INTRODUCTION

Depression is a major global public health concern, clinically defined as a mood disorder characterized by persistent sadness, emptiness, or irritability, often accompanied by somatic and cognitive changes that substantially impair daily functioning [1]. It ranks among the leading causes of disability and disease burden worldwide [2], with an estimated global prevalence of approximately 5% among adults [3]. Thus, identifying the determinants of depression remains a critical research priority.

The etiology of depression involves a complex interplay of multiple factors. Genetic predisposition, epigenetic modifications, and psychological and social stressors have long been recognized as contributing elements [1,4,5]. In recent years, increasing attention has been directed toward environmental determinants [5]. Growing evidence implicates air pollution as a significant risk factor that may influence both the prevalence and severity of depression [5-7].

Among air pollutants, benzene and its organic derivatives—toluene, ethylbenzene, and xylenes—collectively referred to as BTEX [8], are ubiquitous chemicals emitted from industrial processes, fuel combustion, tobacco smoke, consumer products, and building materials [9,10]. Tobacco smoke, in particular, is the primary source of benzene exposure in the general population and contributes to exposure to other BTEX compounds. Their presence is especially notable indoors, where concentrations frequently exceed outdoor levels and where people spend most of their time [9].

BTEX compounds have demonstrated neurotoxic effects even at environmental exposure levels [8,11] and have been shown to induce oxidative stress and disrupt neurotransmitter systems [11-14]. These mechanisms correspond to biological pathways linking air pollutant exposure to depression risk [7]. Specifically, BTEX exposure has been associated with mood disorders, cognitive impairments, and depressive symptoms in both occupational and environmental settings [8,14-20]. Given their widespread occurrence and potential health impacts, numerous countries have implemented specific regulations to limit BTEX exposure [21,22].

However, most previous studies have focused on high-exposure scenarios, such as occupational environments or substance abuse [15-17], while more recent studies using National Health and Nutrition Examination Survey (NHANES) data have explored associations at environmental levels in the United States population [18-20]. Evidence from other populations, however, remains limited, leaving a critical knowledge gap regarding the effects of low-level, chronic exposure. To address this gap, we analyzed data from the nationally representative Korea National Health and Nutrition Examination Survey (KNHANES). Given that BTEX compounds commonly co-exist in daily environments and possess plausible additive neurotoxic potential even at environmental concentrations [8], this study aimed to assess the effects of BTEX exposure on depression risk and to identify the principal contributors within the mixture.

METHODS

Study Design and Participants

KNHANES is a national surveillance system that has assessed the health and nutritional status of Koreans since 1998 [23]. To reflect increased time spent indoors during the coronavirus disease 2019 (COVID-19) pandemic, KNHANES incorporated indoor air quality (IAQ) monitoring and biomarker assessments of environmentally hazardous substances [24,25]. From July 2020 to August 2021, approximately 1200 households were sampled for IAQ measurements, stratified by season, region, and housing type. Biomarker analyses were performed on about 2000 individuals aged 19 years or older from these households who consented to biological sampling.

Of the 1966 adults who participated in the survey, we excluded those who did not complete the depression questionnaire (n=94), those with urinary creatinine levels <0.3 g/L or >3.0 g/L (n=106) [26], and those with missing covariate data (n=33). The final analytic sample included 1733 participants.

Exposure Assessment

To assess individual BTEX exposure, we used well-established urinary biomarkers of exposure: S-phenylmercapturic acid (SPMA) for benzene [27], benzylmercapturic acid (BMA) for toluene [28], phenylglyoxylic acid (PGA) plus mandelic acid (MA) for ethylbenzene [29], and methylhippuric acid (MHA) for xylenes [30]. Urinary BTEX metabolites were quantified using high-performance liquid chromatography (Nexera XR LC-20AD System; Shimadzu, Kyoto, Japan) coupled with mass spectrometry (Triple Quad API 5500; Sciex, Framingham, MA, USA). The lower limits of detection (LLODs) for SPMA, BMA, PGA, MA, 2-MHA, and 3-MHA and 4-MHA were 0.251, 0.010, 1.078, 0.252, 0.985, and 5.337 μg/L, respectively. Values below the LLOD were imputed as LLOD/√2, consistent with KNHANES analytical guidelines. Metabolite concentrations were adjusted for urinary creatinine to account for variations in dilution due to urine output [26]. Because of their right-skewed distributions, all creatinine-adjusted metabolite concentrations were log-transformed prior to statistical analyses.

