Redefining influenza seasonality at a global scale and aligning it to the influenza vaccine manufacturing cycle: A descriptive time series analysis

Objectives: Annual seasonal influenza vaccine manufacturing cycles align with temperate country seasonality in each hemisphere, yet influenza seasonality is poorly defined for many countries. The study introduces a novel and universal approach to defining and classifying seasonality that can be used to classify any country's influenza vaccine cycle alignment.

Methods: Countries reporting to the World Health Organization's FluNet influenza virology database in 90% of weeks during 2011 through 2017 were included. A smoothed, standardised, average proportion of influenza occurring in each week of the year was used to determine degree of seasonality based on the range of average weekly variation. The proportion of activity occurring May through October was used to align influenza activity with a hemisphere's vaccine manufacturing cycle.

Results: From 84 included countries, there were 2,239,208 positive influenza results, of which 26% were influenza type B. Degree of seasonality was moderately positively correlated with absolute value of latitude (r = 0.69, p < 0.0001). Latitude was strongly negatively correlated with the proportion of influenza occurring during May through October (r = -0.83, p < 0.0001). Thirteen countries (12% of the included global population), mainly in tropical zones, had influenza occurrence aligned with the opposite hemisphere's influenza vaccine manufacturing cycle. In tropical zones, concordance in the degree of seasonality and vaccine cycle alignment within regions and between adjacent countries was limited. In temperate zones, on average, influenza B peaked four weeks later than A.

Conclusions: Influenza activity in many countries aligns with the vaccine manufacturing cycle of the opposite hemisphere. In tropical regions, there is limited scope for grouping countries for vaccine distribution and administration.