Next Article in Journal
Association Between Blood Benzene Levels and Periodontal Disease in a Nationally Representative Adult U.S. Population
Previous Article in Journal
Systematic Evaluation of How Indicators of Inequity and Disadvantage Are Measured and Reported in Population Health Evidence Syntheses
Previous Article in Special Issue
Volatile Organic Compound (VOC) Contamination in Hotel Rooms: A Pilot Study to Understand Sources and Health Risks
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJERPH
Volume 22
Issue 6
10.3390/ijerph22060852
ijerph-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Determining Associations Between Levels of Ethylene Oxide Gas Exposure and Neurocognitive Performance for Older U.S. Adults

by
Linda O’Kelley
1,*,
Barbara Swanson
1,
Jessica Bishop-Royse
1,
Joyce W. Tam
2,
Christopher Forsyth
3 and
Susan Buchanan
4
1
College of Nursing, Rush University, Chicago, IL 60612, USA
2
Department of Psychiatry & Behavioral Sciences, Rush Medical College, Rush University, Chicago, IL 60612, USA
3
Department of Internal Medicine, Rush Medical College, Rush University, Chicago, IL 60612, USA
4
School of Public Health, Environmental & Occupational Health Sciences, University of Illinois Chicago, Chicago, IL 60612, USA
*
Author to whom correspondence should be addressed.
Int. J. Environ. Res. Public Health 2025, 22(6), 852; https://doi.org/10.3390/ijerph22060852
Submission received: 8 April 2025 / Revised: 16 May 2025 / Accepted: 23 May 2025 / Published: 29 May 2025
(This article belongs to the Special Issue Environmental and Occupational Health: Exposure to Volatile Organic Compounds (VOC) and Gases)
Versions Notes

Abstract

:
Ethylene oxide (EtO) gas is a widely used industrial chemical and known health hazard. Multiple studies have determined that EtO exposure can be measured via hemoglobin adduct levels, and EtO exposure increases the risk of cancer and neurocognitive deficits, especially with occupational exposure. Emerging studies indicate that neighboring communities are also at risk. The purpose of this study is to explore the relationship of known covariates and EtO hemoglobin adduct levels to neurocognitive performance in older U.S. adults. This exploratory study drew its sample from the publicly available NHANES dataset. The 2013–2014 NHANES measured EtO exposure via hemoglobin adducts and the cognitive domain of neurocognitive function using the CERAD, Animal Fluency, and Digit Symbol Substitution (DSST) tests. Motor function was measured using grip strength. Participants were grouped into background (≤27.36 pmol/gHb) or elevated (>27.36 pmol/gHb) EtO exposure. Hierarchical linear regression, independent t-tests, and logistic regression analyses were performed. A total of 10,175 individuals were sampled: 489 were included in the cognitive analyses, and 436 were included in the motor analyss. Elevated EtO adduct levels significantly predicted low Animal Fluency, DSST, CERAD, and combined grip strength scores. Our findings are supported by the extant literature citing neurotoxic EtO exposure effects. Further study in known EtO-exposed communities is warranted.

1. Introduction

Since ethylene oxide (EtO) gas was introduced in the 1950s for industrial purposes, it has been linked to serious adverse health outcomes, including cancer and neurotoxic effects [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23]. Multiple case series and case control studies have found that chronic occupational EtO exposure is associated with impaired cognition, profound forgetting, slower processing speed, sensory loss in the extremities, and muscle weakness [5,6,7,8,13,17,23]. In response, several countries have banned EtO gas use in food production [24].
Most published toxicology studies have focused on exposure in hospital or occupational settings [10,11,12,14,15,16,18]. However, recent data provided by the U.S. Environmental Protection Agency (EPA) indicate the need to look beyond occupational- and hospital-level exposures to also examine chronic, low-intensity exposures, as seen in communities residing near EtO-emitting facilities [19,20]. The few recent studies that have examined the effects of residential EtO exposures report significantly increased risks of hypertension, diabetes, depression, and cancer [3,4,9,22,25]. It remains unknown if the same neurologic symptoms found in occupationally exposed persons are found in persons with residential exposure.
To explicate the association of EtO gas and adverse health outcomes, several studies have identified valid biomarkers of exposure or biological signs that EtO has been absorbed by the body. As EtO gas is rapidly metabolized and has a short half-life [26], EtO biomarkers are more reliable indicators for long-term exposure than EtO blood levels alone [27]. One biomarker of human exposure to toxins is an adduct, a reaction product formed by the conjugation of the toxin with a human macromolecule [28,29]. Most existing EtO biomarkers involve the conjugation of human DNA, hemoglobin, or other proteins with EtO once the gas is metabolized. Researchers have identified biomarkers specific for EtO gas, which include the hemoglobin adduct N-2-hydroxyethylvaline (HEV), the DNA adduct 7-HEG, and the urinary metabolite HEMA [30,31,32].
The EtO hemoglobin adduct is currently being measured in U.S. residents enrolled in the continuous National Health and Nutrition Examination Survey (NHANES) facilitated by the Centers for Disease Control and Prevention (CDC). In the 2013–2014 cycle of the NHANES, the cognitive and motor domains of neurocognitive function were also concurrently measured for a subsample of participants and included the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) Word Learning subtest, Digit Symbol Substitution test, Animal Fluency test, and grip strength test. These tests measure cognitive and motor domains known to be affected by environmental neurotoxins, such as lead, cadmium, manganese, mercury, and arsenic [33,34,35,36,37,38,39]. Multiple studies have found these tests to be sensitive for detecting the cognitive and/or motor effects of environmental contaminants in large, exposed populations [33,34,35,36,37,38,39]; however, to our knowledge, there have been no studies examining the link between levels of EtO hemoglobin adducts and cognitive/motor function in exposed populations.
Thus, the purpose of this study is to explore the relationship of EtO hemoglobin adduct levels, known predictors of neurocognitive function, and smoking status to cognitive and motor performance in a sample of older U.S. adults using the NHANES database. As there is limited research on the effects of EtO gas at the community level, this novel study will expand the current understanding of the risk posed by community-level EtO gas exposure.