Outcome Definition

Depression was defined as a self-reported physician diagnosis. During structured interviews conducted by trained personnel, participants were asked whether they had ever been diagnosed with depression by a physician.

Covariates

Data on socio-demographic, lifestyle, and health-related factors were collected through standardized interviews. Covariates were selected based on established associations with depression reported in previous studies and included age, sex, education level, occupation type, marital status, household income, residential area, smoking status, drinking status, physical activity, obesity, and subjective health status.

Age was categorized as 19–39 years, 40–59 years, or ≥60 years, and sex as male or female. Education level was classified as <high school, high school, or >high school. Occupation was grouped as white-collar (managers, professionals and related workers, clerks, service workers, sales workers), blue-collar (skilled agricultural/forestry/fishery workers, craft and related trades workers, equipment/machine operators and assemblers, elementary workers), or unemployed. The unemployed category included individuals without economic activity, such as homemakers and students, though further differentiation was not possible with available data. Marital status was categorized as never married, widowed/divorced/separated, or married/living with a partner. Household income was divided into quartiles. Residential area was classified as urban or rural. Smoking status was defined as having smoked at least 100 cigarettes in one’s lifetime. Drinking behavior was based on binge drinking within the past year, defined as consuming ≥7 shots of liquor or ≥5 cans of beer for male, and ≥5 shots of liquor or ≥3 cans of beer for female. Physical activity was dichotomized according to global health recommendations for adults (≥150 minutes of moderate-intensity or 75 minutes of vigorous-intensity aerobic activity per week, or an equivalent combination). Obesity was defined as a body mass index ≥25 kg/m². Subjective health status was categorized as very good/good, fair, or poor/very poor.

Statistical Analysis

To account for the KNHANES complex sampling design, sample weights, strata, and cluster variables were applied to all analyses, except for the weighted quantile sum (WQS) regression. Descriptive statistics were stratified by depression status. Categorical variables were expressed as frequencies (weighted proportions) and compared using the Rao–Scott χ² test. BTEX metabolite concentrations were summarized as weighted geometric means (GMs) and geometric standard deviations (GSDs) and compared using 2-sample t-tests on log-transformed values. Spearman correlation coefficients were calculated to evaluate co-exposure patterns among BTEX metabolites.

Logistic regression was used to examine associations between each BTEX metabolite and depression, modeling exposures both as continuous variables and as categorical variables based on tertiles of their distributions. A series of progressively adjusted models were constructed: Model 1 was unadjusted; Model 2 was adjusted for age and sex; Model 3 additionally adjusted for education level, occupation, marital status, household income, residential area, smoking, drinking, physical activity, obesity, and subjective health status; and Model 4 further adjusted for the other BTEX metabolites. To assess potential non-linearity, restricted cubic spline regression with 3 knots was applied to the continuous metabolite variables [31].

Given that BTEX compounds frequently co-occur due to shared sources and may exert additive neurotoxic effects [8-10], we also evaluated the joint effect of BTEX exposure using WQS regression. This approach constructs a single weighted index representing the overall mixture effect, thereby mitigating issues of multicollinearity and multiple comparisons, while identifying the chemicals contributing most strongly to the association [32,33]. Analyses were conducted using the same covariate sets as the logistic regression models: unadjusted; adjusted for age and sex; and fully adjusted for all covariates. The dataset was randomly divided into a 40% training set (used to estimate metabolite weights via bootstrap, n=1000) and a 60% validation set (used to test the significance of the WQS index regression coefficient) [32]. To address potential instability from a single split, repeated holdout validation was performed by randomly partitioning the dataset 100 times and averaging results to obtain final estimates [33].

Because tobacco smoke is a major source of environmental BTEX exposure, we conducted sensitivity analyses restricting the sample to participants who reported neither current smoking nor environmental tobacco smoke (ETS) exposure, defined as indoor exposure at home, work, or public places within the past 7 days.

