Abstract
Background
Phthalates are synthetic plasticizers widely used in consumer and industrial products, resulting in widespread human exposure and potential adverse reproductive, neurodevelopmental, and metabolic effects. However, the relative contributions of diverse lifestyle-related sources to the overall phthalate burden remain unclear. In this study, we aimed to identify dietary and daily product use factors most prominently associated with joint variations in urinary phthalate metabolite mixtures in a nationally representative sample of Korean adults.
Methods
We analyzed data from 3,760 adults from the Korean National Environmental Health Survey Cycle 4 (2018–2020). Information on dietary consumption, food-contact material use, and consumer product use was collected through questionnaires, and urinary concentrations of eight phthalate metabolites [mono-methyl phthalate, mono-ethyl phthalate, mono-n-butyl phthalate, mono-benzyl phthalate, mono-(2-ethyl-5-hydroxyhexyl) phthalate, mono-(2-ethyl-5-oxohexyl) phthalate, mono-(2-ethyl-5-carboxypentyl) phthalate, and mono-(3-carboxypropyl) phthalate] were measured. Sparse reduced-rank regression was used to identify latent patterns of dietary consumption, food-contact material use, and consumer product use jointly associated with the eight metabolite levels after covariates adjustment, and analyses were repeated by sex.
Results
A single latent pattern was identified, in which all phthalate metabolites showed positive loadings (+0.10 to +0.46). Mineral spring water (+0.08), grains (+0.08), tap water (+0.06), and beverages in plastic pouches (+0.08) exhibited the largest loadings in the same direction as the phthalate metabolites. In sex-stratified analyses, drinking water and grain consumption among men and the use of beverages in plastic pouches and makeup products among women showed prominent loadings in the same direction as the eight metabolites.
Conclusion
Consumption of mineral spring water, tap water, grains, and beverages in plastic pouches were major dietary and daily product use factors associated with higher urinary phthalate metabolite mixtures in Korean adults. Given that current Korean regulations cover a limited subset of phthalates at levels less stringent than international standards, expanding regulatory frameworks for drinking water and food-contact materials and adopting precautionary lifestyle measures are warranted, along with prospective studies to clarify source-specific causal pathways.
Keywords: biomonitoring, dietary exposure, drinking water, exposure assessment, food contact materials, phthalates, sparse reduced-rank regression
1. Introduction
Phthalates are a group of synthetic chemicals derived from ortho-phthalic acid and are widely used as plasticizers to enhance the flexibility and softness of plastics (1). They are used in a broad range of products, including food-contact materials (FCMs), personal care products, building materials, home improvement and cleaning products, and medical devices (1, 2). Phthalates do not form covalent bonds with plastic polymers; hence, they easily leach into the surrounding environment during production, usage, and disposal stages (3). Consequently, phthalates have become ubiquitous in environmental and biological matrices worldwide (1, 2, 4, 5).
Human exposure to phthalates is therefore widespread, continuous, and mediated through multiple pathways across the general population, posing a substantial public health concern (1–4, 6). Phthalates are recognized as endocrine disruptors and have been associated with adverse reproductive (7, 8) and metabolic outcomes (9) and, although the evidence is less consistent, with neurodevelopmental outcomes (10, 11), at the chronic low-level exposures typical of the general population (5, 12). Accordingly, characterizing the background exposure to phthalates is critical.
Background exposure is typically assessed through human biomonitoring, with biomarker selection guided by phthalate metabolism. Parent phthalate diesters are rapidly hydrolyzed by esterases to form their corresponding primary monoester metabolites. Low-molecular-weight phthalates are excreted mainly as these monoesters, whereas high-molecular-weight phthalates undergo additional side-chain oxidation to secondary metabolites (5, 6, 13–15). Thus, urinary metabolites serve as biomarkers of phthalate exposure. They are preferred over the parent diesters because diesters are ubiquitous in laboratory and sampling materials, making their measurement in blood or serum susceptible to external contamination (6, 13).
Effective risk management requires not only the measurement of phthalate exposure levels but also a clear understanding of how different exposure sources contribute to the overall burden of phthalates. Such knowledge enables prioritization of source-specific interventions, risk communication, and public health policies. Despite the growing body of research, substantial gaps remain in quantifying the relative contributions of different sources to human phthalate exposure in the general population (5, 12). These are further compounded by the diversity of exposure pathways, concurrent exposure to multiple phthalates as mixtures, and their pervasive presence in consumer products (6, 16), suggesting the need for joint evaluation of multiple lifestyle-related exposure sources and diverse phthalate metabolites.
To address these gaps, we employed an outcome-wide analysis to examine how multiple lifestyle determinants contribute to phthalate exposures (17). Using a nationally representative sample of Korean adults, we analyzed dietary consumption and daily product use, which are closely linked to individual lifestyles. Given the widespread, multifactorial, and unavoidable nature of phthalate exposure, we aimed to quantify the relative importance of lifestyle factors in explaining the joint variation in urinary phthalate metabolite mixtures. Our findings are expected to contribute to addressing existing knowledge gaps regarding exposure characterization needed to inform exposure-reduction strategies (18).
2. Materials and methods
2.1. Data and study participants
This study analyzed publicly available data from the Korean National Environmental Health Survey (KoNEHS) Cycle 4, a nationwide biomonitoring program conducted by the National Institute of Environmental Research (NIER) under the Ministry of Environment from 2018 to 2020. Since 2009, KoNEHS has assessed internal exposure to hazardous environmental chemicals in the general Korean population, with each cycle spanning 3 years.
The sampling design of KoNEHS Cycle 4 has been described in detail elsewhere (19). The target population of the adult survey comprised Korean residents aged 19 years and older, and the census tract list of the 2015 Population and Housing Census served as the sampling frame for the survey. A two-stage stratified design was applied to ensure the representativeness of the national population. Primary stratification was based on administrative regions (provinces, metropolitan cities, and municipalities) and coastal areas delineated by topography, with sites hosting air pollution monitoring stations incorporated to evaluate the relationships between ambient air pollution and health outcomes. Secondary stratification further divided these areas by socioeconomic characteristics, including housing type and residential area features, yielding a total of 69 strata. Survey sites were distributed across strata via square-root proportional allocation to the number of households per stratum–an approach selected because it increases the sampling fraction in less populous provinces, thereby improving the precision of the region-specific estimates. On this basis, approximately 15 adults were targeted at each of the 245 survey sites.
Data collection included structured questionnaires on sociodemographic and lifestyle factors, physical examinations, and blood and urine sampling. Following standardized field protocols, trained interviewers and staff conducted face-to-face interviews, physical measurements, and biospecimen collection. The study protocol was approved by the Institutional Review Board of the NIER (NIER-2018-BR-003-02), and all participants provided written informed consent.