2. Materials and Methods

  • Study Design
This project is a secondary analysis of the 2013–2014 NHANES dataset, conducted by the National Center for Health Statistics (NCHS), a division of the CDC. The NHANES study is a large, continuous, and publicly available dataset that documents statistics for a wide variety of health- and nutrition-related measures for a sample of approximately 10,000 participants per two-year cycle.
  • Sample and Setting
Using purposive sampling, NHANES recruits a sample that is representative of the U.S. population in terms of age, biological sex, race and ethnicity, and income within each cycle [40,41]. NHANES chooses 30 counties per two-year cycle and randomly selected neighborhoods and households within the chosen counties using U.S. Census Bureau information [42]. If consenting participants meet the eligibility requirements, they undergo an interview and physical examination, including the collection of blood and urine specimens [40]. The CDC anonymized this dataset and removed protected health information to ensure confidentiality for study participants, including zip codes and/or other location data [43,44].
The 2013–2014 cycle is the only known cycle published to date that includes neurocognitive and EtO hemoglobin adduct data. The neurocognitive battery includes (a) three cognitive tests administered to participants aged 60 years and older and (b) a grip strength motor test administered to participants 6 years of age and older. Of the approximately 10,000 NHANES participants aged 6 years and older, the EtO hemoglobin adduct was quantified in a randomly selected subsample of 1/3 of the participants (N ≈ 300) [45].
  • Inclusion and Exclusion Criteria
The inclusion criteria for the NHANES dataset included noninstitutionalized, civilian U.S. residents [40]. The exclusion criteria consisted of the following: (a) non-U.S. citizens, (b) U.S. citizens residing outside of the U.S., (c) all active-duty military personnel, (d) hospitalized persons, and (e) incarcerated persons [40]. In this study, only those participants who had values for EtO hemoglobin adduct levels and who completed all the cognitive and motor tests were included in the final analyses. Cognitive function measures were only administered to participants aged 60 and over. Accordingly, our sample consisted of persons ≥ 60 years of age, and we excluded grip strength data collected from participants younger than age 60.
  • Measures
  • Demographic Variables and Smoking Status
Demographic data and biomarkers of cigarette smoke exposure were collected via interviews or blood draws as part of the NHANES study. The collected demographics included age, biological sex, income, education level, race/ethnicity, and language spoken. The reference categories for the categorical variables included the following: biological sex: female; education level: college graduate; race/ethnicity: non-Hispanic White; and language spoken: English. The sample in the current study was limited to participants aged 60 and over. Participants aged 80 and above were top coded into the 80+ group per NHANES procedure. Cigarette smoke exposure was included in the analyses because smokers have higher levels of EtO biomarkers in their blood [46]. Serum cotinine levels were used to determine cigarette smoke exposure for both active smokers and those exposed to secondhand smoke, with higher levels indicating higher levels of exposure [47]. Serum cotinine levels are widely used, over nicotine alone, as a biomarker of cigarette smoke exposure due to cotinine’s longer half-life and, therefore, more accurate indication of long-term cigarette smoke exposure [47].
  • Cognitive and Grip Strength Measures
The three cognitive assessments administered during the 2013–2014 cycle of NHANES were included in the analyses: the CERAD Word List Learning test (CERAD W-L), Animal Fluency test, and Digit Symbol Substitution test (DSST). These assessments are commonly used to evaluate cognitive function, including memory (CERAD W-L), semantic fluency (Animal Fluency), and processing speed (DSST) [48,49,50]. The CERAD W-L measure contains 3 immediate learning trials and a delayed recall trial, which yield a total learning score (i.e., the sum of the raw scores across the three learning trials) and a delayed recall score. For all three cognitive assessments, higher scores indicate better cognitive performance. All the cognitive tests were administered privately in mobile examination centers by trained interviewers [51].
Under the umbrella of neurocognitive function, both cognitive and motor domains were assessed within the 2013–2014 cycle of NHANES. Similar to cognitive function, motor function is related to brain processing, with some studies indicating that motor dysfunction can precede cognitive dysfunction in the context of dementia [52]. Grip strength, which measures the muscle strength of the hands, represents the motor domain of neurocognitive function. Combined grip strength was calculated as the sum of the largest dynamometer reading from three trials in each hand and measured in kilograms (kg) [53].
  • Ethylene Oxide Blood Levels
The gold standard for quantifying EtO exposure is the hemoglobin adduct, HEV, level in human whole blood [1,30,32,42]. To quantify HEV levels, peripheral venous blood specimens were collected from randomly selected, eligible NHANES participants in mobile examination centers. To ensure consistency among the collection sites, all specimens were handled following a detailed protocol concerning specimen collection, processing, storage, and shipping [42,44]. At the CDC laboratory, a modified Edman reaction was performed to isolate the EtO hemoglobin adduct, followed by high-performance liquid chromatography and tandem mass spectrometry. Analyses were primarily conducted in singlets; however, approximately 2% of the sample was randomly selected to be analyzed in duplicate for a quality analysis. EtO hemoglobin adduct levels are expressed as pmol/g Hb.
The participants were categorized into two groups based on their EtO hemoglobin adduct concentrations: background (≤27.36 pmol/g Hb) or elevated (>27.36 pmol/g Hb). This conversion from a continuous-level to a binary variable was necessary as the highly positive skew violated assumptions for the planned analyses. Log and square root transformation were attempted but unsuccessful. As there are no defined background or normal level cut-offs for EtO hemoglobin adduct levels, these categories were created based on the positively skewed distribution of the EtO hemoglobin adduct data observed in our sample. The cutoff of 27.36 pmol/g Hb was selected as 73.6% of the participants showed hemoglobin adduct levels between the narrow range of 5–27.36 pmol/g Hb. A wider distribution from 27.37 pmol/g Hb to 989.35 pmol/g Hb represents the participants with elevated levels.
  • Procedures
Rush University IRB approval was obtained. Data from the 2013–2014 NHANES cycle were accessed and downloaded from the publicly available website and imported into SPSS.
  • Data Analysis
All statistical analyses were conducted using IBM SPSS (v. 30), using sample weights, as per NHANES guidelines. To preserve an appropriate sample size given the large number of missing variables in the NHANES database, missing data were excluded pairwise. This approach allowed cases with missing data on variables that were irrelevant to the study purpose to remain in the analysis when variables of interest had non-missing values.
The analysis was conducted using corrected laboratory values of EtO hemoglobin adduct levels that were released by the CDC in May of 2023. This correction was necessary due to a discovered calibration error in commercially purchased equipment at the CDC laboratory that led to a systematic bias in original results. The data were corrected via log transformation using the following formula: LBXEOA = 0.58026 (LBXEOA_original 1.03562). The original results were withdrawn from the NHANES website in March 2022 and were not available for analysis [45].
  • Variable Selection Analysis
To evaluate the impact of EtO exposure on cognitive and motor performance, variable selection analyses were undertaken to determine which demographic characteristics significantly related to scores on the neurocognitive measures and to other potential predictors. A larger sample size of 1392 to 1573 participants in the 2013–2014 NHANES (without the inclusion criterion of published EtO hemoglobin adduct levels) was used to verify covariates to increase the power of the analyses. Multivariable linear regression models examined associations between the available predictors and performance on each of the cognitive (CERAD-WL learning trials, CERAD delayed recall, Animal Fluency, and Digit Symbol Substitution) and motor (combined grip strength) tests.
  • Hierarchical Linear Regression
To explore the relationship between the binary EtO hemoglobin adduct levels and neurocognitive function, while controlling for the identified covariates, a Pearson correlation coefficient and R2 value were calculated using hierarchical linear regression models for each neurocognitive measure. Variables were added in blocks, with relevant covariates, including age, income, smoking status, gender, education level, race/ethnicity, and language spoken, being included in the first step of the model. Relevant covariates and EtO adduct levels were included in the second step. This process allowed for the relationship between EtO levels and neurocognitive function to be explored, while controlling for the effects of relevant demographics.
The assumptions of regression models were evaluated prior to analysis [54]. We found little evidence for violations of these assumptions, with the exception of outliers and high leverage points observed in some models. Outliers were subsequently removed to satisfy and facilitate the analysis. Observations representing high leverage points were retained as their removal would have resulted in participants with the “other” race category being dropped from the analysis.
  • Independent samples t-test
To compare the cognitive function of persons with background and elevated EtO hemoglobin adduct levels, an independent samples t-test was performed. A t-test statistic and a significance value were calculated to determine if significant group differences were present for background vs. elevated EtO adduct levels for each of the 5 neurocognitive assessments.
All relevant assumptions were tested prior to analysis and met, with the exception of outliers in each model. Outliers were subsequently removed to satisfy assumptions and facilitate the analysis.
  • Logistic regression analysis
To compare the odds of having low cognitive scores for background vs. elevated EtO hemoglobin adduct levels, an odds ratio, confidence interval, and significance value were calculated using a hierarchical logistic regression for each cognitive test. All cognitive data were adjusted by age based on manual or published normative data [55,56,57]. For the CERAD test, performance was further adjusted by education level (less than a high school degree or a high school degree and beyond) [57]. Z-scores were calculated to identify cognitive impairment, using a binary variable (present/absent). Impaired cognitive function was defined as scoring 1.5 SD below the normative mean, a conventional criterion for impaired cognitive performance [58,59,60]. All the relevant assumptions were tested prior to analysis and met.

