[Show abstract][Hide abstract] ABSTRACT: Archived samples collected from 1995 to 1997 in the National Human Exposure Assessment Survey (NHEXAS) in U.S. Environmental Protection Agency Region 5 (R5) and the Children's Study (CS) in Minnesota were analyzed for total arsenic, arsenate [As(V)], arsenite, dimethyl arsenic acid (DMA), monomethyl arsenic acid (MMA), arsenobetaine (AsB), and arsenocholine. Samples for the CS included drinking water, urine, hair, and dust; both studies included food (duplicate plate, composited 4-day food samples from participants). Except for AsB and As(V), the levels for As species measured in the food and drinking water samples were very low or nonexistent. The analytical methods used for measuring As species were sensitive to < 1 ppb. During the analysis of food and drinking water samples, chromatographic peaks appeared that contained As, but they did not correspond to those being quantified. Thus, in some samples, the sum of the individual As species levels was less than the total As level measured because the unknown forms of As were not quantified. On the other hand, total As was detectable in almost all samples (> 90%) except for hair (47%), indicating that the analytical method was sufficiently sensitive. Population distributions of As concentrations measured in drinking water, food (duplicate plate), dust, urine, and hair were estimated. Exposures to total As in food for children in the CS were about twice as high as in the general R5 population (medians of 17.5 ppb and 7.72 ppb, respectively). In addition, AsB was the most frequently detected form of As in food eaten by the participants, while As(V) was only rarely detected. Thus, the predominant dietary exposure was from an organic form of As. The major form of As in drinking water was As(V). Spearman (rank) correlations and Pearson (log-concentration scale) correlations between the biomarkers (urine, hair) and the other measures (food, drinking water, dust) and urine versus hair were performed. In the NHEXAS CS, total As and AsB in the food eaten were significantly correlated with their levels in urine. Also, levels of As(V) in drinking water correlated with DMA and MMA in urine. Arsenic levels in dust did not show a relationship with urine or hair levels, and no relationship was observed for food, drinking water, and dust with hair. Urine samples were collected on days 3, 5, and 7 of participants' monitoring periods. Total As levels in urine were significantly associated across the three pairwise combinations--i.e., day 3 versus day 5, day 3 versus day 7, and day 5 versus day 7. Because the half-life of As in the body is approximately 3 days, this suggests that some exposure occurred continually from day to day. This trend was also observed for AsB, suggesting that food is primarily responsible for the continual exposure. DMA and MMA in urine were also significantly correlated but not in all combinations.
Environmental Health Perspectives 02/2006; 114(2):220-7. · 7.26 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Multimedia data from two probability-based exposure studies were investigated in terms of how censoring of nondetects affected estimation of population parameters and associations. Appropriate methods for handling censored below-detection-limit(BDL)values in this context were unclear since sampling weights were involved and since bivariate associations/measures were of interest. Both simple substitution(e.g., using 1/2 or 2/3 of the detection limit(DL)for BDL values)and truncation-based strategies were investigated by creating some artificial DLs and comparing resultant estimates with the original studies'uncensored results. The substitution methods generally outperformed the truncation methods, with the(2/3)DL substitution generally performing best.
[Show abstract][Hide abstract] ABSTRACT: Multimedia data from two probability-based exposure studies were investigated in terms of how missing data and measurement-error imprecision affected estimation of population parameters and associations. Missing data resulted mainly from individuals'refusing to participate in certain measurement activities, rather than from field or laboratory problems; it suggests that future studies should focus on methods for maximizing participation rates. Measurement error variances computed from duplicate-sample data were small relative to the inherent variation in the populations; consequently, adjustments in nonparametric percentile estimates to account for measurement imprecision were small. Methods of adjustment based on lognormality assumptions, however, appeared to perform poorly.