Of the 4,239 adults aged ≥19 years, 443 were excluded because urinary creatinine concentrations were outside the acceptable range of 0.3–3.0 g/L, a criterion adopted by the World Health Organization based on the principle that highly dilute or highly concentrated urine specimens are generally unsuitable for biomonitoring (19, 20), and 36 were excluded owing to missing or incomplete covariate data. The final analytical sample comprised 3,760 participants.
2.2. Urinary phthalate metabolite levels
Phthalate exposure was assessed by measuring urinary concentrations of the following metabolites: mono-methyl phthalate (MMP), mono-ethyl phthalate (MEP), mono-n-butyl phthalate (MnBP), mono-benzyl phthalate (MBzP), mono-(2-ethyl-5-hydroxyhexyl) phthalate (MEHHP), mono-(2-ethyl-5-oxohexyl) phthalate (MEOHP), mono-(2-ethyl-5-carboxypentyl) phthalate (MECPP), and mono-(3-carboxypropyl) phthalate (MCPP). Analyses were conducted using an Agilent 1200 series/Agilent 6490 high-performance liquid chromatography–tandem mass spectrometer (HPLC-MS/MS; Agilent, Santa Clara, CA, USA). Sample preparation involved enzymatic deconjugation with β-glucuronidase from Escherichia coli K 12.
Detailed procedures for biological specimen management and quality assurance and control protocols have been described elsewhere (19). Quantification was performed using calibration curves prepared by spiking artificial urine with native analytical standards of the eight target metabolites, with 13C-labeled analogs as internal standards. Calibration standards were prepared at a minimum of four to five concentration levels spanning the range expected in the general population, and calibration curves were expressed as relative response as a function of relative concentration. The calibration ranges were 1.0–120 μg/L for MEHHP and MEOHP, 2.0–240 μg/L for MEP, MnBP, and MECPP, and 0.5–60 μg/L for MMP, MBzP, and MCPP. Calibration standards were prepared and measured for each analytical batch, and calibration curves were accepted only when R2 ≥ 0.995. Calibration verification standards were analyzed after every 20 participant samples and were required to fall within ±15% of their reference values. When this criterion was not met, the calibration curve was reconstructed and the preceding 20 samples were reanalyzed. The limits of detection (LODs) were determined by fortifying pooled or artificial urine at approximately 3–10 times the estimated detection limit and processing seven samples. The LOD was calculated as 3.14 × SD, where SD is the standard deviation of the seven replicate measurements and 3.14 is the t-statistic for six degrees of freedom at a two-sided 98% (equivalently, one-sided 99%) confidence level. The LODs were 0.069 μg/L for MMP, 0.108 μg/L for MEP, 0.084 μg/L for MnBP, 0.030 μg/L for MBzP, 0.069 μg/L for MEHHP, 0.052 μg/L for MEOHP, 0.073 μg/L for MECPP, and 0.072 μg/L for MCPP. Concentrations below the LOD were imputed as LOD/√2 (19). To account for variations in urinary dilution, metabolite concentrations were corrected for creatinine (19). Because these creatinine-corrected concentrations were right-skewed, they were log-transformed prior to statistical analysis.
2.3. Dietary and product-use factors in daily life
Information on dietary consumption and the use of FCMs and consumer products was obtained using comprehensive questionnaires (Table 1).
TABLE 1.
Dietary and daily product-use factors evaluated in the study.
| Category | Subcategory | Specific items |
|---|---|---|
| Dietary consumption | Drinking water | Mineral spring water |
| Tap water | ||
| Filtered water | ||
| Bottled water | ||
| Seafood | Large fish (including tuna) | |
| Fish | ||
| Crustaceans | ||
| Clams | ||
| Seaweed | ||
| Other seafood | ||
| Plant-based food | Fruits | |
| Vegetables | ||
| Kimchi | ||
| Grains | ||
| Beans (including bean products) | ||
| Dairy products and meat | Dairy products | |
| Grilled meat | ||
| Processed food | Cup noodles | |
| Fast food | ||
| Microwaveable packaged food | ||
| Popcorn | ||
| Canned food | ||
| Food contact materials use | Food packaging and containers | Delivery food wrapped in plastic film |
| Plastic containers for storing hot food | ||
| Beverages in plastic pouches | ||
| Beverages in polyethylene terephthalate bottles | ||
| Disposable paper cups | ||
| Non-stick cookware | Coated containers | |
| Coated frying pans | ||
| Coated pots | ||
| Coated rice cookers | ||
| Consumer products use | Personal care products | Toothbrushes |
| Perfumes | ||
| Hair products | ||
| Body wash | ||
| Makeup products | ||
| Nail polish | ||
| Antimicrobial products | ||
| Cleaning products | Air fresheners | |
| Fabric cleaners | ||
| Household cleaners |
Several variables were excluded for methodological reasons. The use of plastic containers for storing hot boiled water and the use of coatings and polishes were excluded due to low prevalence (approximately 5%), which limited their relevance to routine daily use. Ultraviolet-protective sunscreen use was excluded because KoNEHS, conducted throughout the year, records product use over the preceding 3 months, a measure susceptible to misclassification given the strong seasonality of sunscreen use. After these exclusions, 41 dietary and product-use factors were retained for analysis, comprising 22 dietary items (four drinking-water types, six seafood categories, five plant-based food groups, two dairy and meat items, and five processed-food items), nine food-contact material items (five packaging or container types and four non-stick cookware items), and 10 consumer-product items (seven personal care and three cleaning products).
For drinking water, participants were classified as consumers or non-consumers of each water type based on reported use at home or elsewhere. This approach was adopted because the KoNEHS questionnaire records the main type of water consumed separately for home and out-of-home settings but does not distinguish daily-intake volume by location or type. Therefore, the amount of each specific water type consumed could not be quantified. For other dietary and product-use factors, frequency responses were converted to daily equivalents; for instance, a reported frequency of “4–5 times weekly” was converted to 0.64 times per day (= 4.5/7).
2.4. Covariates
Sociodemographic and behavioral factors used as covariates included age, sex, monthly household income, educational attainment, marital status, residential area, smoking status, drinking status, body mass index (BMI), daily indoor time, and recent renovation or repair. Age (years), BMI (kg/m2), and daily indoor time (h) were treated as continuous variables, whereas the remaining factors were categorical.
Sex was classified as male or female. Monthly household income was reported in Korean won (KRW) per month across six ranges: <1 million, 1–<2 million, 2–<3 million, 3–<5 million, 5–<7 million, and ≥7 million. Educational attainment was categorized as below high school, high school, or beyond high school. Marital status was classified as married/living with a partner or never married/divorced/separated/widowed. Residential areas were classified as coastal or non-coastal and included as a covariate to account for broad geographic variations, including environmental background, food origins, water infrastructure, and other regional features not fully represented by the individual-level questionnaire variables alone.
Smoking status was defined as current or non-current, with the latter category including both former- and never-smokers. Drinking status was determined based on binge drinking, defined as consuming ≥7 drinks for men and ≥5 drinks for women at least once a month. Recent renovation or repair was recorded if participants reported undertaking such activities in their primary residence within the past 6 months.