3. Results

The number of participants who met the eligibility criteria and completed each of the neurocognitive tests ranged from 436 (motor) to 489 (cognitive) (Table 1). The mean age of the participants was 68.98 years, and they were equally distributed by biological sex. There were notably larger groups of non-Hispanic White individuals, English speakers, and those with some college or above education among those NHANES participants who completed the five neurocognitive assessments (Table 1). The mean EtO hemoglobin adduct level was 37.17 pmol/g Hb, with 333 of the participants classified as having background levels and 156 of the participants classified as having elevated levels. We found that the NHANES participants with elevated EtO hemoglobin adduct levels were significantly younger, with higher levels of cigarette smoke exposure, lower income, more diverse, and had lower education levels compared to the participants with background levels (Table 2). The participants’ performance on each of the neurocognitive outcome measures is presented in Table 3.

3.1. Variable Selection Analyses

Age, income, biological sex, race/ethnicity, education, and language spoken were significantly associated with each cognitive measure and were, therefore, included as covariates in the final analyses for cognitive performance. Except for language spoken, the same factors were significantly correlated with grip strength and were included as covariates. While smoking status did not correlate with neurocognitive performance, it was included in the final analyses as a covariate due to the known association between higher serum cotinine levels and higher levels of EtO adducts in the blood [46].

3.2. Relationship Between EtO Levels and Neurocognitive Performance

Multiple hierarchical linear regression analyses were performed to investigate if elevated EtO hemoglobin adduct levels were significantly related to neurocognitive performance when controlling for available covariates. The full model of neurocognitive performance (age, smoking status, income, biological sex, race/ethnicity, education, language spoken) and elevated EtO hemoglobin adduct levels were calculated for each of the five neurocognitive measures. For the CERAD delay recall, the participants who were older; of the male biological sex; of Hispanic, Black, or other race descent; and had elevated EtO hemoglobin adduct levels performed more poorly (Table 4, Model 2b). Elevated EtO hemoglobin adduct levels accounted for approximately 2% of the variation in scores, F(1,15) = 9.31, p = 0.008. A significant inverse relationship between elevated EtO hemoglobin adduct levels and motor performance, as measured by combined grip strength, was also found in the full model (Table 5). Elevated EtO hemoglobin adduct levels accounted for less than 1% of the variation in scores, F(1,15) = 8.84, p = 0.009. Models for the CERAD-WL learning trials, Animal Fluency test, and DSST were non-significant (Table 4, Models 1, 3, & 4).

3.3. Group Differences Between Background and Elevated EtO Adduct Levels

Independent t-tests were performed to determine if significant differences existed between the background and elevated EtO groups’ neurocognitive mean scores. The participants with elevated EtO hemoglobin adduct levels performed significantly worse than those with background levels on the CERAD delayed recall, t(15) = 3.53, p = 0.003, Animal Fluency, t(15) = 2.67, p = 0.018, and Digit Symbol Substitution tests, t(15) = 2.24, p = 0.041. No significant group differences were found for the CERAD-WL learning trial sum or the combined grip strength test (Table 6).

3.4. Predictors of Low Cognitive Scores

A logistic regression analysis was performed to assess the association between elevated EtO hemoglobin adduct levels and the likelihood of obtaining a low score on the neurocognitive tests. After adjusting for age and education, elevated EtO hemoglobin adduct levels significantly predicted low cognitive scores (defined as scoring ≥ 1.5 SD below the normative mean) for the CERAD-WL trial #1, χ2(1) = 7.95, p = 0.005, and CERAD delayed recall test, χ2(1) = 4.93, p = 0.026, as compared to background levels. Elevated EtO hemoglobin adduct levels were found to increase the odds of scoring low on the CERAD-WL trial #1 by 2.09× and on the CERAD delayed recall test by 1.76× (Table 7). After adjusting for age, elevated EtO hemoglobin adduct levels significantly predicted low cognitive scores for the Animal Fluency test, χ2(1) = 9.98, p = 0.002, increasing the odds of scoring low on the Animal Fluency test by 1.79× (Table 7). As the available normative data for the Animal Fluency test accounted for age only, additional analyses were performed to account for education. After accounting for age and education (< or ≥12 years of education), the Animal Fluency test was not found to be significant. Models for the CERAD-WL trial 3 and the DSST were not significant (Table 7).
We completed additional analyses to understand the demographics of the participants in our sample who scored in the impaired cognitive function range. We compared the age, education level, and race/ethnicities of the participants who received a higher number of impaired cognitive scores (3–5) to those who received no or few impaired cognitive scores (0–2). A χ2 analysis was performed with all assumptions met. We found significant associations between age (p = 0.014) and race/ethnicity (p = 0.007) and the number of low cognitive scores received, indicating that those who were older and with diverse race/ethnicities had more scores in the impaired cognitive function range. Education differences were non-significant.

4. Discussion

To our knowledge, this study was the first to evaluate associations between EtO gas exposure and cognitive and motor performance in a representative population of older U.S. adults. We found significant inverse associations between elevated EtO hemoglobin adduct levels and neurocognitive performance on the CERAD-WL learning trial, CERAD delayed recall, Animal Fluency, DSST, and combined grip strength tests. Our results suggest that elevated levels of EtO hemoglobin adducts are associated with poorer performance in memory, semantic fluency, processing speed, and muscle strength.
We also found that the participants with elevated EtO hemoglobin adduct levels had significantly increased odds of low cognitive scores when compared to normative samples, adjusting for age and education when possible. This is significant as those participants with elevated EtO hemoglobin adduct levels not only scored lower than those with background levels, but their scores were also lower compared to cognitively healthy samples of individuals.
It was hypothesized that the participants with elevated EtO hemoglobin adduct levels were more likely to have low cognitive scores due to the combination of (a) the neurotoxic effects of EtO gas exposure and (b) the higher representation of socially and economically disadvantaged groups who generally face a higher risk of co-occurring environmental hazards [61,62]. In support of the latter, we found significant differences in race/ethnicities between the background and elevated EtO hemoglobin adduct groups. These findings highlight the need for further studies of residential EtO gas exposure to address both the significant neurocognitive deficits and social justice concerns related to exposure.
The current findings were consistent with the existing literature indicating the neurotoxic effects of EtO exposure. Over several decades of investigation, multiple researchers have published case reports and case series documenting cognitive dysfunction and physical symptoms in occupationally exposed persons [5,6,7,8,13,17,23,63,64,65,66,67,68,69]. Reported symptoms range from headaches and neuropathy to memory loss and poorer psychomotor speed [5,63,70]. These reported effects comprise both cognitive and motor functioning, suggesting an association between elevated EtO adduct levels and dysfunction across multiple neurocognitive domains. Because our study only analyzed four cognitive measures, we were not able to replicate these findings in our sample of community dwelling persons with EtO exposure from unknown sources.
While statistically significant cognitive and motor differences were found between the participants with elevated vs. background EtO hemoglobin adduct levels, the associations were relatively weak, with small effect sizes after accounting for covariates. Combined with the absence of measures of functional outcomes, further inferences regarding clinical relevance cannot be made. The neurocognitive dysfunction described in the extant literature is related to occupational exposure, where EtO gas levels are reported to be orders of magnitude higher than found with residential exposure to the ambient air. Additionally, persons with a lower SES are more likely to be exposed to high levels of multiple co-occurring environmental hazards and chemicals, possibly confounding the explication of effects specific to EtO gas exposure [61,62]. Nevertheless, this exploratory analysis is an important first step to better understand the relationship between EtO gas exposure and neurocognitive sequelae.
Further examination of community-dwelling populations in known areas of EtO exposure is warranted. More targeted studies in a wider age range are recommended as the literature suggests that children and pregnant women are highly susceptible to the effects of environmental exposures, particularly air toxins [71]. In addition to the CERAD, Animal Fluency, DSST, and combined grip strength tests, a more comprehensive test battery should be considered. Recommended tests include those that measure neurocognitive domains of visual memory, visuospatial skills, other aspects of language (e.g., phonemic fluency and naming), and executive functioning, as these domains were not assessed in the 2013–2014 cycle of NHANES.