[Show abstract][Hide abstract] ABSTRACT: The Minnesota Children's Pesticide Exposure Study (MNCPES) of the National Human Exposure Assessment Survey (NHEXAS) was conducted in Minnesota to evaluate children's pesticide exposure. This study complements and extends the populations and chemicals included in the NHEXAS Region V study. One of the goals of the study was to test protocols for acquiring exposure measurements and developing databases for use in exposure models and assessments. Analysis of the data quality is one element in assessing the performance of the collection and analysis protocols used in this study. Data quality information must also be available to investigators to guide analysis of the study data. During the planning phase of MNCPES, quality assurance (QA) goals were established for precision, accuracy, and quantification limits. The data quality was assessed against these goals. The assessment is complex. First, data are not available for all analytes and media sampled. In addition, several laboratories were responsible for the analysis of the collected samples. Each laboratory provided data according to their standard operating procedures (SOPs) and protocols. Detection limits were authenticated for each analyte in each sample type. The approach used to calculate detection limits varied across the different analytical methods. The analytical methods for pesticides in air, food, hand rinses, dust wipe and urine were sufficiently sensitive and met the QA goals, with very few exceptions. This was also true for polynuclear aromatic hydrocarbons (PAHs) in air and food. The analytical methods for drinking water and beverages had very low detection limits; however, there were very little measurable data for these samples. The collection and analysis methods for pesticides in surface press samples and soil, and for PAHs in dust wipes were not sufficiently sensitive. Accuracy was assessed primarily as recovery from field controls. The results were good for pesticides and PAHs in air (75-125% recovery). Recovery was lower (<75%) for pesticides in drinking water and beverages. The recovery of pesticides from hand rinses met QA goals (75-100%), but surface press samples showed lower recovery (50-70%). Analysis by gas chromatography-mass spectrometry (GC-MS) did not confirm the presence of atrazine and other pesticides in hand rinse and surface press samples that had been detected by GC-ECD, but instead GC-MS confirmed background interferences. Assessment of the precision of sample collection and analysis is based on the percent relative standard deviation (%RSD) between the results for duplicate samples. Data are available only for pesticides and PAHs in air. Precision was good (<20% RSD) for analytes with measurable data. There were a few analytes with %RSD >20%, but the number of data pairs was very small in these cases. Precision for instrumental analysis of food sample extracts was excellent, with the median %RSD < 20 for all measurable pesticides. The median %RSD for the analysis of replicate aliquots of food from the same sample composite was considerably higher, indicating the potential for inhomogeneity of food homogenates.
Journal of Exposure Analysis and Environmental Epidemiology 11/2003; 13(6):465-79. · 2.72 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The Minnesota Children's Pesticide Exposure Study (MNCPES) provides exposure, environmental, and biologic data relating to multipathway exposures of children for four primary pesticides (chlorpyrifos, malathion, diazinon, and atrazine), 14 secondary pesticides, and 13 polynuclear aromatic hydrocarbons (PAHs). Monitoring was performed on a probability-based sample of 102 children aged 3-12 in Minneapolis/St. Paul and in a nearby rural area (Goodhue and Rice counties). This paper provides estimated distributions of this population's exposures and exposure-related measurements and examines associations among the various measures via rank (Spearman) correlations. In addition, it provides some aggregate and cumulative exposure estimates for pesticides, and compares the relative intakes from inhalation and dietary ingestion. Intakes for the four primary pesticides appeared to come principally from the ingestion rather than the inhalation route; this was clearly true for chlorpyrifos but was less certain for the other three primary pesticides because of their higher degree of nondetects. Solid food rather than beverages was clearly the main contributor to the ingestion intake. Despite the dominance of the ingestion route, the urinary metabolite of chlorpyrifos exhibited a stronger association with the air measurements than with the dietary measures. Personal-air samples exhibited strong rank correlations with indoor air samples for chlorpyrifos, malathion, and diazinon (0.81, 0.51, and 0.62, respectively), while personal-air atrazine levels correlated well with outdoor levels (0.69); personal-air diazinon levels also correlated well with outdoor levels (0.67). For the PAHs, many significant associations were evident among the various air samples and for the air samples with the dust samples, especially for those compounds with consistently high percent measurable values (particularly fluoranthene, phenanthrene, and pyrene).