2.5. Statistical analysis
Statistical analyses were conducted using R software (version 4.5.1). Continuous variables were reported as medians with interquartile ranges (IQRs), and sex differences were evaluated using the Mann–Whitney U test. Categorical variables were summarized as frequencies and percentages, with differences by sex evaluated using chi-square tests.
Urinary creatinine-corrected phthalate metabolite concentrations were described using geometric means (GMs), geometric standard deviations (GSDs), and the 25th, 50th, and 75th percentiles. Sex differences in log-transformed values were examined using two-sample t-tests. Pearson’s correlation coefficients were calculated to evaluate correlations among log-transformed values.
To identify latent patterns of dietary consumption and daily product use associated with urinary phthalate metabolite levels, we applied sparse reduced-rank regression (sRRR), a dimension-reduction technique (17). Reduced-rank regression (RRR) derives linear combinations of predictors that explain the maximal variation in the multivariate outcomes (21). By imposing a low-rank structure on the regression coefficient matrix, RRR reduces the number of parameters to be estimated and incorporates outcome correlations to enhance predictive performance. To improve stability under predictor collinearity and facilitate variable selection, we used sRRR, a penalized extension of RRR that shrinks coefficients of less informative predictors toward zero while retaining the predictors that contribute most strongly to the outcome structure (22, 23).
In sRRR, dimensionality reduction is achieved by imposing a low-rank constraint on the coefficient matrix, which is factorized as C = UDVT through singular value decomposition (SVD). Each rank corresponds to a latent pattern, defined as a linear combination of predictors jointly explaining outcome variation. The corresponding singular values, given by the diagonal entries of D, reflect the relative importance of each latent pattern. The left singular vectors (columns of U) and right singular vectors (columns of V) represent predictor and outcome loadings, respectively. The absolute value of each loading indicates the strength of the corresponding variable’s association with the latent pattern, and its sign denotes the direction. Because SVD is subject to sign indeterminacy, simultaneously reversing the signs of corresponding columns of U and V does not alter the reconstructed coefficient matrix or the relationships between variables. Accordingly, signs were oriented so that most outcome loadings were positive, facilitating interpretation. Predictors with loadings sharing the same sign as the outcome loadings indicate positive contributions to the corresponding metabolite concentrations. The sRRR analysis was conducted using the rrpack package (24).
All variables were centered and scaled before modeling. Categorical covariates were converted into ordinal scales by assigning increasing integer values to represent higher category levels. Covariates were adjusted by exempting them from penalties during model fitting (17), ensuring that latent patterns reflected dietary and daily product-use factors after adjustment for confounders. Model selection involved determining the optimal rank based on the Bayesian information criterion. Because sex may influence exposure patterns and metabolite levels, subgroup analyses were stratified by sex.
3. Results
3.1. Study population
The sociodemographic and behavioral characteristics of the 3,760 participants are presented in Table 2. The median age was 54 years (IQR: 41–64 years), and women comprised 53.0% of the sample. The most common monthly household income was 3–5 million KRW. Educational attainment beyond high school was reported by 43.5% of the participants, whereas 74.9% were never married, divorced, separated, or widowed. Men had a significantly higher proportion of individuals with higher educational attainment than women, and women had a significantly higher proportion of married individuals than men (p < 0.01). Most participants (95.0%) resided in non-coastal areas.
TABLE 2.
Characteristics of the study participants.
| Characteristics | Total (n = 3,760) | Male (n = 1,766) | Female (n = 1,994) | P-value |
|---|---|---|---|---|
| Age (years)a | 54 (41–64) | 54 (41–65) | 54 (41–63) | 0.11 |
| Household income (KRW/month)b | 0.19 | |||
| <1 | 514 (13.7) | 228 (12.9) | 286 (14.3) | |
| 1–2 million | 573 (15.2) | 253 (14.3) | 320 (16.0) | |
| 2–3 million | 643 (17.1) | 304 (17.2) | 339 (17.0) | |
| 3–5 million | 1,008 (26.8) | 503 (28.5) | 505 (25.3) | |
| 5–7 million | 625 (16.6) | 286 (16.2) | 339 (17.0) | |
| ≥7 million | 397 (10.6) | 192 (10.9) | 205 (10.3) | |
| Educational attainmentb | <0.01 | |||
| <High school | 981 (26.1) | 367 (20.8) | 614 (30.8) | |
| High school | 1,142 (30.4) | 561 (31.8) | 581 (29.1) | |
| >High school | 1,637 (43.5) | 838 (47.5) | 799 (40.1) | |
| Marital statusb | <0.01 | |||
| Never married, divorced, separated, or widowed | 2,817 (74.9) | 1,369 (77.5) | 1,448 (72.6) | |
| Married, or living with a partner | 943 (25.1) | 397 (22.5) | 546 (27.4) | |
| Residential locationb | 0.98 | |||
| Non-coastal | 3,572 (95.0) | 1,677 (95.0) | 1,895 (95.0) | |
| Coastal | 188 (5.0) | 89 (5.0) | 99 (5.0) | |
| Smokingb | <0.01 | |||
| No | 3,122 (83.0) | 1,189 (67.3) | 1,933 (96.9) | |
| Yes | 638 (17.0) | 577 (32.7) | 61 (3.1) | |
| Drinkingb | <0.01 | |||
| No | 2,798 (74.4) | 1,071 (60.6) | 1,727 (86.6) | |
| Yes | 962 (25.6) | 695 (39.4) | 267 (13.4) | |
| BMI (kg/m2)a | 24.6 (22.4–26.9) | 25.1 (23.2–27.3) | 23.9 (21.7–26.5) | <0.01 |
| Indoor time (h/day)a | 22.0 (20.7–22.9) | 21.6 (20.0–22.7) | 22.1 (21.0–23.0) | <0.01 |
| Renovations or repairsb | 0.69 | |||
| No | 3,189 (84.8) | 1,493 (84.5) | 1,696 (85.1) | |
| Yes | 571 (15.2) | 273 (15.5) | 298 (14.9) |
KRW, Korean won; BMI, body mass index. Data are presented as amedian (25th percentile–75th percentile) or bcount (proportion, %).
Most participants were non-current smokers (83.0%) and did not engage in binge drinking (74.4%). The median BMI was 24.6 kg/m2 (IQR: 22.4–26.9). Participants spent a median of 22.0 h indoors daily (IQR: 20.7–22.9). Recent renovations or repairs were uncommon, with 84.8% reporting none. Men had significantly higher proportions of current smoking and binge drinking and a higher BMI than women (p < 0.01). Women spent more time indoors (p < 0.01).
