent the problem and study trajectories of annual (sub)type compositions of countries. First, we examined global trends from 2000 to 2023. We identified a few seasons which stood out for the strong within-country (sub)type dominance due to either a new virus/clade taking over (2003/2004 season, A/H1N1pdm pandemic) or (sub)types spatial segregation (COVID-19 pandemic). Second, we showed that geographical factors, most notably international mobility, concurred in shaping countries composition trajectories between 2010 and 2019. Trajectories clustered in two macroregions characterized by (sub)type alternation vs. persistent mixing. Finally, we defined five algorithms for forecasting the next years composition and found that incorporating the global history of (sub)type composition in a Bayesian Hierarchical Vector AutoRegressive model improved predictions compared with naive methods. The joint analysis of spatiotemporal dynamics of influenza (sub)types worldwide revealed a hidden structure in (sub)type circulation that can be used to improve predictions of the (sub)type composition of next years epidemic according to place.