Limitations

While this study may be the first to examine neurocognitive EtO exposure effects in the U.S. population, multiple limitations arise from working exclusively with NHANES data. First, the study sample was limited to noninstitutionalized, older U.S. adults, and the results cannot be generalized to the general U.S. population. Second, we were unable to determine if the participants had occupational exposure to EtO gas. Additionally, it is unknown if the participants sampled from the 2013–2014 cycle of NHANES resided near an EtO-emitting facility. As NHANES demographic data are restricted, we were unable to access geographic data at the zip code level to determine the EtO exposure source. Third, NHANES participants represent a sample with an unknown cognitive status. While NHANES restricts eligibility to noninstitutionalized persons, some participants completing the cognitive tests may have had some degree of existing cognitive impairment.
Additionally, there are analytical limitations that should be noted. First, normative data for each neurocognitive measure were not co-normed. While all three neurocognitive normative test scores accounted for age, only the CERAD scores also accounted for education and biological sex. Additionally, the normative scores were typically several decades old and likely calculated from homogeneous populations, thus raising questions about how accurately they represent the true normative scores in the U.S. population [72]. Second, no normative scores were found by the authors for grip strength, excluding this test from the normative analyses. Third, while most assumptions were met or corrected by removing outliers, cases with high leverage points remained in the final hierarchical linear regression analyses to retain individuals identifying as other races in the analytical model.

5. Conclusions

We found significant inverse associations between EtO hemoglobin adduct levels and both cognitive and motor domains of neurocognitive performance in older U.S. adults. In some cases, these findings are above and beyond the significant influences of known covariates including age, race, income, biological sex, smoking status, education level, and language spoken. These findings are supported by a large body of extant literature on the neurotoxic effects of EtO gas exposure, but further research is still needed. Additional studies using a sample representative of the U.S. population in terms of age and in known areas of EtO gas exposure are recommended.

Author Contributions

L.O., B.S., J.B.-R., and J.W.T. conceived of and designed the study. L.O. extracted the data, conducted the analysis, created the tables, and wrote the manuscript. L.O., B.S., J.B.-R., and J.W.T. developed the methodology and interpreted the results. L.O., B.S., J.B.-R., J.W.T., C.F., and S.B. reviewed the manuscript and provided substantial feedback. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

This study was reviewed, acknowledged to be non-human subject research, and exempted from human subject research approval, as determined by the Office of Research Affairs, Rush University.

Data Availability Statement

The data analyzed in this study can be found within the publicly available NHANES database, https://www.cdc.gov/nchs/nhanes/, accessed on 5 June 2024.

Acknowledgments

The responsibility for the content remains exclusively with the authors.

Conflicts of Interest

The authors declare they have no conflicts of interest related to this work to disclose.

Abbreviations

The following abbreviations are used in this manuscript:
EtOEthylene Oxide Gas
CERADConsortium to Establish a Registry for Alzheimer’s Disease
NHANESNational Health and Nutrition Examination Survey
DSSTDigit Symbol Substitution test