Journal of Exposure Analysis and Environmental Epidemiology 04/2003; 13(2):100-11. · 2.72 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The ability of questionnaires to predict children's exposure to pesticides was examined as part of the Minnesota Children's Pesticide Exposure Study (MNCPES). The MNCPES focused on a probability sample of 102 children between the ages of 3 and 13 years living in either urban (Minneapolis and St. Paul, MN) or nonurban (Rice and Goodhue Counties in Minnesota) households. Samples were collected in a variety of relevant media (air, food, beverages, tap water, house dust, soil, urine), and chemical analyses emphasized three organophosphate insecticides (chlorpyrifos, diazinon, malathion) and a herbicide (atrazine). Results indicate that the residential pesticide-use questions and overall screening approach used in the MNCPES were ineffective for identifying and oversampling children/households with higher levels of individual target pesticides.
Environmental Health Perspectives 02/2003; 111(1):123-8. · 7.26 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: A statistical mixed effects model was applied to data from a two-phased experiment in which cholinesterase activity (CA) in
red blood cells (RBCs) of rats was monitored following single-and double-gavagead ministrations of the carbamate pesticide,
Aldicarb. Our goals were to develop and estimate a nonlinear mixed effects model for describing the inhibition and recovery
pattern of CA as a continuous function of time and Aldicarb dose, and to use the results to characterize both the typical-animal
CA response and its variability across animals. The Phase 1 experiment involved adult male CD® rats randomly assigned to a
control and two dose groups. Blood samples were taken via jugular cannula prior to exposure and at 11 post-dose time points
(over six hours) for determination of RBC CA. The Phase II experiment was similar, but involved a second administration approximately
4.5 hours after the first. An open, one-compartment model with first-order kinetics appeared to provide an adequate structural
basis for the statistical nonlinear mixed effects model, which was fit to the RBC data of the 26 treatment-group rats from
both phases. For the “typical” animal, the model predicts that about a 20% (40%) maximal reduction in activity will occur
for the low (high) dose group and that recovery to 90% of the baseline level will occur in 105 minutes (179 minutes) for the
low (high) dose group. Based on simulated distributions derived from the model’s variance-covariance matrix of random effects,
we estimated that about 99% of animals are expected to require less than 143 minutes (267 minutes) to recover to 90% of their
baseline level for the low (high) dose group. Because virtually all animals had returned to near pre-exposure, levels by the
time of the second administration, the level of the first dosing had little impact on the CA patterns following the second
Journal of Agricultural Biological and Environmental Statistics 01/2003; 8(4):420-437. · 1.24 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The National Human Exposure Assessment Survey (NHEXAS) Phase I field study conducted in EPA Region 5 (Great Lakes Area) provides extensive exposure data on a representative sample of approximately 250 residents of the region. Associated environmental media and biomarker (blood, urine) concentration data were also obtained for the study participants to aid in understanding of the relationships of exposures to both contaminant pathways and doses. Besides fulfilling the primary NHEXAS objectives, the NHEXAS data provided an opportunity to explore secondary usages, such as examining pathway to route of exposure relationships. A generic type of structural equation model was used to define the anticipated relationships among the various data types for both arsenic (As) and lead (Pb). Since, by design, only a few participants provided data for all sample types, implementing this model required that some media concentrations (outdoor air and soil) be imputed for subjects with missing information by using measurements collected in the same geographic area and time period. The model, and associated pairwise correlations, generally revealed significant but weak associations among the concentrations, exposures, and doses; the strongest associations occurred for the various air measurements (indoor versus outdoor and personal). The generally weak associations were thought to be partly due to the absence of complete coverage of nonresidential environmental media and to nonsynchronization of relevant measurement times and integration periods of collection across the various sample types. In general, relationships between the NHEXAS questionnaire data and the various concentration, exposure, and body-burden measures were also weak. The model results and the modeling exercise suggest several ways for optimizing the design of future exposure assessment studies that are aimed at supporting structural modeling activities.
