Reporting and evaluating influenza virus surveillance data: an argument for incidence by single year of age
Vaccine, 36:42, 6249–6252 (2018)
The observation that past exposures to influenza virus shape the outcome of subsequent exposures has been understood since the mid-1900’s – when Thomas Francis Jr. described the concept of “original antigenic sin.” Over the past several years, substantial evidence has accumulated that cohort effects have a profound impact on the rates of influenza virus infection and mortality based on priming with viruses encountered earlier in life. However, the detection of these cohort effects in epidemiological data remains challenging due to the antiquated methods currently employed to report surveillance data. Most agencies currently report surveillance data in broad age categories, which implicitly prioritize “age effects” over “cohort effects.” Here, based on recent surveillance data, we show that the relative risk of infection from H3N2 to H1N1 viruses varies considerably by birth cohort, a result that is simply not visible when using large age group data, such as made available by the CDC. We argue that the current status quo lacks rigorous scientific rationale and hampers scientific progress by obscuring the detection of important epidemiological phenomena. We suggest that reporting and dissemination of influenza virus surveillance data by single-year of age, over larger geographic areas if need be, should become the new norm.
Keywords: Canada, cohort analysis, infectious diseases, influenza