References

  1. Agency for Toxic Substances and Disease Registry (ATSDR). Evaluation of Potential Health Impacts from Ethylene Oxide Emissions. U.S. Department of Health and Human Services. 2018, August 2. Available online: https://www.atsdr.cdc.gov/HAC/pha/sterigenic/Sterigenics_International_Inc-508.pdf (accessed on 20 January 2023).
  2. Bisanti, L.; Maggini, M.; Raschetti, R.; Spila Alegiani, S.; Menniti Ippolito, F.; Caffari, B.; Segnan, N.; Ponti, A. Cancer mortality in ethylene oxide workers. Br. J. Ind. Med. 1993, 50, 317–324. [Google Scholar] [CrossRef] [PubMed]
  3. Bulka, C.; Nastoupil, L.J.; Koff, J.L.; Bernal-Mizrachi, L.; Ward, K.C.; Williams, J.N.; Bayakly, A.R.; Switchenko, J.M.; Waller, L.A.; Flowers, C.R. Relations between residential proximity to EPA-designated toxic release sites and diffuse large b-cell lymphoma incidence. South. Med. J. 2016, 109, 606–614. [Google Scholar] [CrossRef]
  4. Chen, X. A temporal analysis of the association between breast cancer and socioeconomic and environmental factors. GeoJournal 2018, 83, 1239–1256. [Google Scholar] [CrossRef]
  5. Crystal, H.A.; Schaumburg, H.H.; Grober, E.; Fuld, P.A.; Lipton, R.B. Cognitive impairment and sensory loss associated with chronic low-level ethylene oxide exposure. Neurology 1988, 38, 567–569. [Google Scholar] [CrossRef] [PubMed]
  6. Estrin, W.J.; Cavalieri, S.A.; Wald, P.; Becker, C.E.; Jones, J.R.; Cone, J.E. Evidence of neurologic dysfunction related to long-term ethylene oxide exposure. Arch. Neurol. 1987, 44, 1283–1286. [Google Scholar] [CrossRef]
  7. Feitosa, A.; Fustes, O.H.; Assunção, C.M.; Lurkiv, R.A.; Hümmelgen, E.; Hernandez-cossio, O.; Taques, C. Occupational exposure to ethylene oxide as a cause of polyneuropathy. J. Neurol. Sci. 2017, 381, 498. [Google Scholar] [CrossRef]
  8. Finelli, P.F.; Morgan, T.F.; Yaar, I.; Granger, C.V. Ethylene oxide- induced polyneuropathy: A clinical and electrophysiologic study. Arch. Neurol. 1983, 40, 419–421. [Google Scholar] [CrossRef]
  9. Guo, J.; Wan, Z.; Cui, G.; Pan, A.; Liu, G. Association of exposure to ethylene oxide with risk of diabetes mellitus: Results from NHANES 2013–2016. Environ. Sci. Pollut. Res. 2021, 28, 68551–68559. [Google Scholar] [CrossRef]
  10. Hagmar, L.; Mikoczy, Z.; Welinder, H. Cancer incidence in Swedish sterilant workers exposed to ethylene oxide. Occup. Environ. Med. 1995, 52, 154–156. [Google Scholar] [CrossRef]
  11. Hogstedt, C.; Rohlén, O.; Berndtsson, B.S.; Axelson, O.; Ehrenberg, L. A cohort study of mortality and cancer incidence in ethylene oxide production workers. Br. J. Ind. Med. 1979, 36, 276–280. [Google Scholar] [CrossRef]
  12. Kiran, S.; Cocco, P.; Mannetje, A.; Satta, G.; D’Andrea, I.; Becker, N.; de Sanjosé, S.; Foretova, L.; Staines, A.; Kleefeld, S.; et al. Occupational exposure to ethylene oxide and risk of lymphoma. Epidemiology 2010, 21, 905–910. [Google Scholar] [CrossRef] [PubMed]
  13. Kuzuhara, S.; Kanazawa, I.; Nakanishi, T.; Egashira, T. Ethylene oxide polyneuropathy. Neurology 1983, 33, 377. [Google Scholar] [CrossRef]
  14. Mikoczy, Z.; Tinnerberg, H.; Björk, J.; Albin, M. Cancer incidence and mortality in Swedish sterilant workers exposed to ethylene oxide: Updated cohort study findings 1972–2006. Int. J. Environ. Res. Public Health 2011, 8, 2009–2019. [Google Scholar] [CrossRef]
  15. Norman, S.A.; Berlin, J.A.; Soper, K.A.; Middendorf, B.F.; Stolley, P.D. Cancer incidence in a group of workers potentially exposed to ethylene oxide. Int. J. Epidemiol. 1995, 24, 276–284. [Google Scholar] [CrossRef] [PubMed]
  16. Park, R.M. Associations between exposure to ethylene oxide, job termination, and cause—specific mortality risk. Am. J. Ind. Med. 2020, 63, 577–588. [Google Scholar] [CrossRef]
  17. Patch, P.C.; Hartlage, L.C. Neurological and emotional sequelae of exposure to ethylene oxide. Int. J. Neurosci. 2001, 106, 101–107. [Google Scholar] [CrossRef] [PubMed]
  18. Steenland, K.; Stayner, L.; Deddens, J. Mortality analyses in a cohort of 18 235 ethylene oxide exposed workers: Follow up extended from 1987 to 1998. Occup. Environ. Med. 2004, 61, 2–7. [Google Scholar]
  19. United States Environmental Protection Agency (U.S. EPA). Evaluation of the Inhalation Carcinogenicity of Ethylene Oxide (Final Report); EPA/635/R-16/350F; U.S. Environmental Protection Agency: Washington, DC, USA, 2016. Available online: https://cfpub.epa.gov/ncea/iris/iris_documents/documents/toxreviews/1025tr.pdf (accessed on 20 January 2023).
  20. United States Environmental Protection Agency (U.S. EPA). Hazardous Air Pollutants: Ethylene Oxide. Our Current Understanding of the Human Health and Environmental Health Risks of Ethylene Oxide. 8 December 2022. Available online: https://www.epa.gov/hazardous-air-pollutants-ethylene-oxide/our-current-understanding-human-health-and-environmental (accessed on 20 January 2023).
  21. World Health Organization (WHO). International Programme on Chemical Safety, United Nations Environment Programme, International Labour Organization, World Health Organization & WHO Task Force on Ethylene Oxide: Ethylene Oxide. 1985. Available online: https://apps.who.int/iris/handle/10665/38815 (accessed on 9 June 2023).
  22. Wu, N.; Cao, W.; Wang, Y.; Liu, X. Association between blood ethylene oxide levels and the prevalence of hypertension. Environ. Sci. Pollut. Res. 2022, 29, 76937–76943. [Google Scholar] [CrossRef]
  23. Zampollo, A.; Zacchetti, O.; Pisati, G. On ethylene oxide neurotoxicity: Report of two cases of peripheral neuropathy. Ital. J. Neurol. Sci. 1984, 5, 59–62. [Google Scholar] [CrossRef]
  24. Dudkiewicz, A.; Dutta, P.; Kołożyn-Krajewska, D. Ethylene oxide in foods: Current approach to the risk assessment and practical considerations based on the European food business operator perspective. Eur. Food Res. Technol. 2022, 248, 1951–1958. [Google Scholar] [CrossRef]
  25. Wang, H.; Chen, X.; Lin, F.; Zheng, J.; Chen, K.; Wang, X.; Lin, X.; Chen, X.; Ye, Q.; Wang, Y.; et al. Association between ethylene oxide levels and depressive symptoms: A cross-sectional study based on NHANES 2013–2018 database. J. Affect. Disord. 2024, 348, 135–142. [Google Scholar] [CrossRef] [PubMed]
  26. Filser, J.G.; Denk, B.; Törnqvist, M.; Kessler, W.; Ehrenberg, L. Pharmacokinetics of ethylene in man; body burden with ethylene oxide and hydroxyethylation of hemoglobin due to endogenous and environmental ethylene. Arch. Toxicol. 1992, 66, 157–163. [Google Scholar] [CrossRef] [PubMed]
  27. Ogawa, M.; Oyama, T.; Isse, T.; Yamaguchi, T.; Murakami, T.; Endo, Y.; Kawamoto, T. Hemoglobin adducts as a marker of exposure to chemical substances, especially PRTR class I designated chemical substances. J. Occup. Health 2006, 48, 314–328. [Google Scholar] [CrossRef]
  28. Ehrenberg, L.; Granath, F.; Törnqvist, M. Macromolecule adducts as biomarkers of exposure to environmental mutagens in human populations. Environ. Health Perspect. 1996, 104 (Suppl. 3), 423–428. [Google Scholar] [PubMed]
  29. Neumann, H.G.; Albrecht, O.; Van Dorp, C.; Zwirner-Baier, I. Macromolecular adducts caused by environmental chemicals. Clin. Chem. 1995, 41, 1835–1840. [Google Scholar] [CrossRef]
  30. Bader, M.; Will, W.; Frey, G.; Nasterlack, M. Analysis of protein adducts as biomarkers of short-term exposure to ethylene oxide and results of follow-up biomonitoring. Arh. Za Hig. Rada I Toksikologiju 2012, 63, 107–115. [Google Scholar] [CrossRef] [PubMed]
  31. Mráz, J.; Hanzlíková, I.; Linhart, I.; Dušková, Š.; Dabrowská, L.; Hejl, K. N-(2-Hydroxyethyl)-l-valyl-l-leucine in rat urine as a hydrolytic cleavage product of ethylene oxide adduct with globin. Arch. Toxicol. 2019, 93, 603–613. [Google Scholar] [CrossRef]
  32. Wu, K.-Y.; Chiang, S.-Y.; Shih, W.-C.; Huang, C.-C.J.; Chen, M.-F.; Swenberg, J.A. The application of mass spectrometry in molecular dosimetry: Ethylene oxide as an example. Mass Spectrom. Rev. 2011, 30, 733–756. [Google Scholar] [CrossRef]
  33. Anger, W.K. Neurobehavioural tests and systems to assess neurotoxic exposures in the workplace and community. Occup. Environ. Med. 2003, 60, 531–538. [Google Scholar] [CrossRef]
  34. Bakulski, K.M.; Seo, Y.A.; Hickman, R.C.; Brandt, D.; Vadari, H.S.; Hu, H.; Park, S.K. Heavy metals exposure and Alzheimer’s disease and related dementias. J. Alzheimer’s Dis. JAD 2020, 76, 1215–1242. [Google Scholar] [CrossRef]