All analyses were conducted using R version 4.3.2 (R Foundation for Statistical Computing, Vienna, Austria). The ‘gWQS’ package was used for WQS regression. Statistical significance was defined as p-value <0.05 (two-sided).

Ethics Statement

We conducted a cross-sectional analysis using publicly available data from the KNHANES, which was approved by the Institutional Review Board of the Korea Disease Control and Prevention Agency (2018-01-03-2C-A; 2018-01-03-5C-A).

RESULTS

Participant Characteristics

Participant characteristics are summarized in Table 1. Among the 1733 participants, 85 (5.5%) reported a physician-confirmed diagnosis of depression. The depression group included a lower proportion of participants aged 40–59 years (19.7%) than the non-depression group (41.8%) and a higher proportion of females (73.2 vs. 49.5%). Individuals with depression were less likely to have education beyond high school (27.0%), to be married or living with a partner (50.8%), or to belong to the highest household income quartile (16.6%). Conversely, participants with depression were more likely to be unemployed (51.4%) and to reside in urban areas (94.3%). No significant differences were observed between the groups in smoking status, drinking behavior, physical activity, or obesity. However, a significantly higher proportion of individuals with depression reported “poor” or “very poor” subjective health status (39.2%) compared with those without depression (13.6%).

Distribution of Benzene, Toluene, Ethylbenzene, and Xylenes Metabolites

Table 2 presents the weighted distribution of urinary BTEX metabolites. The GMs (GSDs) of each metabolite (μg/g creatinine) were as follows: SPMA 0.77 (1.71), BMA 5.01 (2.37), PGA+MA 404.44 (1.70), and MHA 127.62 (2.20). These levels were substantially lower than the Biological Exposure Indices (BEIs) recommended by the American Conference of Governmental Industrial Hygienists (ACGIH), which are intended to represent occupational exposure concentrations unlikely to cause adverse effects in nearly all workers: 25 μg/g creatinine for SPMA, 150 000 μg/g creatinine for PGA+MA, and 300 000 μg/g creatinine for MHA [26]. BMA was not compared because it is not included among ACGIH-designated BEI biomarkers [26].

Only SPMA concentrations were significantly higher in the depression group (GM, 0.97; GSD, 1.70) compared with the non-depression group (GM, 0.76; GSD, 1.70) (p<0.01). No significant between-group differences were found for other BTEX metabolites.

Correlations Among Benzene, Toluene, Ethylbenzene, and Xylenes Metabolites

Figure 1 displays the correlations among BTEX metabolites. Significant positive correlations were observed across all pairs. Specifically, SPMA was weakly correlated with BMA (r=0.38) and PGA+MA (r=0.35); BMA correlated with PGA+MA (r=0.31); and PGA+MA correlated with MHA (r=0.35).

Association Between Individual Benzene, Toluene, Ethylbenzene, and Xylenes Metabolites and Depression

Table 3 summarizes the associations between BTEX metabolites and depression. SPMA exhibited a significant positive association with depression risk. In continuous models, each unit increase in ln-transformed SPMA was associated with higher odds of depression: unadjusted (odds ratio [OR], 2.27; 95% CI, 1.38 to 3.72; p<0.01), age-adjusted and sex-adjusted: (OR, 2.70; 95% CI, 1.54 to 4.74, p<0.01), and fully adjusted (OR 2.35; 95% CI, 1.21 to 4.54; p=0.01). The association remained statistically significant after additional adjustment for the other BTEX metabolites (OR, 2.62; 95% CI, 1.34 to 5.13; p<0.01). No statistically significant nonlinearity was observed for any BTEX metabolite (p for non-linearity >0.05). Categorical analyses based on tertiles of metabolite distributions also showed a consistent positive association between SPMA and depression. None of the other BTEX metabolites were significantly associated with depression risk in any of the models.

Association Between Benzene, Toluene, Ethylbenzene, and Xylenes Mixture and Depression

Figure 2 presents the results of the WQS regression analysis assessing BTEX mixture exposure in relation to depression. The BTEX mixture was significantly associated with an increased risk of depression in the unadjusted model (OR, 1.51; 95% CI, 1.12 to 2.03). However, the association was attenuated and became non-significant after adjusting for age and sex (OR, 1.34; 95% CI, 0.91 to 1.99) and in the fully adjusted model (OR, 1.12; 95% CI, 0.76 to 1.67).

