Comparison of PM2.5 concentrations estimated using eleven different methods and their associations with cognitive outcomes in the ARIC cohort
Year: 2022
Type: conference-abstract
Abstract: Background and Aim: Evidence of associations between fine particulate matter (< 2.5 µm aerodynamic diameter) (PM₂.₅) and dementia-related outcomes remains mixed. Existing studies have used a variety of methods to estimate exposure, but the literature comparing the impact of estimation method on exposure and health effects is limited. This study aims to assess agreement between PM₂.₅ concentrations across 11 estimation methods and to compare estimated associations between PM₂.₅ and dementia-related outcomes using data from the Atherosclerosis Risk in Communities (ARIC) study cohort, which recruited in 1987-1989 from four regional sites: Forsyth County, NC; Jackson, MS; the suburbs of Minneapolis, MN; and Washington County, MD. Methods: Using the ARIC cohort, we estimated address-based mean PM₂.₅ concentrations (2000-2007) using 11 different methods. We assessed agreement between method-specific PM₂.₅ concentrations using descriptive statistics and plots, overall and by site. We used adjusted linear and logistic regression to estimate associations between method-specific PM₂.₅ concentrations and Visit 5 (2011-2013) cognitive scores and brain MRI outcomes, stratified by study site and then combined with random-effects meta-analysis. Results: PM₂.₅ concentrations from most estimation methods had good agreement across sites and lower agreement within each site. Within-site exposure variation was limited for some methods. Point estimates for PM₂.₅-cognitive outcome associations varied by method and generally did not support associations between PM₂.₅ and cognitive (n=4,460) or MRI (n=1,427-1,441) outcomes. Small, but statistically significant effect estimates were found for a few exposure method-outcome associations, but these associations were null when using PM₂.₅ estimates from most other methods. Conclusion: Exposure estimation approaches agreed more across ARIC sites than within each site. Variability between exposure estimation methods for PM₂.₅ within sites likely contributed to heterogeneity in estimated associations with cognitive outcomes. Choice of estimation methods may influence estimated exposure and findings in epidemiological studies when participants are concentrated in small geographic areas.
Source: ISEE Conference Abstracts
Language: English
Subfield: Health, Toxicology and Mutagenesis
Field: Environmental Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4317178482