Exposome-wide association study of cognition among older adults in the National Health and Nutrition Examination Survey

Lauren M Middleton,Erika Walker,Scarlet Cockell,John Dou,Vy K Nguyen,Mitchell Schrank,Chirag J Patel,Erin J Ware,Justin A Colacino,Sung Kyun Park,Kelly M Bakulski

DOI: https://doi.org/10.1101/2024.07.19.24310725

2024-07-21

Abstract:Cognitive impairment among older adults is a growing public health challenge and environmental chemicals may be modifiable risk factors. A wide array of chemicals has not yet been tested for association with cognition in an environment-wide association framework. In the US National Health and Nutrition Examination Survey (NHANES) 1999-2000 and 2011-2014 cross-sectional cycles, cognition was assessed using the Digit Symbol Substitution Test (DSST, scores 0-117) among participants aged 60 years and older. Concentrations of environmental chemicals measured in blood or urine were log2 transformed and standardized. Chemicals with at least 50% of measures above the lower limit of detection were included (nchemicals=147, nclasses=14). We tested for associations between chemical concentrations and cognition using parallel survey-weighted multivariable linear regression models adjusted for age, sex, race/ethnicity, education, smoking status, fish consumption, cycle year, urinary creatinine, and cotinine. Participants with at least one chemical measurement (n=4,982) were mean age 69.8 years, 55.0% female, 78.2% non-Hispanic White, and 77.0% at least high school educated. The mean DSST score was 50.4 (standard deviation (SD)=17.4). In adjusted analyses, 5 of 147 exposures were associated with DSST at p-value<0.01. Notably, a SD increase in log2-scaled cotinine concentration was associated with 2.71 points lower DSST score (95% CI -3.69, -1.73). A SD increase in log2-scaled urinary tungsten concentration was associated with 1.34 points lower DSST score (95% CI -2.11, -0.56). Exposure to environmental chemicals, particularly heavy metals and tobacco smoke, may be modifiable factors for cognition among older adults.

Epidemiology

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the relationship between environmental chemical exposure and cognitive function among the elderly population in the United States. Specifically, researchers systematically explored the associations between the exposure levels of 147 different chemicals and the cognitive function of the elderly through the method of environmental - wide association study (ExWAS), using data from the National Health and Nutrition Examination Survey (NHANES). This study aims to identify environmental chemicals that may affect the cognitive health of the elderly, especially those that may be neurotoxic, thereby providing new intervention targets for the prevention of cognitive impairment and dementia.

The main findings of the study include:
- High concentrations of serum cotinine and urinary tungsten are associated with lower cognitive scores.
- Increases in the concentrations of urinary benzophenone - 3, blood m - /p - xylene, and urinary 2 - methylhippuric acid are associated with higher cognitive scores.
- The study also found that mono - benzyl phthalate is associated with an increased risk of cognitive impairment, while the combination of perfluorooctanoic acid (PFOA) and 3 - methylhippuric acid/4 - methylhippuric acid is associated with a lower risk of cognitive impairment.

These findings not only enhance our understanding of how specific chemicals affect the cognitive function of the elderly, but also provide new hypotheses for future research, especially in exploring the effects of these chemicals on neurodegenerative diseases.