Despite the loss of statistical significance in adjusted models, SPMA consistently emerged as the dominant contributor to the overall mixture effect, accounting for 69.8% of the total weight in the unadjusted model, 62.1% in the age-adjusted and sex-adjusted model, and 76.1% in the fully adjusted model.

Sensitivity Analysis

Given that both logistic regression and WQS regression identified SPMA as the principal compound associated with depression, we performed sensitivity analyses to test the robustness of this association while controlling for the potential confounding effects of tobacco smoke. Logistic regression analyses for SPMA were repeated in 2 subgroups: participants who were not current smokers (n=1472) and participants who were neither current smokers nor exposed to ETS (n=1286) (Supplemental Material 1). The significant positive association between urinary SPMA and depression persisted in both subgroups across all models, confirming the robustness of the findings.

DISCUSSION

In this nationally representative study of Korean adults, we found that higher exposure to benzene, assessed through its urinary metabolite SPMA, was significantly and consistently associated with increased odds of depression. This association remained robust after adjusting for a comprehensive set of potential confounders, including co-exposure to other BTEX compounds. Although analysis of the BTEX mixture using WQS regression did not reach statistical significance after covariate adjustment, SPMA consistently emerged as the predominant contributor to depression risk. Importantly, the association between benzene exposure and depression persisted even among individuals unexposed to tobacco smoke.

Our findings are consistent with recent research examining the effects of volatile organic compound (VOC) exposure on depression. Cross-sectional studies using data from the United States NHANES have reported significant associations between exposure to VOC mixtures and depression [18-20]. Although the specific VOCs identified as key contributors varied across studies, BTEX compounds were consistently among them [18-20]. These population-based findings extend earlier occupational health research, which has long documented that high-level exposure to organic solvents containing BTEX is associated with elevated rates of mood disorders [15-17].

Notably, NHANES studies that measured parent compounds in blood consistently identified benzene as a primary contributor [18,19], aligning with our finding that benzene is a major component of the BTEX effect on depression. In contrast, another NHANES study using urinary biomarkers identified PGA and 3,4-MHA as the principal contributors, likely explained by their higher urinary concentrations (GM: PGA, 2.56 ng/0.01 mg creatinine; 3,4-MHA, 4.67 ng/0.01 mg creatinine) compared with those observed in our study (GM: PGA, 218.01 μg/g creatinine; 3,4-MHA, 105.87 μg/g creatinine) [20]. However, that study’s biomarker panel did not include a specific urinary indicator of benzene exposure, limiting direct comparison [20]. Our findings are further supported by evidence from the UK Biobank, a large-scale cohort study that reported a significant association between long-term ambient benzene exposure and increased risk of major depression [34]. Similarly, research conducted among Gulf state residents found positive associations between blood BTEX levels (particularly benzene) and central nervous system symptoms [10]. Animal studies corroborate these findings, demonstrating that benzene exposure decreases hippocampal acetylcholine and midbrain dopamine and norepinephrine, neurotransmitters implicated in the pathophysiology of depression [35].

A key aspect of our analytical strategy was the evaluation of BTEX compounds as a chemical mixture. This approach is critical because BTEX compounds often originate from shared sources such as fuel combustion and industrial emissions [9,10], leading to correlated exposures in the general population. Moreover, due to their structural similarities, BTEX compounds are believed to exert additive neurotoxic effects even at environmental levels, as noted by the Agency for Toxic Substances and Disease Registry [8]. To address these interdependencies, we employed WQS regression, a statistical method specifically developed for mixture analysis under the assumption of additive effects [32,33]. WQS regression is particularly useful because it mitigates the multicollinearity inherent in correlated exposure data, yielding a single, interpretable estimate of the mixture’s overall effect [32,33]. Another strength of this approach is its capacity to identify the individual chemicals contributing most strongly to the association by assigning empirically derived weights to each component [32]. This enabled us not only to estimate the combined risk of BTEX exposure but also to pinpoint benzene as the principal driver.