Journal of Exposure Analysis and Environmental Epidemiology 01/2002; 12(1):29-43. · 2.72 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The Minnesota Children's Pesticide Exposure Study is a probability-based sample of 102 children 3-13 years old who were monitored for commonly used pesticides. During the summer of 1997, first-morning-void urine samples (1-3 per child) were obtained for 88% of study children and analyzed for metabolites of insecticides and herbicides: carbamates and related compounds (1-NAP), atrazine (AM), malathion (MDA), and chlorpyrifos and related compounds (TCPy). TCPy was present in 93% of the samples, whereas 1-NAP, MDA, and AM were detected in 45%, 37%, and 2% of samples, respectively. Measured intrachild means ranged from 1.4 microg/L for MDA to 9.2 microg/L for TCPy, and there was considerable intrachild variability. For children providing three urine samples, geometric mean TCPy levels were greater than the detection limit in 98% of the samples, and nearly half the children had geometric mean 1-NAP and MDA levels greater than the detection limit. Interchild variability was significantly greater than intrachild variability for 1-NAP (p = 0.0037) and TCPy (p < 0.0001). The four metabolites measured were not correlated within urine samples, and children's metabolite levels did not vary systematically by sex, age, race, household income, or putative household pesticide use. On a log scale, mean TCPy levels were significantly higher in urban than in nonurban children (7.2 vs. 4.7 microg/L; p = 0.036). Weighted population mean concentrations were 3.9 [standard error (SE) = 0.7; 95% confidence interval (CI), 2.5, 5.3] microg/L for 1-NAP, 1.7 (SE = 0.3; 95% CI, 1.1, 2.3) microg/L for MDA, and 9.6 (SE = 0.9; 95% CI, 7.8, 11) microg/L for TCPy. The weighted population results estimate the overall mean and variability of metabolite levels for more than 84,000 children in the census tracts sampled. Levels of 1-NAP were lower than reported adult reference range concentrations, whereas TCPy concentrations were substantially higher. Concentrations of MDA were detected more frequently and found at higher levels in children than in a recent nonprobability-based sample of adults. Overall, Minnesota children's TCPy and MDA levels were higher than in recent population-based studies of adults in the United States, but the relative magnitude of intraindividual variability was similar for adults and children.
Environmental Health Perspectives 07/2001; 109(6):583-90. · 7.26 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The distribution of PM(2.5) and manganese (Mn) personal exposures was determined over a 4-month period in Indianapolis, IN, at a time when the gasoline additive, methylcyclopentadienyl manganese tricarbonyl (MMT), was not being used. The data collection period coincided with the data collection period in the Toronto, ON, study, where MMT had been used as a gasoline additive for over 20 years. The inferential or target population consisted of noninstitutionalized residents of the Indianapolis area during the monitoring period (from May 1996 through August 1996) who were at least 16 years old. The survey instruments used in this study (and also in Toronto) included a household screener form (HSF), a study questionnaire (SQ), and a time and activity questionnaire (TAQ). The SQ was administered to elicit information about the participant and his/her activities, occupation, and surroundings that might be relevant to his/her exposure to particles and Mn. In addition to the personal particulate matter (PM) and elemental 3-day monitoring, 240 participants completed a TAQ on a daily basis during the actual monitoring period. Also, a subset of participants had 3-day outdoor and indoor stationary monitoring at their home (approximately 58 observations), and sampling was conducted at a fixed site (approximately thirty-three 3-day observations). The quality of data was assessed and compared to the Toronto study in terms of linearity of measurement, instrument and method sensitivity, measurement biases, and measurement reproducibility. Twenty-six of the sample filters were subjected to two analyses to characterize the within-laboratory component of precision in terms