Lag associations of four types of pollens with respiratory mortality in Michigan 2006-2021

Abstract

BACKGROUND AND AIM: Research on associations between pollen and mortality from respiratory-related causes, such as pneumonia and chronic obstructive pulmonary disease (COPD), is limited. Levels of plant-based aeroallergens such as common ragweed are rising as growing seasons lengthen and CO2 levels rise. We determined associations of four species of pollens, ragweed, deciduous trees, grass and evergreen trees with pneumonia and COPD-related mortality in Michigan, USA. METHODS: Death records were obtained for all deaths within Michigan for the years 2006 through 2021 from the Michigan Department of Health and Human Services (MDHHS). The dataset was filtered on specific causes of death based on International Classification of Diseases (ICD) codes. Pollen data came from a prognostic model based raster of daily pollen counts at 25km resolution. Case-crossover models with distributed lag non-linear bases for pollen were used to test associations between lags of daily pollen counts with mortality. RESULTS:Among 1,399,714 deaths between 2006 and 2021, 130,584 were from all respiratory causes, 71,427 from COPD and 22,723 from pneumonia. Of the four pollen types, only ragweed was found to be associated with all-cause respiratory mortality, and only for high concentrations (100-200 grains/m3) at lags 5-8. Similarly, only ragweed was associated with deaths from COPD and only at lags 4-8 for high concentrations (100-300 grains/m3.) Short term exposures (lags 0-2) of increasing intensity to grass pollens were suggestive of increased risk of death from pneumonia, but this relationship was not significant. CONCLUSIONS:Our results suggest relationships of exposures of pollens of specific plants with specific causes of death. Intense exposure to ragweed pollens may increase risk for mortality from COPD or other respiratory causes, and grass pollens might be associated with death from pneumonia. Future work will consider how model-based predictions of pollen counts can be used to test associations of pollen exposure with mortality risk over wide areas. KEYWORDS: allergy, pollen, climate change, respiratory, copd, mortality