The absence of a statistically significant BTEX mixture effect after covariate adjustment may reflect the relatively low exposure levels in our study population [20,26], which could result in subclinical neuropsychiatric effects below the threshold of detection. Additionally, using physician-diagnosed depression, as opposed to symptom-based screening tools employed in NHANES studies [18-20], may have limited sensitivity for identifying mild or subthreshold depressive states. The reduced sample size after adjustment for multiple covariates may also have constrained statistical power to detect modest associations. Nevertheless, the consistent pattern implicating benzene exposure across multiple analytical frameworks underscores its potential role in depression etiology. Larger and longitudinal studies are warranted to confirm these findings and to clarify the biological pathways through which low-level benzene exposure may contribute to depressive disorders.

This study has several notable strengths. To our knowledge, it is the first to demonstrate a potential association between BTEX exposure and depression in a large, nationally representative sample of Korean adults, enhancing the generalizability of the findings. Second, we employed specific urinary metabolites as biomarkers of exposure, providing an integrated measure of internal dose from all exposure routes and sources [27-30]. The use of SPMA—the most specific and widely validated biomarker for benzene exposure [27,36]—adds credibility to our results. However, interpretation of some biomarkers warrants caution. BMA is a metabolite of benzyl alcohol, which is both an oxygenated product of toluene and a solvent commonly used in paints, dyes, and cosmetics [28]. PGA and MA are metabolites of both ethylbenzene and styrene [29]. Thus, caution is needed when interpreting their specificity as exposure indicators. Third, the application of multiple statistical models produced consistent results, reinforcing the conclusion that benzene is the principal toxicant associated with depression.

Nonetheless, several limitations should be acknowledged. First, the cross-sectional design limits the ability to infer causality. Second, exposure assessment was based on a single spot urine sample. Because the urinary biomarkers analyzed have short biological half-lives, they primarily reflect recent BTEX exposure [37]. This limitation may introduce exposure misclassification; however, the consistency of our findings after excluding individuals exposed to tobacco smoke—a major and variable source of acute exposure [38,39]—supports the robustness of our results. Third, depression status was determined using self-reported physician diagnoses. Although this measure has been shown to correspond closely to clinical diagnoses [40], it may underestimate prevalence by missing undiagnosed cases. Future studies would benefit from incorporating standardized diagnostic interviews or validated symptom scales, such as the Patient Health Questionnaire-9, which was included in KNHANES only for limited survey years and therefore had restricted applicability. Fourth, WQS regression assumes directional homogeneity, meaning that all exposures are expected to have effects in the same direction or none at all. In our data, PGA+MA, although not statistically significant, exhibited an opposite direction of association compared with the other compounds, potentially violating this assumption. Therefore, results from mixture analyses should be interpreted with caution. Finally, despite adjusting for a comprehensive set of covariates, residual confounding due to unmeasured factors, such as genetic vulnerability or psychosocial stress [1,4,5], cannot be entirely excluded.

In conclusion, this study provides novel evidence from a nationally representative Korean population that environmental exposure to benzene is associated with an increased risk of depression, independent of tobacco smoke exposure. The robustness of this association across multiple models and sensitivity analyses underscores benzene’s potential role as a neurotoxicant contributing to depressive disorders. These findings highlight the need for strengthened environmental policies and public health interventions aimed at reducing benzene exposure as a means to help mitigate the growing mental health burden of depression. Future research should prioritize longitudinal designs with repeated biomarker assessments to establish causal relationships and better characterize long-term exposure–response dynamics over time.

Supplemental Materials

Notes

Conflict of Interest

The authors have no conflicts of interest associated with the material presented in this paper.

Funding

None.

Acknowledgements

This research was supported by the Chung-Ang University Graduate Research Scholarship in 2023.

Author Contributions

Conceptualization: Park HW, Choi BS, Park B. Data curation: Park HW, Lee WY. Formal analysis: Park HW, Choi BS, Lee W, Lee WY. Funding acquisition: None. Methodology: Park B, Lee W, Lee WY. Project administration: Park B, Lee WY. Visualization: Park HW, Choi BS, Lee W. Writing – original draft: Park HW. Writing – review & editing: Park HW, Choi BS, Park B, Lee W, Lee WY.