of relative standard deviations (RSDs). The median RSD for Mn was 8.7%, as compared to 2.2% for Toronto. The quality assurance (QA) laboratory exhibited a clear positive bias relative to the primary laboratory for Al and Ca, but no systematic difference was evident for Mn. A high interlaboratory correlation (>0.99) was also attained for Mn. Mean field blank results for PM and Mn were 0.87 microg/m(3) and 0.71 ng/m(3), respectively, which were comparable to the Toronto study. The median RSDs for colocated fixed site and residential samples ranged from 2.2% to 9.0% for PM and from 8.8% to 15.3% for Mn, which were close to those observed in Toronto. For the PM(10), the 90th percentile indoors was 124 microg/m(3) compared with 54 microg/m(3) outdoors. This pattern was even more pronounced for the PM(2.5) data (90th percentiles of 92 microg/m(3) indoors vs 30 microg/m(3) outdoors). Personal PM(2.5) was somewhat higher than the indoor levels, but the percentiles seemed to follow the more highly skewed pattern of the indoor distribution. This difference was largely due to the presence of some smokers in the sample; e.g., exclusion of smokers led to a personal exposure distribution that was more similar to the outdoor distribution. The estimated 90th percentile for the nonsmokers' personal exposures to PM was 43 microg/m(3) compared with 84 microg/m(3) for the overall population. In general, the Indianapolis PM levels of a given type and cut size were somewhat higher than the levels observed in Toronto, e.g., the median and 90th percentile for the personal PM(2.5) exposures were 23 and 85 microg/m(3), respectively, in Indianapolis, while in Toronto, the corresponding percentiles were 19 and 63 microg/m(3). The cities' distributions of the proportion of the PM(10) mass in the 2.5-microm fraction appeared similar for the residential outdoor data (medians of 0.67 and 0.65 for Indianapolis and Toronto, respectively, and 90th percentiles of 0.83 for both cities). For the indoor data, Indianapolis tended to have a larger portion of the mass in the fine fraction (median of 0.80 compared to 0.70 for Toronto). Unlike the PM, the Indianapolis indoor Mn concentration levels were substantially lower than the outdoor levels for both PM sizes, and the median personal levels for Mn in PM(2.5) appeared to fall between the median indoor and outdoor levels. The personal Mn exposure distributions exhibited more skewness than the indoor or outdoor distributions (e.g., the means for the personal, indoor, and outdoor distributions were 7.5, 2.6, and 3.5 ng/m(3), respectively, while the medians were 2.8, 2.2, and 3.2 ng/m(3), respectively). At least a substantial portion of the high end of the personal exposure distribution appeared to be associated with occupational exposures to Mn. In general, the Mn levels in both cut sizes in Indianapolis were approximately 5 ng/m(3) smaller than those in Toronto (e.g., the estimated median and mean levels for personal Mn exposures in PM(2.5) were 2.8 and 7.5 ng/m(3), respectively, in Indianapolis, but were 8.0 and 13.1 ng/m(3) in Toronto). For the nonoccupational subgroups with no exposure to smoking and no subway riders in the two cities, the medians (2.6 ng/m(3) in Indianapolis and 7.8 ng/m(3) in Toronto) were similar to those for the overall populations, but the means were substantially smaller (3.1 ng/m(3) in Indianapolis and 9.2 ng/m(3) in Toronto). The median proportion of Mn in the fine fraction (relative to the PM(10) Mn) for Indianapolis was 0.39 for outdoors and 0.55 for indoors; these ratios were somewhat smaller than the corresponding Toronto medians (0.52 and 0.73). The study found high correlations for particulates and Mn between personal exposures and indoor concentrations, and between outdoor and fixed site concentrations, and low correlations of personal and indoor levels with outdoor and fixed site levels. The pattern was similar to that observed for Toronto, but slightly more pronounced. The PM(10) Mn concentrations (log scale) generally exhibited stronger associations among these various measures than the PM(2.5) Mn concentrations. Comparisons of the particulate distributions between PTEAM (Riverside, CA) and the Indianapolis and Toronto studies were also made.
