Global patterns and drivers of influenza decline during the COVID-19 pandemic

Abstract

Objectives: The influenza circulation reportedly declined during the COVID-19 pandemic in many countries. The occurrence of this change has not been studied worldwide nor its potential drivers.

Methods: The change in the proportion of positive influenza samples reported by country and trimester was computed relative to the 2014-2019 period using the FluNet database. Random forests were used to determine predictors of change from demographical, weather, pandemic preparedness, COVID-19 incidence, and pandemic response characteristics. Regression trees were used to classify observations according to these predictors.

Results: During the COVID-19 pandemic, the influenza decline relative to prepandemic levels was global but heterogeneous across space and time. It was more than 50% for 311 of 376 trimesters-countries and even more than 99% for 135. COVID-19 incidence and pandemic preparedness were the two most important predictors of the decline. Europe and North America initially showed limited decline despite high COVID-19 restrictions; however, there was a strong decline afterward in most temperate countries, where pandemic preparedness, COVID-19 incidence, and social restrictions were high; the decline was limited in countries where these factors were low. The "zero-COVID" countries experienced the greatest decline.

Conclusion: Our findings set the stage for interpreting the resurgence of influenza worldwide.

Keywords: COVID-19 pandemic; Global analysis; Influenza; Regression trees.

Figure 1

Change in influenza circulation during the COVID-19 pandemic relative to the prepandemic period. (a) Weekly counts of processed and positive tests of influenza reported to FluNet for all 166 countries included in the database from December 2014 to January 2022. The green shaded area indicates the COVID-19 pandemic period considered in the study. The six blocks indicate the trimesters. The week in which COVID-19 was declared a pandemic by World Health Organization is reported as reference. (b) Percentage of positive tests for the prepandemic and COVID-19 pandemic periods (December 2014-December 2019 and March 2020-September 2021, respectively), for all 376 countries and trimesters satisfying the filtering criteria on the FluNet data. For each trimester-country, the x coordinate is the average percentage of positive tests of the 5 years included in the prepandemic period, while the y coordinate is the percentage of positive tests during the COVID-19 pandemic period. The size of the dots is proportional to the number of samples found in FluNet for the pandemic period. Dots’ color indicates the log relative influenza level. As guides to the eyes, the three dashed lines indicate the level curves of the L.R.I.L. equal to -2, -1 and -0.69, which correspond to flu reductions of 99%, 90% and 50%, respectively. L.R.I.L., log relative influenza level; nb, number.

Figure 2

Influenza decline during the first 18 months of COVID-19 pandemic by trimesters-countries. Maps of the log relative influenza level for the six trimesters considered in the analysis. The gray color indicates trimesters-countries not included in the analysis.

Figure 3

Importance of covariates predicting influenza decline in random forest analysis. Importance of covariates as predictors of the log relative influenza level. In green the 11 covariates selected as significant to build the model with the minimum prediction error following the Breiman's rule. Black segments show the standard deviations of the importance. IDVI, infectious disease vulnerability index; nb, number; RH, relative humidity; T, temperature.

Figure 4

Regression tree analysis of influenza decline and characteristics of the identified subgroups. (a) Regression tree obtained with the variables selected in Figure 3. We report here the first four splits, which partition the observations in five groups. The full tree is reported in Figure S4 of the Supplementary material. For each node the average log relative influenza level and the number of observations are reported (the former is also indicated with a color scale). (b) Characteristics of each group. For each variable the color of the circle indicates the percentile of the whole dataset distribution the median of the group corresponds to. The percentile value is also indicated inside the circle. The size of the circle increases with the number of observations of the group. IDVI, infectious disease vulnerability index; nb, number; RH, relative humidity; T, temperature.