  35. Gbemavo, M.C.J.; Bouchard, M.F. Concentrations of lead, mercury, selenium, and manganese in blood and hand grip strength among adults living in the United States (NHANES 2011–2014). Toxics 2021, 9, 189. [Google Scholar] [CrossRef] [PubMed]
  36. O’Bryant, S.E.; Edwards, M.; Menon, C.V.; Gong, G.; Barber, R. Long-term low-level arsenic exposure is associated with poorer neuropsychological functioning: A Project FRONTIER study. Int. J. Environ. Res. Public Health 2011, 8, 861–874. [Google Scholar] [CrossRef] [PubMed]
  37. Przybyla, J.; Houseman, E.A.; Smit, E.; Kile, M.L. A path analysis of multiple neurotoxic chemicals and cognitive functioning in older US adults (NHANES 1999–2002). Environ. Health 2017, 16, 19. [Google Scholar] [CrossRef] [PubMed]
  38. Sasaki, N.; Carpenter, D.O. Associations between metal exposures and cognitive function in American older adults. Int. J. Environ. Res. Public Health 2022, 19, 2327. [Google Scholar] [CrossRef]
  39. Zoni, S.; Albini, E.; Lucchini, R. Neuropsychological testing for the assessment of manganese neurotoxicity: A review and a proposal. Am. J. Ind. Med. 2007, 50, 812–830. [Google Scholar] [CrossRef]
  40. Johnson, C.L.; Dohrmann, S.M.; Burt, V.L.; Mohadjer, L.K. National Health and Nutrition Examination Survey: Sample design, 2011–2014. In National Center for Health Statistics; Vital Health Stat 2014, 2; Centers for Disease Control and Prevention: Hyattsville, MD, USA, 2013. Available online: https://www.cdc.gov/nchs/data/series/sr_02/sr02_162.pdf (accessed on 30 June 2023).
  41. National Health and Nutrition Examination Survey (NHANES). 2013–2014 Data Documentation, Codebook, and Frequencies: Demographic Variables and Sample Weights (DEMO_H). 2015. Available online: https://wwwn.cdc.gov/Nchs/Data/Nhanes/Public/2013/DataFiles/DEMO_H.htm (accessed on 22 May 2025).
  42. Center for Disease Control and Prevention (CDC). National Center for Health Statistics: National Health and Nutrition Examination Survey: Why Was I Selected? 2021. Available online: https://www.cdc.gov/nchs/nhanes-participants/selection.html?CDC_AAref_Val=https://www.cdc.gov/nchs/nhanes/participant/participant-selected.htm (accessed on 22 May 2025).
  43. Center for Disease Control and Prevention (CDC). National Center for Health Statistics: National Health and Nutrition Examination Survey: NHANES 2013–2014 Laboratory Data Overview. n.d. Available online: https://wwwn.cdc.gov/nchs/nhanes/continuousnhanes/overviewlab.aspx?BeginYear=2013 (accessed on 2 July 2023).
  44. Center for Disease Control and Prevention (CDC). National Center for Health Statistics: National Health and Nutrition Examination Survey: NHANES 2013–2014 Overview. n.d. Available online: https://wwwn.cdc.gov/nchs/nhanes/continuousnhanes/overview.aspx?BeginYear=2013#print (accessed on 2 July 2023).
  45. National Health and Nutrition Examination Survey (NHANES). 2013–2014 Data Documentation, Codebook, and Frequencies: Ethylene Oxide (ETHOX_H). 2023. Available online: https://wwwn.cdc.gov/Nchs/Data/Nhanes/Public/2013/DataFiles/ETHOX_H.htm (accessed on 30 June 2023).
  46. Szwiec, E.; Friedman, L.; Buchanan, S. Levels of ethylene oxide biomarker in an exposed residential community. Int. J. Environ. Res. Public Health 2020, 17, 8646. [Google Scholar] [CrossRef]
  47. National Health and Nutrition Examination Survey (NHANES). NHANES 2013–2014 Laboratory Methods: Cotinine and Hydroxycotinine. 2017. Available online: https://wwwn.cdc.gov/nchs/data/nhanes/public/2013/labmethods/cot_h_met.pdf (accessed on 19 November 2024).
  48. Fillenbaum, G.G.; Mohs, R. CERAD (Consortium to Establish a Registry for Alzheimer’s Disease) Neuropsychology Assessment Battery: 35 years and counting. J. Alzheimer’s Dis. 2023, 93, 1–27. [Google Scholar] [CrossRef]
  49. Jaeger, J. Digit Symbol Substitution test. J. Clin. Psychopharmacol. 2018, 38, 513–519. [Google Scholar] [CrossRef]
  50. Sebaldt, R.; Dalziel, W.; Massoud, F.; Tanguay, A.; Ward, R.; Thabane, L.; Melnyk, P.; Landry, P.A.; Lescrauwaet, B. Detection of cognitive impairment and dementia using the animal fluency test: The DECIDE study. Can. J. Neurol. Sci. 2009, 36, 599–604. [Google Scholar] [CrossRef]
  51. National Health and Nutrition Examination Survey (NHANES). 2013–2014 Data Documentation, Codebook, and Frequencies: Cognitive Functioning (CFQ_H). 2017. Available online: https://wwwn.cdc.gov/Nchs/Data/Nhanes/Public/2013/DataFiles/CFQ_H.htm (accessed on 30 June 2023).
  52. Sayyid, Z.N.; Wang, H.; Cai, Y.; Gross, A.L.; Swenor, B.K.; Deal, J.A.; Lin, F.R.; Wanigatunga, A.A.; Dougherty, R.J.; Tian, Q.; et al. Sensory and motor deficits as contributors to early cognitive impairment. Alzheimer’s Dement. 2024, 20, 2653–2661. [Google Scholar] [CrossRef]
  53. National Health and Nutrition Examination Survey (NHANES). 2013–2014 Data Documentation, Codebook, and Frequencies: Muscle Strength—Grip Test (MGX_H). 2016. Available online: https://wwwn.cdc.gov/Nchs/Data/Nhanes/Public/2013/DataFiles/MGX_H.htm (accessed on 6 July 2023).
  54. Tabachnick, B.G.; Fidell, L.S. Using Multivariate Statistics, 7th ed.; Allyn & Bacon: Boston, MA, USA, 2019. [Google Scholar]
  55. Ardila, A. Normal aging increases cognitive heterogeneity: Analysis of dispersion in WAIS-III scores across age. Arch. Clin. Neuropsychol. 2007, 22, 1003–1011. [Google Scholar] [CrossRef] [PubMed]
  56. Weintraub, S.; Besser, L.; Dodge, H.H.; Teylan, M.; Ferris, S.; Goldstein, F.C.; Giordani, B.; Kramer, J.P.; Loewenstein, D.; Marson, D.; et al. Version 3 of the Alzheimer Disease Centers’ neuropsychological test battery in the Uniform Data Set (UDS). Alzheimer Dis. Assoc. Disord. 2018, 32, 10–17. [Google Scholar] [CrossRef]
  57. Welsh, K.A.; Butters, N.; Mohs, R.C.; Beekly, D.; Edland, S.; Fillenbaum, G.; Heyman, A. The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part V. A normative study of the neuropsychological battery. Neurology 1994, 44, 609. [Google Scholar] [CrossRef] [PubMed]
  58. Hooghiemstra, A.M.; Leeuwis, A.E.; Bertens, A.S.; Biessels, G.J.; Bots, M.L.; Rocca, H.-P.B.-L.; Greving, J.P.; Kappelle, L.J.; van Oostenbrugge, R.J.; van Rossum, A.C.; et al. Frequent cognitive impairment in patients with disorders along the heart-brain axis. Stroke 2019, 50, 3369–3375. [Google Scholar] [CrossRef]
  59. Jak, A.J.; Preis, S.R.; Beiser, A.S.; Seshadri, S.; Wolf, P.A.; Bondi, M.W.; Au, R. Neuropsychological criteria for mild cognitive impairment and dementia risk in the Framingham Heart Study. J. Int. Neuropsychol. Soc. 2016, 22, 937–943. [Google Scholar] [CrossRef] [PubMed]
  60. Trittschuh, E.H.; Crane, P.K.; Larson, E.B.; Cholerton, B.; McCormick, W.C.; McCurry, S.M.; Bowen, J.D.; Baker, L.D.; Craft, S. Effects of varying diagnostic criteria on prevalence of mild cognitive impairment in a community based sample. J. Alzheimer’s Dis. 2011, 25, 163–173. [Google Scholar] [CrossRef]
  61. Azzouz, M.; Hasan, Z.; Rahman, M.M.; Gauderman, W.J.; Lorenzo, M.; Lurmann, F.W.; Eckel, S.P.; Palinkas, L.; Johnston, J.; Hurlburt, M.; et al. Does socioeconomic and environmental burden affect vulnerability to extreme air pollution and heat? A case-crossover study of mortality in California. J. Expo. Sci. Environ. Epidemiol. 2024, 35, 294–302. [Google Scholar] [CrossRef]
  62. Wood, A.; Howarth, M. How federal and state regulatory systems perpetuate environmental injustice in the United States: Industrial ethylene oxide emissions as a case study. Environ. Justice 2023, 16, 297–308. [Google Scholar] [CrossRef]
  63. Brashear, A.; Unverzagt, F.W.; Farber, M.O.; Bonnin, J.M.; Garcia, J.G.; Grober, E. Ethylene oxide neurotoxicity: A cluster of 12 nurses with peripheral and central nervous system toxicity. Neurology 1996, 46, 992–998. [Google Scholar] [CrossRef]
  64. Bryant, H.E.; Visser, N.D.; Yoshida, K. Ethylene oxide sterilizer use and short-term symptoms amongst workers. Occup. Med. 1989, 39, 101–106. [Google Scholar] [CrossRef]
  65. Gross, J.A.; Haas, M.L.; Swift, T.R. Ethylene oxide neurotoxicity: Report of four cases and review of the literature. Neurology 1979, 29, 978–983. [Google Scholar] [CrossRef] [PubMed]
  66. Klees, J.E.; Lash, A.; Bowler, R.M.; Shore, M.; Becker, C.E. Neuropsychologic “impairment” in a cohort of hospital workers chronically exposed to ethylene oxide. Clin. Toxicol. 1990, 28, 21–28. [Google Scholar] [CrossRef] [PubMed]