Journal of Exposure Analysis and Environmental Epidemiology 01/2001; 11(6):423-40. · 2.72 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: A National Human Exposure Assessment Survey (NHEXAS) was performed in U.S. Environmental Protection Agency (U.S. EPA) Region V, providing population-based exposure distribution data for metals and volatile organic chemicals (VOCs) in personal, indoor, and outdoor air, drinking water, beverages, food, dust, soil, blood, and urine. One of the principal objectives of NHEXAS was the testing of protocols for acquiring multimedia exposure measurements and developing databases for use in exposure models and assessments. Analysis of the data quality is one element in assessing the performance of the collection and analysis protocols used in NHEXAS. In addition, investigators must have data quality information available to guide their analyses of the study data. At the beginning of the program quality assurance (QA) goals were established for precision, accuracy, and method quantification limits. The assessment of data quality was complicated. First, quality control (QC) data were not available for all analytes and media sampled, because some of the QC data, e.g., precision of duplicate sample analysis, could be derived only if the analyte was present in the media sampled in at least four pairs of sample duplicates. Furthermore, several laboratories were responsible for the analysis of the collected samples. Each laboratory provided QC data according to their protocols and standard operating procedures (SOPs). Detection limits were established for each analyte in each sample type. The calculation of the method detection limits (MDLs) was different for each analytical method. The analytical methods for metals had adequate sensitivity for arsenic, lead, and cadmium in most media but not for chromium. The QA goals for arsenic and lead were met for all media except arsenic in dust and lead in air. The analytical methods for VOCs in air, water, and blood were sufficiently sensitive and met the QA goals, with very few exceptions. Accuracy was assessed as recovery from field controls. The results were excellent (> or = 98%) for metals in drinking water and acceptable (> or = 75%) for all VOCs except o-xylene in air. The recovery of VOCs from drinking water was lower, with all analytes except toluene (98%) in the 60-85% recovery range. The recovery of VOCs from drinking water also decreased when comparing holding times of < 8 and > 8 days. Assessment of the precision of sample collection and analysis was based on the percent relative standard deviation (% RSD) between the results for duplicate samples. In general, the number of duplicate samples (i.e., sample pairs) with measurable data were too few to assess the precision for cadmium and chromium in the various media. For arsenic and lead, the precision was excellent for indoor, and outdoor air (< 10% RSD) and, although not meeting QA goals, it was acceptable for arsenic in urine and lead in blood, but showed much higher variability in dust. There were no data available for metals in water and food to assess the precision of collection and analysis.
Journal of Exposure Analysis and Environmental Epidemiology 01/2001; 11(2):140-54. · 2.72 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The purpose of this manuscript is to describe the practical strategies developed for the implementation of the Minnesota Children's Pesticide Exposure Study (MNCPES), which is one of the first probability-based samples of multi-pathway and multi-pesticide exposures in children. The primary objective of MNCPES was to characterize children's exposure to selected pesticides through a combination of questionnaires, personal exposure measurements (i.e., air, duplicate diet, hand rinse), and complementary monitoring of biological samples (i.e., pesticide metabolites in urine), environmental samples (i.e., residential indoor/outdoor air, drinking water, dust on residential surfaces, soil), and children's activity patterns. A cross-sectional design employing a stratified random sample was used to identify homes with age-eligible children and screen residences to facilitate oversampling of households with higher potential exposures. Numerous techniques were employed in the study, including in-person contact by locally based interviewers, brief and highly focused home visits, graduated subject incentives, and training of parents and children to assist in sample collection. It is not feasible to quantify increases in rates of subject recruitment, retention, or compliance that resulted from the techniques employed in this study. Nevertheless, results indicate that the total package of implemented procedures was instrumental in obtaining a high percentage of valid samples for targeted households and environmental media.Keywords: children's health, exposure assessment, National Human Exposure Assessment Survey (NHEXAS)