3.2. Urinary creatinine-corrected phthalate metabolite concentrations
Urinary creatinine-corrected phthalate metabolite concentrations are shown in Table 3. Detection rates were 96.91% (MMP), 88.88% (MEP), 98.30% (MnBP), 97.85% (MBzP), 99.95% (MEHHP), 99.81% (MEOHP), 99.95% (MECPP), and 96.84% (MCPP). The GMs for urinary creatinine-corrected MMP, MEP, MnBP, MBzP, MEHHP, MEOHP, MECPP, and MCPP were 3.41, 4.63, 24.26, 1.10, 14.14, 7.69, 18.27, and 0.46 μg/g creatinine, respectively. Concentrations of all metabolites were higher in women than in men.
TABLE 3.
Distribution of urinary creatinine-corrected phthalate metabolites (μg/g creatinine) in the study population.
| Phthalate metabolites | Total (n = 3,760) | Male (n = 1,766) | Female (n = 1,994) | P-value | ||||
|---|---|---|---|---|---|---|---|---|
| Detection rate | GM (GSD) | Median (Q1–Q3) | GM (GSD) | Median (Q1–Q3) | GM (GSD) | Median (Q1–Q3) | ||
| MMP | 96.91% | 3.41 (2.94) | 3.96 (2.16–6.72) | 2.89 (2.94) | 3.33 (1.84–5.76) | 3.94 (2.88) | 4.62 (2.59–7.52) | <0.01 |
| MEP | 88.88% | 4.63 (9.19) | 3.77 (1.42–15.30) | 3.93 (9.53) | 3.10 (1.17–13.29) | 5.35 (8.81) | 4.49 (1.74–18.01) | <0.01 |
| MnBP | 98.30% | 24.26 (3.03) | 24.93 (15.59–42.28) | 21.19 (3.20) | 21.79 (14.02–35.71) | 27.35 (2.83) | 28.20 (17.48–45.88) | <0.01 |
| MBzP | 97.85% | 1.10 (3.55) | 1.06 (0.50–2.49) | 1.00 (3.52) | 1.00 (0.46–2.19) | 1.19 (3.55) | 1.14 (0.55–2.68) | <0.01 |
| MEHHP | 99.95% | 14.14 (2.03) | 13.98 (9.09–21.98) | 12.65 (1.98) | 12.58 (8.32–19.18) | 15.61 (2.05) | 15.55 (10.12–24.01) | <0.01 |
| MEOHP | 99.81% | 7.69 (2.17) | 7.78 (4.85–12.38) | 6.64 (2.15) | 6.78 (4.29–10.53) | 8.75 (2.15) | 8.88 (5.51–14.01) | <0.01 |
| MECPP | 99.95% | 18.27 (1.99) | 17.96 (11.75–27.96) | 15.87 (1.93) | 15.77 (10.60–23.91) | 20.70 (1.99) | 20.40 (13.41–31.89) | <0.01 |
| MCPP | 96.84% | 0.46 (2.26) | 0.45 (0.28–0.73) | 0.38 (2.22) | 0.38 (0.23–0.61) | 0.54 (2.23) | 0.51 (0.33–0.84) | <0.01 |
MMP, mono-methyl phthalate; MEP, mono-ethyl phthalate; MnBP, mono-n-butyl phthalate; MBzP, mono-benzyl phthalate; MEHHP, mono-(2-ethyl-5-hydroxyhexyl) phthalate; MEOHP, mono-(2-ethyl-5-oxohexyl) phthalate; MECPP, mono-(2-ethyl-5-carboxypentyl) phthalate; MCPP, mono-(3-carboxypropyl) phthalate; GM, geometric mean; GSD, geometric standard deviation; Q1, 25th percentile; Q3, 75th percentile.
Pearson’s correlation coefficients for the log-transformed, urinary creatinine-corrected metabolite concentrations are shown in Supplementary Figure 1. Significant positive correlations were observed among the metabolites. Strong correlations were found among MEHHP, MEOHP, and MECPP, which are metabolites of di(2-ethylhexyl) phthalate (DEHP) (r = 0.89–0.93). In contrast, MEP, a metabolite of diethyl phthalate, showed relatively weaker correlations with other metabolites (r = 0.11–0.22).
3.3. Associations of dietary and daily product-use factors with urinary phthalate metabolite patterns
The sRRR analysis revealed a single underlying pattern characterizing the associations between dietary and daily product-use factors and urinary phthalate metabolite levels. The model selectively retained key variables while effectively excluding 26 lifestyle-related factors that contributed minimally to the emergent structure.
As shown in Figure 1, the pattern was characterized by all eight metabolites exhibiting loadings in the same direction (MMP: +0.33, MEP: +0.10, MnBP: +0.27, MBzP: +0.34, MEHHP: +0.43, MEOHP: +0.44, MECPP: +0.46, and MCPP: +0.31). Several lifestyle factors exhibited loadings in the same direction as the outcome loadings, indicating that they were associated with higher urinary phthalate metabolite mixtures. For dietary consumption, mineral spring water (+0.08), grains (+0.08), and tap water (+0.06) showed substantial loadings in the same direction. For FCMs use, beverages in plastic pouches (+0.08) exhibited substantial loadings in the same direction.
FIGURE 1.
Contributions of dietary and daily product-use factors and urinary creatinine-corrected phthalate metabolites to the latent pattern. The x-axis represents the loading values derived from the sparse reduced-rank regression model. These values correspond to the predictor loadings (left singular vectors) and the response loadings (right singular vectors), where the magnitude indicates the strength of association with the latent pattern and the sign indicates the direction of association. The upper panel (predictor loadings) represents the association between exposure sources and the latent pattern, whereas the lower panel (response loadings) represents the association between urinary creatinine-corrected phthalate metabolites and the latent pattern. Signs were adjusted for interpretability. MMP, mono-methyl phthalate; MEP, mono-ethyl phthalate; MnBP, mono-n-butyl phthalate; MBzP, mono-benzyl phthalate; MEHHP, mono-(2-ethyl-5-hydroxyhexyl) phthalate; MEOHP, mono-(2-ethyl-5-oxohexyl) phthalate; MECPP, mono-(2-ethyl-5-carboxypentyl) phthalate; MCPP, mono-(3-carboxypropyl) phthalate.
Sex-stratified analyses revealed distinct latent patterns (Figure 2). Among men, dietary consumption, including grains, mineral spring water, and tap water, exhibited substantial loadings in the same direction as the eight metabolites. In contrast, among women, the use of FCMs and consumer products, such as beverages in plastic pouches and makeup products, exhibited substantial loadings in the same direction as the eight metabolites.
FIGURE 2.
Contributions of dietary and daily product-use factors and urinary creatinine-corrected phthalate metabolites to the latent pattern in panel (A) males and (B) females. The x-axis represents the loading values derived from the sparse reduced-rank regression model. These values correspond to the predictor loadings (left singular vectors) and the response loadings (right singular vectors), where the magnitude indicates the strength of association with the latent pattern and the sign indicates the direction of association. The upper panels (predictor loadings) represent the association between exposure sources and the latent pattern, whereas the lower panels (response loadings) represent the association between urinary creatinine-corrected phthalate metabolites and the latent pattern. Signs were adjusted for interpretability. MMP, mono-methyl phthalate; MEP, mono-ethyl phthalate; MnBP, mono-n-butyl phthalate; MBzP, mono-benzyl phthalate; MEHHP, mono-(2-ethyl-5-hydroxyhexyl) phthalate; MEOHP, mono-(2-ethyl-5-oxohexyl) phthalate; MECPP, mono-(2-ethyl-5-carboxypentyl) phthalate; MCPP, mono-(3-carboxypropyl) phthalate.
4. Discussion
Using a nationally representative sample of Korean adults and an sRRR approach, this study identified a lifestyle-related pattern associated with higher urinary concentrations of multiple phthalate metabolites. Previous studies have reported phthalate exposure distributions and evaluated individual determinants (25–31); however, these were limited by their focus on isolated sources, separate evaluation of individual metabolites without adequately accounting for simultaneous co-exposure, and susceptibility to false positives from multiple testing (6). Given that phthalate exposure arises from multiple sources simultaneously (1–4, 16), such approaches cannot adequately characterize correlated lifestyle behaviors that drive co-exposure. In contrast, by jointly modeling multiple metabolites, our sRRR approach identified the lifestyle behaviors associated with the shared internal burden of multiple phthalates.
The central finding of this study was that, among routine behaviors, drinking mineral spring water and tap water, consuming grains, and using beverages packaged in plastic pouches were associated with elevated urinary phthalate metabolite levels. These findings align with the established understanding that ingestion is a predominant exposure route (5, 32). Notably, whereas conventional analyses often yield divergent source attributions across metabolites, complicating prioritization, our approach identified a common subset of potential sources that contribute to multiple metabolites. This integrated source profile offers population-specific empirical evidence relevant to public health recommendations and regulatory consideration (4, 5).
Drinking mineral spring and tap water was associated with higher levels of urinary phthalate metabolite mixtures. Contamination can occur at multiple stages within the water supply chain (33, 34). For instance, raw water may be affected by inadequately treated wastewater, poor waste management, or wastewater utilization for agricultural irrigation (34, 35). Furthermore, the supply system can exacerbate contamination (33, 34). Conventional drinking water treatment plants, primarily designed to remove suspended solids, bacteria, and natural organic matter, are ineffective against phthalates (34, 36), and residues often persist (36). Secondary contamination can occur downstream through leaching from plastic pipelines and point-of-use treatment units (34). Notably, in Korea, only DEHP is regulated in drinking water, with a permissible limit of 80 μg/L, higher than the standards set by the U.S. Environmental Protection Agency and the World Health Organization, and DEHP removal rates in Korean water treatment facilities are reported to be ineffective, ranging from 8.2% to 33.4% (36).
Grain consumption, a major component of the Korean diet (37), was also associated with higher levels of urinary phthalate metabolite mixtures. Although we did not measure phthalate concentrations in participants who consumed specific grains, the observed association may be partly explained by contamination arising from agricultural practices, including plastic mulch films that can leach chemicals into the soil and subsequently into crops (38–40). Another possible explanation is environmental deposition from air and dust during processing and migration from packaging materials, such as plastic bags or containers used in storage and transport (41, 42).
The consumption of beverages packaged in plastic pouches was also associated with higher levels of urinary phthalate metabolite mixtures, which may be related to the widespread use of phthalates as plasticizers in food-contact materials and their potential migration into foods and beverages during processing and storage (41, 42). Plastic pouches, which often undergo heat sealing and prolonged storage (43, 44), may have a greater potential for leaching, given that migration rates from FCMs are influenced by temperature, contact duration, and the physicochemical properties of the food matrix (41, 45). Under Korea’s common manufacturing standard for FCMs, DEHP should not be used if migration into food may occur (46). For articles intended for infants and young children, as well as polyvinyl chloride (PVC) materials, regulations also apply to several other phthalate species (46).
Furthermore, the lifestyle factors most prominently associated with urinary phthalate metabolite mixtures differed by sex: among men, these were the consumption of mineral spring water, tap water, and grains, whereas among women, these were the consumption of beverages packaged in plastic pouches and the use of makeup products. The latter likely reflects sex-specific patterns of makeup product use, with women typically using more makeup products than men. These sex-specific differences underscore the need for tailored approaches to exposure reduction that account for differential product-use patterns.
This study has several strengths. First, the application of sRRR enabled simultaneous consideration of diverse lifestyle factors while mitigating multicollinearity across predictors (17, 22), thereby characterizing phthalate exposure as the result of key combined behaviors. Furthermore, by jointly modeling multiple metabolites, this approach offered a more integrated representation of real-world exposure and enabled the identification of a lifestyle pattern contributing to the overall phthalate burden, which provides more interpretable insights for decision-making. Second, the use of a large, nationally representative sample ensured that the findings were robust and highly generalizable to Korean adults. Third, the detailed and well-structured questionnaire of KoNEHS enabled comprehensive assessments of multiple daily lifestyle factors, including diet, FCMs, and consumer products, allowing for a more complete understanding of phthalate exposure pathways.
However, this study has several limitations. First, the cross-sectional design limited our ability to draw causal inferences. Second, exposure assessments based on questionnaire data and urinary biomarkers have inherent limitations. Dietary consumption and daily product use were self-reported and may therefore be subject to recall bias and measurement error. Moreover, questionnaire-based dietary and lifestyle variables used in this study represent habitual long-term behaviors, whereas urinary metabolites capture recent exposure owing to the short biological half-lives of phthalates. This temporal mismatch may lead to exposure misclassification and complicate causal interpretation; accordingly, the findings should be understood as identifying habitual exposure patterns rather than direct short-term dose determinants. Third, potential seasonal variation in phthalate exposure could not be accounted for, as information on the month or season of biospecimen collection was unavailable to adjust for such confounding. Fourth, phthalate concentrations were not directly measured in the drinking water, foods, or products reported by participants. As phthalate concentrations in potential sources may vary substantially, the factors identified as being associated with urinary phthalate mixture exposure represent behaviors or lifestyle characteristics associated with exposure rather than confirmed direct exposure sources. Moreover, without source- or product-specific chemical measurements, the precise origins and pathways of contamination remain undetermined. For instance, it is unclear whether drinking water contamination stems from raw source water, the water supply system, or packaging materials. Future studies should incorporate longitudinal designs, integrate environmental and product-specific measurements with biomonitoring, and identify the origins of contamination to better inform regulatory actions.
The eight metabolites measured in this study characterize exposure to the six principal parent phthalates, although MCPP is not specific to a single parent compound and is commonly attributed to di-n-octyl phthalate (DnOP) (6, 47). MMP, MEP, MnBP, and MBzP are commonly used as biomarkers of dimethyl phthalate (DMP), diethyl phthalate (DEP), di-n-butyl phthalate (DnBP), and benzyl butyl phthalate (BBzP), respectively, whereas MEHHP, MEOHP, and MECPP are oxidative metabolites of DEHP (6, 14). These parent phthalates have distinct uses and exposure profiles. DMP and DEP have been used as solvents or fragrance fixatives in PCPs and other non-PVC applications, whereas DnBP has been used in PCPs, particularly in nail polish, as well as in lacquers, varnishes, adhesives, printing inks, and coatings (41, 48, 49). DEHP has been widely used as a plasticizer in PVC-containing consumer products, including certain FCMs, whereas BBzP spans both domains, being used in paints, adhesives, and sealants and as a plasticizer in PVC articles ranging from vinyl flooring to some FCMs (41, 48, 49). Consistent with these use patterns, consumer products and indoor sources are important contributors to DMP and DEP exposure, with DEP exposure being particularly associated with fragranced PCPs (48, 49), whereas food contributes substantially to DnBP and DEHP exposure, reflecting diffuse environmental contamination together with migration into food from packaging adhesives and printing inks and from processing equipment (41, 48, 49). For BBzP, consumer products and indoor sources as well as food have been identified as important contributors (48, 50). MCPP is a nonspecific oxidative metabolite. In experimental rats, it was formed most abundantly after DnOP exposure but was also detected after exposure to di-isooctyl phthalate, di-isononyl phthalate (DiNP), di-isodecyl phthalate (DiDP), DEHP, and DnBP, and its specificity as a human biomarker remains uncertain (47). DiNP and DiDP have increasingly been used as substitutes for DEHP in plasticized products (48, 50, 51).
Korean regulations cover these compounds unevenly (46). Among the phthalates analyzed in this study, DEHP is the only one with an established national drinking water monitoring criterion (36). Under common manufacturing standards for FCMs, DEHP is prohibited when migration into food may occur (46). DnBP and BBzP are additionally prohibited in the manufacture of utensils, containers, and packages intended for infants and young children, and PVC-specific migration limits apply to DnBP, BBzP, DEHP, DnOP, and the sum of DiNP, and DiDP (46). In cosmetics, the combined concentrations of DnBP, BBzP, and DEHP are limited to 100 μg/g (46). DMP and DEP are not listed in drinking water monitoring, FCM, or cosmetics provisions (36, 46). Regulatory attention is thus concentrated on DEHP, whereas the lifestyle pattern identified here, although led by the DEHP metabolites, was also appreciably associated with the remaining metabolites. This indicates that the internal burden associated with these behaviors extends well beyond the phthalates that current provisions primarily target and suggests that the expansion of the list of regulated phthalate species should be considered (52).
In summary, this study explored the dietary and daily product-use factors associated with exposure to multiple phthalates among Korean adults. Using a predictive and exploratory analytical framework, we provide a population-level characterization of the patterns of dietary consumption and daily product use associated with urinary phthalate metabolite levels. In this nationally representative sample, drinking water, grains, and beverages in plastic pouches made the largest contributions to the lifestyle-related pattern associated with higher urinary phthalate metabolite concentrations. These findings provide empirical grounds for further investigation through prospective studies that integrate biomonitoring with environmental and product-specific measurements to clarify source-specific exposure pathways and inform evidence-based exposure-reduction strategies. If future studies confirm causal relationships between the lifestyle factors identified in our study and phthalate exposure, those findings could help inform policies for reducing cumulative exposure, including regulations for drinking water and FCMs, improvements in water treatment infrastructure, and expansion of the list of regulated phthalate species (52). Even though definitive recommendations await further confirmation, consuming filtered or bottled water instead of tap or mineral spring water and reducing reliance on plastic-based FCMs to minimize leaching may be considered as precautionary approaches to help lower exposure (52–54).
Acknowledgments
This study used Korean National Environmental Health Survey Cycle 4 (2018–2020) data provided by the National Institute of Environmental Research (NIER-2020-01-01- 016).
Funding Statement
The author(s) declared that financial support was received for this work and/or its publication. H-WP was supported as a trainee of the environmental health training program provided by the Environmental Health Center of the Catholic University of Korea, funded by the Ministry of Environment, Republic of Korea (2025).
Footnotes
Edited by: Aleksandra Nikolic, Univerzitet u Sarajevu, Bosnia and Herzegovina
Reviewed by: Jianshe Wang, Yantai University, China
Giulia Galani, Federal University of Santa Catarina, Brazil
Data availability statement
The datasets analyzed in this study are available from the Korean National Environmental Health Survey (KoNEHS) and can be obtained from the National Institute of Environmental Research (NIER) upon reasonable request. Additional information is available at https://www.ehtis.or.kr/. Requests to access these datasets should be directed to knehs@korea.kr.
Ethics statement
The studies involving humans were approved by National Institute of Environmental Research. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
Author contributions
H-WP: Formal analysis, Visualization, Resources, Writing – original draft, Data curation, Conceptualization, Writing – review & editing. DL: Formal analysis, Resources, Software, Data curation, Writing – review & editing, Methodology. BP: Supervision, Conceptualization, Investigation, Validation, Writing – review & editing, Project administration.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that Generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fnut.2026.1929257/full#supplementary-material
References
- 1.Heudorf U, Mersch-Sundermann V, Angerer J. Phthalates: toxicology and exposure. Int J Hyg Environ Health. (2007) 210:623–34. 10.1016/j.ijheh.2007.07.011 [DOI] [PubMed] [Google Scholar]
- 2.Kamrin MA. Phthalate risks, phthalate regulation, and public health: a review. J Toxicol Environ Health B Crit Rev. (2009) 12:157–74. 10.1080/10937400902729226 [DOI] [PubMed] [Google Scholar]
- 3.Net S, Sempere R, Delmont A, Paluselli A, Ouddane B. Occurrence, fate, behavior and ecotoxicological state of phthalates in different environmental matrices. Environ Sci Technol. (2015) 49:4019–35. 10.1021/es505233b [DOI] [PubMed] [Google Scholar]
- 4.Prasad B. Phthalate pollution: environmental fate and cumulative human exposure index using the multivariate analysis approach. Environ Sci Process Impacts. (2021) 23:389–99. 10.1039/d0em00396d [DOI] [PubMed] [Google Scholar]
- 5.Wang Y, Zhu H, Kannan KA. Review of biomonitoring of phthalate exposures. Toxics. (2019) 7:21. 10.3390/toxics7020021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Johns LE, Cooper GS, Galizia A, Meeker JD. Exposure assessment issues in epidemiology studies of phthalates. Environ Int. (2015) 85:27–39. 10.1016/j.envint.2015.08.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Radke EG, Glenn BS, Braun JM, Cooper GS. Phthalate exposure and female reproductive and developmental outcomes: a systematic review of the human epidemiological evidence. Environ Int. (2019) 130:104580. 10.1016/j.envint.2019.02.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Radke EG, Braun JM, Meeker JD, Cooper GS. Phthalate exposure and male reproductive outcomes: a systematic review of the human epidemiological evidence. Environ Int. (2018) 121:764–93. 10.1016/j.envint.2018.07.029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Radke EG, Galizia A, Thayer KA, Cooper GS. Phthalate exposure and metabolic effects: a systematic review of the human epidemiological evidence. Environ Int. (2019) 132:104768. 10.1016/j.envint.2019.04.040 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Jurewicz J, Hanke W. Exposure to phthalates: reproductive outcome and children health. A review of epidemiological studies. Int J Occup Med Environ Health. (2011) 24:115–41. 10.2478/s13382-011-0022-2 [DOI] [PubMed] [Google Scholar]
- 11.Radke EG, Braun JM, Nachman RM, Cooper GS. Phthalate exposure and neurodevelopment: a systematic review and meta-analysis of human epidemiological evidence. Environ Int. (2020) 137:105408. 10.1016/j.envint.2019.105408 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wang Y, Qian H. Phthalates and their impacts on human health. Healthcare (Basel). (2021) 9:603. 10.3390/healthcare9050603 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hogberg J, Hanberg A, Berglund M, Skerfving S, Remberger M, Calafat AM, et al. Phthalate diesters and their metabolites in human breast milk, blood or serum, and urine as biomarkers of exposure in vulnerable populations. Environ Health Perspect. (2008) 116:334–9. 10.1289/ehp.10788 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Frederiksen H, Skakkebaek NE, Andersson AM. Metabolism of phthalates in humans. Mol Nutr Food Res. (2007) 51:899–911. 10.1002/mnfr.200600243 [DOI] [PubMed] [Google Scholar]
- 15.Kurata Y, Shimamura N, Katoh M. Metabolite profiling and identification in human urine after single oral administration of DEHP. J Toxicol Sci. (2012) 37:401–14. 10.2131/jts.37.401 [DOI] [PubMed] [Google Scholar]
- 16.Meeker JD, Sathyanarayana S, Swan SH. Phthalates and other additives in plastics: human exposure and associated health outcomes. Philos Trans R Soc Lond B Biol Sci. (2009) 364:2097–113. 10.1098/rstb.2008.0268 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Anguita-Ruiz A, Amine I, Stratakis N, Maitre L, Julvez J, Urquiza J, et al. Beyond the single-outcome approach: a comparison of outcome-wide analysis methods for exposome research. Environ Int. (2023) 182:108344. 10.1016/j.envint.2023.108344 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Martin L, Zhang Y, First O, Mustieles V, Dodson R, Rosa G, et al. Lifestyle interventions to reduce endocrine-disrupting phthalate and phenol exposures among reproductive age men and women: a review and future steps. Environ Int. (2022) 170:107576. 10.1016/j.envint.2022.107576 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hong S, Kim OJ, Jung SK, Jeon HL, Kim S, Kil J. The exposure status of environmental chemicals in South Korea: the Korean National Environmental Health Survey 2018-2020. Toxics. (2024) 12:829. 10.3390/toxics12110829 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.American Conference of Governmental Industrial Hygienists. Threshold Limit Values for Chemical Substances and Physical Agents & Biological Exposure Indices. Cincinnati, OH: ACGIH Worldwide; (2025). [Google Scholar]
- 21.Kiers HAL, Smilde AK. A comparison of various methods for multivariate regression with highly collinear variables. Stat Method Appl. (2006) 16:193–228. 10.1007/s10260-006-0025-5 [DOI] [Google Scholar]
- 22.Chen LS, Huang JHZ. Sparse reduced-rank regression for simultaneous dimension reduction and variable selection. J Am Stat Assoc. (2012) 107:1533–45. 10.1080/01621459.2012.734178 [DOI] [Google Scholar]
- 23.Yuan M, Ekici A, Lu ZS, Monteiro R. Dimension reduction and coefficient estimation in multivariate linear regression. J R Stat Soc Ser B Stat Methodol. (2007) 69:329–46. 10.1111/j.1467-9868.2007.00591.x [DOI] [Google Scholar]
- 24.Chen K, Wang W. Rrpack: Reduced-Rank Regression. R Package Version 0.1-13 ed; (2022). 10.32614/CRAN.package.rrpack [DOI] [Google Scholar]
- 25.Allotey JA, Boyle M, Sapkota A, Zhu L, Peng RD, Garza MA, et al. Determinants of phthalate exposure among a U.S.-Based Group of Latino Workers. Int J Hyg Environ Health. (2021) 234:113739. 10.1016/j.ijheh.2021.113739 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Bang DY, Kyung M, Kim MJ, Jung BY, Cho MC, Choi SM, et al. Human risk assessment of endocrine-disrupting chemicals derived from plastic food containers. Compr Rev Food Sci Food Saf. (2012) 11:453–70. 10.1111/j.1541-4337.2012.00197.x [DOI] [Google Scholar]
- 27.Kang J, Cho SY, Yoon S. Relationship between the use of plastics in refrigerator food storage and urine phthalate metabolites: the Korean National Environmental Health Survey (KoNEHS) cycle 3. Ann Occup Environ Med. (2023) 35:e53. 10.35371/aoem.2023.35.e53 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kim L, Huh DA, Park K, Lee J, Hwang SH, Choi HJ, et al. Dietary exposure to environmental phenols and phthalates in Korean adults: data analysis of the Korean National Environmental Health Survey (KoNEHS) 2018-2020. Int J Hyg Environ Health. (2025) 267:114597. 10.1016/j.ijheh.2025.114597 [DOI] [PubMed] [Google Scholar]
- 29.Lee G, Lee J, Park NY, Jung S, Lee I, Kwon BR, et al. Exposure to phthalates and alternative plasticizers in patients with impaired kidney function in Korea: temporal trend during 2011-2020 and its association with chronic kidney disease. Environ Sci Technol. (2024) 58:19128–40. 10.1021/acs.est.4c03625 [DOI] [PubMed] [Google Scholar]