  67. Kloth, S.; Baur, X.; Göen, T.; Budnik, L.T. Accidental exposure to gas emissions from transit goods treated for pest control. Environ. Health A Glob. Access Sci. Source 2014, 13, 110. [Google Scholar] [CrossRef] [PubMed]
  68. Lamontagne, A.D.; Christiani, D.C.; Kelsey, K.T. Utility of the complete blood count in routine medical surveillance for ethylene oxide exposure. Am. J. Ind. Med. 1993, 24, 191–206. [Google Scholar] [CrossRef]
  69. Zey, J.N.; Mortimer, V.D.; Elliott, L.J. Ethylene oxide exposures to hospital sterilization workers from poor ventilation design. Appl. Occup. Environ. Hyg. 1994, 9, 633–641. [Google Scholar] [CrossRef]
  70. Estrin, W.J.; Bowler, R.M.; Lash, A.; Becker, C.E. Neurotoxicological evaluation of hospital sterilizer workers exposed to ethylene oxide. J. Toxicol. Clin. Toxicol. 1990, 28, 1–20. [Google Scholar] [CrossRef]
  71. Gilliland, F.; Avol, E.; Kinney, P.; Jerrett, M.; Dvonch, T.; Lurmann, F.; Buckley, T.; Breysse, P.; Keeler, G.; de Villiers, T.; et al. Air pollution exposure assessment for epidemiologic studies of pregnant women and children: Lessons learned from the Centers for Children’s Environmental Health and Disease Prevention Research. Environ. Health Perspect. 2005, 113, 1447–1454. [Google Scholar] [CrossRef]
  72. Fernández, A.L.; Abe, J. Bias in cross-cultural neuropsychological testing: Problems and possible solutions. Cult. Brain 2018, 6, 1–35. [Google Scholar] [CrossRef]
Table 1. Weighted descriptive statistics of predictors: NHANES demographics and serum cotinine levels for participants completing cognitive (N = 489) and motor (N = 436) tests.
Table 1. Weighted descriptive statistics of predictors: NHANES demographics and serum cotinine levels for participants completing cognitive (N = 489) and motor (N = 436) tests.
CharacteristicN%MSD
HbEtO Adduct Levels 37.173.04
  Background Levels33373.6
  Elevated Levels15626.4
Age (Measured in Years) (N = 489) 68.980.50
Serum Cotinine Levels (N = 480) 37.056.03
Ratio of Family Income/Poverty (N = 456) 3.050.11
Gender (N = 489)
  Men23849.0
  Women25151.0
Race/Ethnicity (N = 489)
  Hispanic917.0
  Non-Hispanic White25681.0
  Non-Hispanic Black927.8
  Non-Hispanic Asian463.8
  Other Race (including multi-racial)40.4
Education (N = 489)
  Less Than 12th Grade11213.1
  Highschool Grade or Equivalent12323.2
  Some college or AA degree13631.1
  College Graduate or Above11832.6
Language for Cognitive Tests (N = 489)
  English42995.1
  Spanish483.9
  Asian Languages121.0
Language Used for Grip Strength (N = 436)
  English40597.6
  Spanish312.4
Table 2. Weighted demographics of sample, comparing participants with background (N = 333) and elevated (N = 156) EtO hemoglobin adduct levels.
Table 2. Weighted demographics of sample, comparing participants with background (N = 333) and elevated (N = 156) EtO hemoglobin adduct levels.
BackgroundElevatedGroup Differences
MSDMSDFp
Age (Measured in Years) (N = 489)69.020.5868.880.775.710.017
Serum Cotinine levels (N = 480)8.145.75117.2127.4787.36<0.001
Ratio of Family Income/Poverty (N = 456)3.180.132.700.159.310.002
N%N%χ2p
Biological Sex (N = 489) 2.820.073
  Men14846.89055.4
  Women18553.26644.6
Race/Ethnicity (N = 489) 12.30<0.001
  Hispanic636.6288.0
  Non-Hispanic White19184.36571.9
  Non-Hispanic Black526.04012.6
  Non-Hispanic Asian252.7217.0
  Other Race (including multi-racial)20.320.6
Education (N = 489) 15.230.030
  Less Than 12th Grade7112.64114.7
  Highschool or Equivalent7619.94732.5
  Some College10234.13422.5
  College Graduate or Above8433.43430.3
Language for Cognitive Tests (N = 489) 2.860.018
  English29995.813092.9
  Spanish293.6195.0
  Asian Languages50.672.2
Language Used for Grip Strength (N = 436) 0.230.841
  English27997.712697.4
  Spanish222.392.6
Table 3. Descriptive statistics of NHANES scores for cognitive (N = 489) and motor function tests (N = 436).
Table 3. Descriptive statistics of NHANES scores for cognitive (N = 489) and motor function tests (N = 436).
Outcome VariableM SD
CERAD List Learning Total20.560.29
CERAD Delayed Score6.540.16
Animal Fluency Test Score18.060.33
Digit Symbol Substitution Test Score51.241.05
Combined Grip Strength Score64.751.42
Table 4. Hierarchical linear regression of available predictors of cognitive performance and elevated HbEtO adduct levels on cognitive function.
Table 4. Hierarchical linear regression of available predictors of cognitive performance and elevated HbEtO adduct levels on cognitive function.
Model1a1b2a2b3a3b4a4b
CERAD-WL Learning TotalCERAD Delay RecallAnimal FluencyDigit Symbol Substitution Test
Step 1Step 2Step 1Step 2Step 1Step 2Step 1Step 2
Age (Years)−0.22 (0.03) ***−0.22 (0.03) ***−0.12 (0.02) ***−0.12 (0.02) ***−0.29 (0.05) ***−0.29 (0.04) ***−0.81 (0.07) ***−0.81 (0.07) ***
Serum Cotinine   0.00 (0.00)  0.00 (0.00)  0.00 (0.00)  0.00 (0.00)−0.00 (0.00)−0.00 (0.00)−0.01 (0.00) **−0.01 (0.01) *
Income/Poverty Ratio  0.40 (0.12) **  0.39 (0.12) **  0.18 (0.08) *  0.16 (0.07) *  0.28 (0.23)  0.25 (0.24)  3.60 (0.40) ***  3.60 (0.40) ***
Male Gender (ref = female)−1.98 (0.32) ***−1.93 (0.30) ***−0.84 (0.11) ***−0.79 (0.11) ***−0.46 (0.57)−0.38 (0.60)−6.78 (1.64) ***−6.79 (1.71) **
Race/Ethnicity (ref = white)
  Hispanic−0.82 (0.56)−0.77 (0.53)−0.69 (0.26) *−0.62 (0.26) *−0.52 (1.04)−0.43 (1.07)−7.48 (2.35) **−7.49 (2.36) **
  Black−1.02 (0.64) −0.94 (0.63)−0.66 (0.21) ** −0.56 (0.21) *−3.04 (0.80) ** −2.90 (0.79) **−8.75 (1.11) ***−8.76 (1.10) ***
  Asian−1.43 (0.82)−1.28 (0.82)−0.63 (0.40)−0.44 (0.40)−3.43 (0.60) ***−3.17 (0.71) ***−3.86 (2.79)−3.88 (3.04)
  Other−4.43 (1.62) *−4.39 (1.63) *−1.53 (0.49) **−1.48 (0.53) *−1.69 (2.46)−1.62 (2.39)−5.63 (2.00) *−5.64 (1.95) *
Education (ref = college grad)
  <12th Grade−1.26 (1.02)−1.35 (0.98)−0.16 (0.52)−0.27 (0.53)−3.80 (1.09) **−3.95 (1.07) **−7.48 (3.44) *−7.46 (3.42) *
  Highschool−0.82 (0.60)−0.77 (0.61)−0.50 (0.38)−0.44 (0.37)−3.93 (0.81) ***−3.84 (0.80) ***−3.62 (2.81)−3.63 (2.90)
  Some College−0.17 (0.68)−0.26 (0.65)−0.06 (0.41)−0.17 (0.39)−1.78 (1.05)−1.93 (1.03)  0.75 (2.81)  0.76 (2.60)
Language (ref = English)
  Spanish−1.25 (0.59) *−1.24 (0.57) * −0.68 (0.37)−0.66 (0.34) −1.26 (1.58)−1.24 (1.66) −6.69 (1.76) **−6.69 (1.77) **
  Asian  5.13 (1.22) ***  5.22 (1.29) **  2.55 (0.43) ***  2.66 (0.51) ***−0.19 (1.46)−0.05 (1.27)  9.30 (3.35) *  9.29 (3.27) *
Elevated HbEtO levels −0.62 (0.48)   0.75 (0.25) ** −1.05 (0.91)   0.09 (2.35)
Constant36.11 (2.38) ***36.19 (2.37) ***15.27 (1.29) ***15.37 (1.24) ***39.75 (3.35) ***39.88 (3.36) ***103.12 (6.02) ***103.11 (6.03) ***
R20.240.240.190.210.250.260.440.44
* p < 0.05, ** p < 0.01, *** p < 0.001.
Table 5. Hierarchical linear regression of predictors of neurocognitive performance and HbEtO levels on combined grip strength.
Table 5. Hierarchical linear regression of predictors of neurocognitive performance and HbEtO levels on combined grip strength.
Step 1Step 2
VariableBSEpBSEp
Age (years)−0.960.10<0.001−0.960.10<0.001
Serum Cotinine Levels−0.010.010.450−0.000.010.853
Poverty Ratio1.060.350.0080.980.360.016
Male Gender (ref = female)28.691.20<0.00128.911.23<0.001
Race/Ethnicity (ref = white)
  Hispanic−7.542.350.006−7.272.300.006
  Black2.012.540.4402.422.520.353
  Asian−6.701.770.002−5.941.710.003
  Other−7.068.290.408−6.727.580.389
Education (ref = college grad)
  <12th Grade0.072.460.979−0.382.390.876
  Highschool−0.681.570.670−0.441.600.786
  Some College1.281.290.3400.811.220.518
Elevated HbEtO levels −3.151.060.009
R2 0.69 0.69
Table 6. Independent t-tests for background vs. elevated HbEtO levels.
Table 6. Independent t-tests for background vs. elevated HbEtO levels.
HbEtO LevelsBackgroundElevated
MSDMSDtdfp
CERAD-WL Learning Trials20.760.3520.010.381.84150.085
CERAD Delayed Recall6.740.155.990.253.53150.003
Animal Fluency18.540.3716.720.562.67150.018
Digit Symbol Substitution Test52.441.3047.911.292.24150.041
Combined Grip Strength65.121.4363.682.400.60150.559
Table 7. Logistic regressions for elevated EtO levels predicting low cognitive scores.
Table 7. Logistic regressions for elevated EtO levels predicting low cognitive scores.
Elevated EtO Levels Predicting Low Scores on Cognitive Tests:BSEOR95% CIp
CERAD-WL Trial # 10.740.262.09[1.20, 3.65]0.013
CERAD-WL Trial # 3 0.360.311.43[0.74, 2.79]0.270
CERAD Delayed Recall0.560.251.76[1.02, 3.02]0.042
Animal Fluency0.580.181.79[1.21, 2.64]0.006
Digit Symbol Substitution Test0.400.281.50[0.82, 2.72]0.172
CI = confidence interval for odds ratio (OR).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

