Regional characteristics of influenza seasonality patterns in mainland China, 2005-2017: a statistical modeling study

Background: The seasonal and antigenic characteristics of influenza are crucial to help understanding influenza activity and inform vaccine recommendations.

Methods: We employed a generalized linear model with harmonic terms to quantify the seasonal pattern of influenza in China during 2005-2017, including amplitude (circulatory intensity), semiannual periodicity (given two peaks a year), annual peak time, and epidemic duration. The antigenic differences were distinguished as antigenic similarity between 2009 and 2020. We categorized regions above 33° N, between 27° N and 33° N, and below 27° N as the north, central, and south regions, respectively.

Results: We estimated that the amplitude in the north region (median: 0.019, 95% CI: 0.018-0.021) was significantly higher than that in the central (median: 0.011, 95% CI: 0.01-0.012, p < 0.001) and south region (median: 0.008, 95% CI: 0.007-0.008, p < 0.001) for A/H3N2. The A/H3N2 in the central region had a semiannual periodicity (median: 0.548, 95% CI: 0.517-0.577), while no semiannual pattern was found in other regions or subtypes/lineages. The antigenic similarity was low (below 50% in the 2009-10, 2014-15, 2016-18, and 2019-20 seasons) for A/H3N2.

Conclusions: Our study depicted the seasonal pattern differences and antigenic differences of influenza in China, which provides information for vaccination strategies.