Figure. 1.

Correlations among urinary benzene, toluene, ethylbenzene, and xylenes metabolite levels. SPMA, S-phenyl-mercapturic acid; BMA, benzyl-mercapturic acid; PGA, phenyl-glyoxylic acid; MA, mandelic acid; MHA, methylhippuric acid.

Figure. 2.

ORs for the association between benzene, toluene, ethylbenzene, and xylenes (BTEX) mixture exposure and depression risk, and the weighted contributions of individual BTEX metabolites estimated by weighted quantile sum regression. (A) Model 1: unadjusted. (B) Model 2: adjusted for age and sex. (C) Model 3: adjusted for age, sex, education, occupation, marital status, household income, residential area, smoking status, drinking status, physical activity, obesity, and subjective health status. SPMA, S-phenyl-mercapturic acid; BMA, benzyl-mercapturic acid; PGA, phenyl-glyoxylic acid; MA, mandelic acid; MHA, methylhippuric acid; OR, odds ratio; CI, confidence interval.

REFERENCES

• 1. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, fifth edition, text revision: DSM-5-TRTM. Washington, DC: American Psychiatric Association Publishing; 2022. https://doi.org/10.1176/appi.books.9780890425787
• 2. GBD 2019 Mental Disorders Collaborators. Global, regional, and national burden of 12 mental disorders in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Psychiatry 2022;9(2):137-150. https://doi.org/10.1016/S2215-0366(21)00395-3ArticlePubMedPMC
• 3. World Health Organization. Depressive disorder (depression) [cited 2023 Dec 16]. Available from: https://www.who.int/news-room/fact-sheets/detail/depression
• 4. Hammen C. Risk factors for depression: an autobiographical review. Annu Rev Clin Psychol 2018;14: 1-28. https://doi.org/10.1146/annurev-clinpsy-050817-084811ArticlePubMed
• 5. van den Bosch M, Meyer-Lindenberg A. Environmental exposures and depression: biological mechanisms and epidemiological evidence. Annu Rev Public Health 2019;40: 239-259. https://doi.org/10.1146/annurev-publhealth-040218-044106ArticlePubMed
• 6. Borroni E, Pesatori AC, Bollati V, Buoli M, Carugno M. Air pollution exposure and depression: a comprehensive updated systematic review and meta-analysis. Environ Pollut 2022;292(Pt A):118245. https://doi.org/10.1016/j.envpol.2021.118245ArticlePubMed
• 7. Zundel CG, Ryan P, Brokamp C, Heeter A, Huang Y, Strawn JR, et al. Air pollution, depressive and anxiety disorders, and brain effects: a systematic review. Neurotoxicology 2022;93: 272-300. https://doi.org/10.1016/j.neuro.2022.10.011ArticlePubMedPMC
• 8. Agency for Toxic Substances and Disease Registry. Interaction profiles for toxic substances: benzene, toluene, ethylbenzene, and xylenes (BTEX); 2004 [cited 2024 Jan 17]. Available from: https://www.atsdr.cdc.gov/interaction-profiles/about/benzene-toluene-ethylbenzene-and-xylenes-btex.html#cdc_generic_section_3-accessing-interaction-profile-pdfs
• 9. Bolden AL, Kwiatkowski CF, Colborn T. New look at BTEX: are ambient levels a problem? Environ Sci Technol 2015;49(9):5261-5276. https://doi.org/10.1021/es505316fArticlePubMed
• 10. Davidson CJ, Hannigan JH, Bowen SE. Effects of inhaled combined benzene, toluene, ethylbenzene, and xylenes (BTEX): toward an environmental exposure model. Environ Toxicol Pharmacol 2021;81: 103518. https://doi.org/10.1016/j.etap.2020.103518ArticlePubMed
• 11. Werder EJ, Engel LS, Blair A, Kwok RK, McGrath JA, Sandler DP. Blood BTEX levels and neurologic symptoms in Gulf states residents. Environ Res 2019;175: 100-107. https://doi.org/10.1016/j.envres.2019.05.004ArticlePubMedPMC