O’Kelley, L.; Swanson, B.; Bishop-Royse, J.; Tam, J.W.; Forsyth, C.; Buchanan, S. Determining Associations Between Levels of Ethylene Oxide Gas Exposure and Neurocognitive Performance for Older U.S. Adults. Int. J. Environ. Res. Public Health 2025, 22, 852. https://doi.org/10.3390/ijerph22060852

AMA Style

O’Kelley L, Swanson B, Bishop-Royse J, Tam JW, Forsyth C, Buchanan S. Determining Associations Between Levels of Ethylene Oxide Gas Exposure and Neurocognitive Performance for Older U.S. Adults. International Journal of Environmental Research and Public Health. 2025; 22(6):852. https://doi.org/10.3390/ijerph22060852

Chicago/Turabian Style

O’Kelley, Linda, Barbara Swanson, Jessica Bishop-Royse, Joyce W. Tam, Christopher Forsyth, and Susan Buchanan. 2025. "Determining Associations Between Levels of Ethylene Oxide Gas Exposure and Neurocognitive Performance for Older U.S. Adults" International Journal of Environmental Research and Public Health 22, no. 6: 852. https://doi.org/10.3390/ijerph22060852

APA Style

O’Kelley, L., Swanson, B., Bishop-Royse, J., Tam, J. W., Forsyth, C., & Buchanan, S. (2025). Determining Associations Between Levels of Ethylene Oxide Gas Exposure and Neurocognitive Performance for Older U.S. Adults. International Journal of Environmental Research and Public Health, 22(6), 852. https://doi.org/10.3390/ijerph22060852

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop