Higher airborne pollen concentrations correlated with increased SARS-CoV-2 infection rates, as evidenced from 31 countries across the globe

Significance Coexposure to airborne pollen enhances susceptibility to respiratory viral infections, regardless of the allergy status. We hypothesized this could be also true for SARS-CoV-2 infections. To investigate this, we tested for relationships between SARS-CoV-2 infection rates and pollen concentrations, along with humidity, temperature, population density, and lockdown effects. Our unique dataset derives from 130 sites in 31 countries and across five continents. We found that pollen, sometimes in synergy with humidity and temperature, explained, on average, 44% of the infection rate variability. Lockdown halved infection rates under similar pollen concentrations. As we cannot completely avoid pollen exposure, we suggest wide dissemination of pollen−virus coexposure information to encourage high-risk individuals to wear particle filter masks during high springtime pollen concentrations.

. Switzerland as case study to illustrate the magnitude of the pollen effect. Number of COVID-19 cases, infection rates (7-day moving average) and airborne pollen concentrations (pollen m -3 ) are shown as a function of date for three Swiss cantons.

Supplementary Information Text for Fig. S1
The example of Switzerland illustrates the relative importance of the effects of pollen, population density and lockdown. Switzerland was one of the countries with the highest pollen concentrations for several days during the exponential phase of the pandemic, which made it possible to compare three cities in Switzerland located close to each other and having comparable climate and population densities, but different pollen concentrations.
The pollen spike on 12 March in Zurich was 5 times that of Geneva and 8 times higher than in Lausanne. As a result, the overall tendency of reduction of the infection rate from 11 March was broken in Zurich (ovals mark the spike and its effect) -and the anomaly faded out only a week later. Of note, the pollen spikes that occurred during the lockdown phase exhibited a by far less pronounced effect. Figure S2. Donut charts of the biodiversity of all monitored airborne pollen per region, in 70 regions across the Northern Hemisphere. Focus was given on allergenic pollen primarily (color depiction), while the rest were denoted as 'Others' (black color depiction). The numbers in the middle of each chart indicate the total number of all pollen grains recorded during the study period for each region. The Southern Hemisphere has been excluded here, as the daily average pollen concentration was only 17 pollen grains per cubic meter of air, the majority belonging to grass pollen (in contrast to a daily average of more than 200 pollen in the Northern Hemisphere). Figure S3. Scatterplot of the distribution of the total confirmed COVID-19 cases in relation to pollen totals. To analyze, in parallel, the infection cases in association with pollen abundances, we differentiated the 1 st quartile of pollen data (low concentrations, as in the Southern Hemisphere), which accounted for a threshold 1980 pollen.

Supplementary Information Text for Fig. S3
To evaluate whether the total amount of pollen per region is the most representative and appropriate for our research question, we assessed the whole diversity of airborne pollen taxa for each and every site out of 248 originally acquired and for all the regions and all 31 countries included in the analysis. It is clearly demonstrated that the effect of our originally calculated totals of pollen are almost identical to the newly calculated allergenic pollen by 90.7%. Regarding a potential change in the originally calculated pollen signal in the whole analysis, if we apply a fitting line, the difference of the coefficient of determination is only of the magnitude of 0.03 (significant and positive in both scenarios). The above similarity is simply due to the fact that our study period refers almost exclusively (for the Northern Hemisphere) to airborne pollen from winter-spring trees and shrubs. Table S1. Overview of data sources for pollen and COVID-19 cases. Data from a total of 130 single aerobiological measurement stations included in the analysis. When more than 2 sites were equally eligible, we picked the site(s) with the highest population. Last data access on 10 May 2020.

Country
Köppen climate