- 30.Park C, Hwang M, Baek Y, Jung S, Lee Y, Paek D, et al. Urinary phthalate metabolite and bisphenol a levels in the Korean adult population in association with sociodemographic and behavioral characteristics: Korean National Environmental Health Survey (KoNEHS) 2012-2014. Int J Hyg Environ Health. (2019) 222:903–10. 10.1016/j.ijheh.2019.02.003 [DOI] [PubMed] [Google Scholar]
- 31.Rosen EM, Stevens DR, Ramos AM, McNell EE, Wood ME, Engel SM, et al. Personal care product use patterns in association with phthalate and replacement biomarkers across pregnancy. J Expo Sci Environ Epidemiol. (2024) 34:591–600. 10.1038/s41370-023-00627-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.United States. Environmental Protection Agency. America’s Children and the Environment. 3rd ed. Washington, D.C.: United States Environmental Protection Agency; (2013). iv,304 p. [Google Scholar]
- 33.Sulentic RO, Dumitrascu I, Deziel NC, Gurzau AE. Phthalate exposure from drinking water in Romanian adolescents. Int J Environ Res Public Health. (2018) 15:2109. 10.3390/ijerph15102109 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Wang L, Yuan T, Kwa YC, Jong MC. Chronic exposure to phthalates in drinking water: the implications on public health. Curr Opin Env Sci Hl. (2025) 44:100602. 10.1016/j.coesh.2025.100602 [DOI] [Google Scholar]
- 35.Duenas-Moreno J, Vazquez-Tapia I, Mora A, Cervantes-Aviles P, Mahlknecht J, Capparelli MV, et al. Occurrence, ecological and health risk assessment of phthalates in a polluted urban river used for agricultural land irrigation in Central Mexico. Environ Res. (2024) 240:117454. 10.1016/j.envres.2023.117454 [DOI] [PubMed] [Google Scholar]
- 36.Chung Y, Jang D, Kim H, Oh G, Tak H, Son C, et al. Occurrence and treatment of DEHP in water treatment plants: case study in South Korea. ACS ES T Eng. (2024) 4:580–90. 10.1021/acsestengg.3c00396 [DOI] [Google Scholar]
- 37.Kwon DY. Diet in Korea. In: Meiselman HL. editor. Handbook of Eating and Drinking. Cham: Springer International Publishing; (2020). p. 1435–65. [Google Scholar]
- 38.Feng NX, Pan B, Huang HJ, Huang YT, Lyu H, Xiang L, et al. Uptake, translocation, and biotransformation of phthalate acid esters in crop plants: a comprehensive review. J Hazard Mater. (2025) 489:137580. 10.1016/j.jhazmat.2025.137580 [DOI] [PubMed] [Google Scholar]
- 39.Scopetani C, Pellinen J, Selonen S. Phthalates and other organic chemicals in agricultural soils after use of different types of conventional and biodegradable plastics. Environ Res. (2024) 255:119177. 10.1016/j.envres.2024.119177 [DOI] [PubMed] [Google Scholar]
- 40.Sun Q, Liu C, Zhang X, Wang Z, Guan P, Wang Z, et al. Phthalate Ester (PAES) accumulation in wheat tissues and dynamic changes of rhizosphere microorganisms in the field with plastic-film residue. Sci Total Environ. (2024) 931:172833. 10.1016/j.scitotenv.2024.172833 [DOI] [PubMed] [Google Scholar]
- 41.Giuliani A, Zuccarini M, Cichelli A, Khan H, Reale M. Critical review on the presence of phthalates in food and evidence of their biological impact. Int J Environ Res Public Health. (2020) 17:5655. 10.3390/ijerph17165655 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Serrano SE, Braun J, Trasande L, Dills R, Sathyanarayana S. Phthalates and diet: a review of the food monitoring and epidemiology data. Environ Health. (2014) 13:43. 10.1186/1476-069X-13-43 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Bamps B, Buntinx M, Peeters R. Seal materials in flexible plastic food packaging: a review. Packag Technol Sci. (2023) 36:507–32. 10.1002/pts.2732 [DOI] [Google Scholar]
- 44.Ilhan I, Turan D, Gibson I, ten Klooster R. Understanding the factors affecting the seal integrity in heat sealed flexible food packages: a review. Packag Technol Sci. (2021) 34:321–37. 10.1002/pts.2564 [DOI] [Google Scholar]
- 45.Xu Q, Yin X, Wang M, Wang H, Zhang N, Shen Y, et al. Analysis of phthalate migration from plastic containers to packaged cooking oil and mineral water. J Agric Food Chem. (2010) 58:11311–7. 10.1021/jf102821h [DOI] [PubMed] [Google Scholar]
- 46.Heo E, Lee JY, Park M, Son A. Occurrence, detection technologies, and regulations of phthalic acid esters at atmospheric and aquatic environments. J Korean Soc Environ Eng. (2024) 46:319–34. 10.4491/ksee.2024.46.6.319 [DOI] [Google Scholar]
- 47.Calafat AM, Silva MJ, Reidy JA, Earl Gray L, Samandar E, Preau JL, et al. Mono-(3-carboxypropyl) phthalate, a metabolite of di-N-octyl phthalate. J Toxicol Environ Health A. (2006) 69:215–27. 10.1080/15287390500227381 [DOI] [PubMed] [Google Scholar]
- 48.Wormuth M, Scheringer M, Vollenweider M, Hungerbuhler K. What are the sources of exposure to eight frequently used phthalic acid esters in Europeans? Risk Anal. (2006) 26:803–24. 10.1111/j.1539-6924.2006.00770.x [DOI] [PubMed] [Google Scholar]
- 49.Duty SM, Ackerman RM, Calafat AM, Hauser R. Personal care product use predicts urinary concentrations of some phthalate monoesters. Environ Health Perspect. (2005) 113:1530–5. 10.1289/ehp.8083 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.EFSA Panel on Food Contact Materials, Enzymes and Processing Aids (CEP), Silano V, Barat Baviera JM, Bolognesi C, Chesson A, Cocconcelli PS, et al. Update of the risk assessment of di-butylphthalate (DBP), butyl-benzyl-phthalate (BBP), bis(2-ethylhexyl)phthalate (DEHP), di-isononylphthalate (DINP) and di-isodecylphthalate (DIDP) for use in food contact materials. EFSA J. (2019) 17:e05838. 10.2903/j.efsa.2019.5838 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Lee KJ, Choi K. Environmental occurrence, human exposure, and endocrine disruption of di-iso-nonyl phthalate and di-iso-decyl phthalate: a systematic review. Crit Rev Env Sci Tec. (2024) 54:603–40. 10.1080/10643389.2023.2261815 [DOI] [Google Scholar]
- 52.Sieck NE, Bruening M, van Woerden I, Whisner C, Payne-Sturges DC. Effects of behavioral, clinical, and policy interventions in reducing human exposure to bisphenols and phthalates: a scoping review. Environ Health Perspect. (2024) 132:036001. 10.1289/EHP11760 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Wang J, Weng X, Liu S, Zhang H, Zhu Q, Fu Q, et al. Occurrence, fate, and reduction measures of phthalates in the cooking process: a review. Environ Health (Wash). (2023) 1:300–14. 10.1021/envhealth.3c00091 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Zhang M, Hou J. Phthalate contamination in food: occurrence, health risks, biomarkers for detection, and mitigation strategies to enhance food safety. J Agric Food Chem. (2025) 73:13178–94. 10.1021/acs.jafc.5c02394 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets analyzed in this study are available from the Korean National Environmental Health Survey (KoNEHS) and can be obtained from the National Institute of Environmental Research (NIER) upon reasonable request. Additional information is available at https://www.ehtis.or.kr/. Requests to access these datasets should be directed to knehs@korea.kr.