• 12. Sainio MA Sr. Neurotoxicity of solvents. Handb Clin Neurol 2015;131: 93-110. https://doi.org/10.1016/B978-0-444-62627-1.00007-XArticlePubMed
• 13. Wang F, Fangfang Z, Guo X, Chen W, Yao W, Liu H, et al. Effects of volatile organic compounds and carbon monoxide mixtures on learning and memory, oxidative stress, and monoamine neurotransmitters in the brains of mice. Toxicol Ind Health 2018;34(3):178-187. https://doi.org/10.1177/0748233717747504ArticlePubMed
• 14. Khan A, Kanwal H, Bibi S, Mushtaq S, Khan A, Khan YH, et al. Volatile organic compounds and neurological disorders: from exposure to preventive interventions. In: Akash MS, Rehman K, editors. Environmental contaminants and neurological disorders. Cham: Springer; 2021. p. 201-230
• 15. Condray R, Morrow LA, Steinhauer SR, Hodgson M, Kelley M. Mood and behavioral symptoms in individuals with chronic solvent exposure. Psychiatry Res 2000;97(2-3):191-206. https://doi.org/10.1016/s0165-1781(00)00217-1ArticlePubMed
• 16. Morrow LA, Gibson C, Bagovich GR, Stein L, Condray R, Scott A. Increased incidence of anxiety and depressive disorders in persons with organic solvent exposure. Psychosom Med 2000;62(6):746-750. https://doi.org/10.1097/00006842-200011000-00002ArticlePubMed
• 17. Saraei M, Golshan M, Aminian O, Eftekhari S. Association between solvent exposure and depression among industrial workers. J Public Health (Berl) 2022;30: 953-958. https://doi.org/10.1007/s10389-020-01376-yArticle
• 18. Zhuang Y, Zhang X, Sun X, Liu Z, Yu Q, Dong C, et al. Association of environmental volatile organic compounds with depression in adults: NHANES 2013-2018. Hyg Environ Health Adv 2023;6: 100058. https://doi.org/10.1016/j.heha.2023.100058Article
• 19. Zhu Y, Ju Y, Wang M, Yang Y, Wu R. Association of volatile organic compounds exposure with the risk of depression in U.S. adults: a cross‑sectional study from NHANES 2013-2016. Int Arch Occup Environ Health 2023;96(8):1101-1111. https://doi.org/10.1007/s00420-023-01993-6ArticlePubMed
• 20. Tang L, Liu M, Tian J. Volatile organic compounds exposure associated with depression among U.S. adults: results from NHANES 2011-2020. Chemosphere 2024;349: 140690. https://doi.org/10.1016/j.chemosphere.2023.140690ArticlePubMed
• 21. Tsai WT. An overview of health hazards of volatile organic compounds regulated as indoor air pollutants. Rev Environ Health 2019;34(1):81-89. https://doi.org/10.1515/reveh-2018-0046ArticlePubMed
• 22. Settimo G, Manigrasso M, Avino P. Indoor air quality: a focus on the European legislation and state-of-the-art research in Italy. Atmosphere 2020;11(4):370. https://doi.org/10.3390/atmos11040370Article
• 23. Kweon S, Kim Y, Jang MJ, Kim Y, Kim K, Choi S, et al. Data resource profile: the Korea National Health and Nutrition Examination Survey (KNHANES). Int J Epidemiol 2014;43(1):69-77. https://doi.org/10.1093/ije/dyt228ArticlePubMedPMC
• 24. Kim SJ, Kim Y, Hwang Y, Oh K, Park J, Lee K. Development of the indoor air quality monitoring model based on Korea National Health and Nutrition Examination Survey (KNHANES). Public Health Wkly Rep 2021;14(5):239-247. (Korean)
• 25. Kim SJ, Oh K. Research on biomarkers related to environmentally hazardous substances based on the Korea National Health and Nutrition Examination Survey (KNHANES). Public Health Wkly Rep 2022;15(9):588-593. (Korean)
• 26. American Conference of Governmental Industrial Hygienists (ACGIH). Threshold limit values for chemical substances and physical agents & biological exposure indices. Cincinnati: ACGIH Worldwide; 2025
