2,4-DCP (81.2–90.5%) and 2,5-DCP (97.4–98.3%) were frequently detected during the four survey periods examined, with GMs ranging from 0.803 μg/L to 1.04 μg/L for 2,4-DCP, and 6.1 μg/L to 12.9 μg/L for 2,5-DCP (). Distributions according to NHANES cycle, age, sex, and race/ethnicity are provided in Supplemental Material, Tables S1 and S2 (for 2,5-DCP and 2,4-DCP, respectively) and Tables S3 and S4 (for creatinine-corrected DCPs). The urinary concentrations of 2,5-DCP and 2,4-DCP among NHANES 2003–2010 participants (n = 10,426) were highly correlated (Pearson correlation coefficient (r) = 0.95, p < 0.0001); by contrast, the correlation between the urinary concentrations of 2,4-DCP and triclosan, another chlorinated chemical monitored in NHANES, was relatively low (r = 0.35, p < 0.0001). Furthermore, 1,4-D blood concentrations and urinary DCP concentrations were correlated among the 1,381 NHANES 2003–2004 and 2005–2006 adult participants with available data (2,5-DCP: r = 0.74; 2,4-DCP: r = 0.69, both p < 0.0001).
GMs and selected percentiles (95% CIs) of urinary concentrations (μg/L), and detection frequency of 2,4-DCP and 2,5-DCP in the U.S. population ≥ 6 years of age, NHANES 2003–2010.
Estimated GM concentrations of DCPs and corresponding 95% CIs (derived using weighted values for other model covariates) are listed in , and p-values for pair-wise comparisons between categories of each model predictor are provided in Supplemental Material, Table S5. Based on models adjusted for survey cycle and race/ethnicity × family income, GM concentrations of both 2,4-DCP and 2,5-DCP were significantly higher among 6- to 11-year-old children and older adults (≥ 60 years of age) than among 12- to 19- or 20- to 59-year-old participants (all p < 0.01). Adjusted GM concentrations of 2,5-DCP declined from 12.27 μg/L (NHANES 2003–2004) to 6.07 μg/L (NHANES 2009–2010) (, ). 2,4-DCP concentrations also decreased over time, although the decline was not monotonic and not as pronounced.
Adjusted GM concentrations (95% CIs) of 2,4-DCP and 2,5-DCP (μg/L) according to age, NHANES cycle, and race/ethnicity × family income, NHANES 2003–2010.
Figure 1 Temporal trend for adjusted GM urinary concentrations (μg/L) of 2,5-DCP (A) and 2,4-DCP (B). Adjusted GMs were estimated using the regression equation with the intercept and regression coefficient for a given level of the categorical variable (more ...)
There was little variation in adjusted DCP concentrations among non-Hispanic whites according to income (, ; see also Supplemental Material, Table S5). Among Mexican Americans, DCPs decreased monotonically with increasing family income, from 31.95 μg/L (95% CI: 22.54, 45.3) for those below poverty to 9.28 μg/L (95% CI: 6.12, 14.08) in the highest-income group for 2,5-DCP, and from 1.95 μg/L (95% CI: 1.86, 2.04) to 0.95 μg/L (95% CI: 0.83, 1.07) for 2,4-DCP (both p < 0.001). DCP concentrations among non-Hispanic blacks also were lowest among those with the highest incomes, but the highest concentrations were in the low-income group (). DCP concentrations were consistently lower in non-Hispanic whites than in Mexican Americans or non-Hispanic blacks, regardless of income, although the differences were not as pronounced among the high-income participants and no longer statistically significant for Mexican Americans and non-Hispanic whites. Concentrations of both DCPs were higher among Mexican Americans than non-Hispanic blacks among those below the poverty line, but higher among non-Hispanic blacks than Mexican Americans in all other income groups. Non-Hispanic whites lived in single-unit houses more often (75.2%) than Mexican Americans (60.5%) or non-Hispanic blacks (52.5%) (see Supplemental Material, Table S6). The percentage of people living in apartments or attached family housing decreased as family income increased, from 51.3% to 28.8% for non-Hispanic blacks, and from 32.8% to 12.4% for Mexican Americans (see Supplemental Material, Table S7). Among those with high incomes, fewer Mexican Americans (12.4%) than non-Hispanic blacks (28.8%) reported living in apartments or attached family houses.
Figure 2 Adjusted GM urinary concentrations (μg/L) of 2,5-DCP (A) and 2,4-DCP (B) by family income categories. Below poverty, PIR < 1; low, PIR = 1–1.93; middle, PIR = 1.93–3.71; high, PIR (more ...)
Compared with non-Hispanic blacks, non-Hispanic whites and Mexican Americans were 1.95 times (95% CI: 1.44, 2.66) and 1.66 times (95% CI: 1.01, 2.74) more likely to live in single-family houses, respectively, based on logistic regression models adjusted for income. Compared with participants living below the poverty level, the odds of living in a single-family house increased with income [odds ratios (ORs) (95% CIs) were 1.62 (1.12, 2.35); 3.26 (2.18, 4.89); and 5.27 (3.5, 7.92) for low-, middle-, and high-income categories, respectively]. However, associations between living in a single-family home and race/ethnicity did not differ significantly according to income, and vice versa (data not shown).
The odds of having urinary DCP concentrations above the 95th percentile in NHANES 2003–2004 and 2005–2006 were significantly associated with type of housing (single vs. multiunit) and race/ethnicity, but not with income. After adjusting for race/ethnicity, participants living in multiunit housing were about 1.5 times more likely than participants living in single-family houses to have urinary concentrations of both DCPs above the 95th percentile (). Compared with non-Hispanic whites, Mexican Americans, non-Hispanic blacks, and persons of other races were 1.5–4.7 and 2.0–6.0 times more likely to have 2,4-DCP and 2,5-DCP concentrations above the 95th percentile, respectively ().
Adjusted ORs (95% CIs) of the likelihood of participants having 2,4-DCP and 2,5-DCP urinary concentrations above the 95th percentile (NHANES 2003–2004 and NHANES 2005–2006).