• 27. Agency for Toxic Substances and Disease Registry. Toxicological profile for benzene; 2007 [cited 2024 Jan 17]. Available from: https://www.epa.gov/sites/default/files/2014-03/documents/benzene_toxicological_profile_tp3_3v.pdf
• 28. Agency for Toxic Substances and Disease Registry. Toxicological profile for toluene; 2017 [cited 2024 Jan 17]. Available from: https://www.atsdr.cdc.gov/toxprofiles/tp56-p.pdf
• 29. Agency for Toxic Substances and Disease Registry. Toxicological profile for ethylbenzene; 2010 [cited 2024 Jan 17]. Available from: https://www.atsdr.cdc.gov/toxprofiles/tp110.pdf
• 30. Agency for Toxic Substances and Disease Registry. Toxicological profile for xylene; 2007 [cited 2024 Jan 17]. Available from: https://www.atsdr.cdc.gov/toxprofiles/tp71.pdf
• 31. Desquilbet L, Mariotti F. Dose-response analyses using restricted cubic spline functions in public health research. Stat Med 2010;29(9):1037-1057. https://doi.org/10.1002/sim.3841ArticlePubMed
• 32. Carrico C, Gennings C, Wheeler DC, Factor-Litvak P. Characterization of weighted quantile sum regression for highly correlated data in a risk analysis setting. J Agric Biol Environ Stat 2015;20(1):100-120. https://doi.org/10.1007/s13253-014-0180-3ArticlePubMedPMC
• 33. Tanner EM, Bornehag CG, Gennings C. Repeated holdout validation for weighted quantile sum regression. MethodsX 2019;6: 2855-2860. https://doi.org/10.1016/j.mex.2019.11.008ArticlePubMedPMC
• 34. Li Y, Bao Y, Qiang N, Zhong M, Han Z, Li Y, et al. Long-term exposure to ambient benzene and brain disorders among urban adults. Nat Cities 2024;1: 830-841. https://doi.org/10.1038/s44284-024-00156-zArticle
• 35. Kanada M, Miyagawa M, Sato M, Hasegawa H, Honma T. Neurochemical profile of effects of 28 neurotoxic chemicals on the central nervous system in rats (1). Effects of oral administration on brain contents of biogenic amines and metabolites. Ind Health 1994;32(3):145-164. https://doi.org/10.2486/indhealth.32.145ArticlePubMed
• 36. Li AJ, Pal VK, Kannan K. A review of environmental occurrence, toxicity, biotransformation and biomonitoring of volatile organic compounds. Environ Chem Ecotoxicol 2021;3: 91-116. https://doi.org/10.1016/j.enceco.2021.01.001Article
• 37. Louis LM, Kavi LK, Boyle M, Pool W, Bhandari D, De Jesús VR, et al. Biomonitoring of volatile organic compounds (VOCs) among hairdressers in salons primarily serving women of color: a pilot study. Environ Int 2021;154: 106655. https://doi.org/10.1016/j.envint.2021.106655ArticlePubMedPMC
• 38. Oh HH, Park H, Kim DH, Son BC, Lee CK, Kim K, et al. The relationship between urinary BTEX metabolites and residence setting among Korean homemakers: the first Korea National Environmental Health Survey (2009–2011). Ann Occup Environ Med 2017;29: 38. https://doi.org/10.1186/s40557-017-0189-5Article
• 39. Vardoulakis S, Giagloglou E, Steinle S, Davis A, Sleeuwenhoek A, Galea KS, et al. Indoor exposure to selected air pollutants in the home environment: a systematic review. Int J Environ Res Public Health 2020;17(23):8972. https://doi.org/10.3390/ijerph17238972ArticlePubMedPMC
• 40. Sanchez-Villegas A, Schlatter J, Ortuno F, Lahortiga F, Pla J, Benito S, et al. Validity of a self-reported diagnosis of depression among participants in a cohort study using the Structured Clinical Interview for DSM-IV (SCID-I). BMC Psychiatry 2008;8: 43. https://doi.org/10.1186/1471-244X-8-43ArticlePubMedPMC

Figure & Data

References

Citations

Citations to this article as recorded by