Journal of Exposure Analysis and Environmental Epidemiology 10/2000; · 2.72 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Methylcyclopentadienyl manganese tricarbonyl (MMT) is a manganese-based gasoline additive used to enhance automobile performance. MMT has been used in Canadian gasoline for about 20yr. Because of the potential for increased levels of Mn in particulate matter resulting from automotive exhausts, a large-scale population-based exposure study (∼1000 participant periods) was conducted in Toronto, Canada, to estimate the distribution of 3-day average personal exposures to particulate matter (PM2.5 and PM10) and Mn. A stratified, three-stage, two-phase probability, longitudinal sample design of the metropolitan population was employed. Residential indoor and outdoor, and ambient levels (at a fixed site and on a roof) of PM2.5, PM10, and Mn were also measured. Supplementary data on traffic counts, meteorology, MMT levels in gasoline, personal occupations, and activities (e.g. amount of vehicular usage) were collected. Overall precision (%RSD) for analysis of duplicate co-located samples ranged from 2.5 to 5.0% for particulate matter and 3.1 to 5.5% for Mn. The detection limits were 1.47 and 3.45μgm-3 for the PM10 and PM2.5 fractions, respectively, and 5.50 and 1.83ngm-3 for Mn in PM10 and PM2.5, respectively. These low detection limits permitted the reporting of concentrations for >98% of the samples. For PM10, the personal particulate matter levels (median 48.5μgm-3) were much higher than either indoor (23.1μgm-3) or outdoor levels (23.6μgm-3). The median levels for PM2.5 for personal, indoor, and outdoor were 28.4, 15.4 and 13.2μgm-3, respectively. The correlation between PM2.5 personal exposures and indoor concentrations was high (0.79), while correlations between personal and the outdoor, fixed site and roof site were low (0.16–0.27). Indoor Mn concentration distributions (in PM2.5 and PM10), unlike particulate matter, exhibited much lower and less variable levels that the corresponding outdoor data. The median personal exposure was 8.0ngm-3, compared with 4.7 and 8.6ngm-3, respectively, for the indoor and outdoor distributions. The highest correlations occurred for personal vs indoor data (0.56) and for outdoor vs roof site data (0.66), and vs fixed site data (0.56). The concentration of Mn in particulate matter, expressed in ppm (w/w), revealed that the fixed site was the highest, followed by the roof site, outdoor, indoor, and personal. The personal and indoor data showed a statistically significant correlation (0.68) while all other correlations between personal or indoor data and outdoor or fixed-site data were quite small. The low correlations of personal and indoor levels with outdoor levels suggest that different sources in the indoor and outdoor microenvironments produce particle matter with dissimilar composition. The correlation results indicate that neither the roof- nor fixed-site concentrations can adequately predict personal particulate matter or Mn exposures.
[Show abstract][Hide abstract] ABSTRACT: For the Phase I field test of the National Human Exposure Assessment Survey (NHEXAS) in U.S. Environmental Protection Agency (EPA) Region 5, this paper presents the survey sampling design, the response rates achieved, and the sample weighting procedure implemented to compensate for unit nonresponse. To enable statistically defensible inferences to the entire region, a sample of about 250 members of the household population in EPA Region 5 was selected using a stratified multistage probability-based survey sampling design. Sample selection proceeded in four nested stages: (1) sample counties; (2) area segments based on Census blocks within sample counties; (3) housing units (HUs) within sample segments; and (4) individual participants within sample households. Each fourth-stage sample member was asked to participate in 6 days of exposure monitoring. A subsample of participants was asked to participate in two rounds of longitudinal follow-up data collection. Approximately 70% of all sample households participated in household screening interviews in which rosters of household members were developed. Over 70% of the sample subjects selected from these households completed the Baseline Questionnaire regarding their demographic characteristics and potential for exposures. And, over 75% of these sample members went on to complete at least the core environmental monitoring, including personal exposures to volatile organic compounds (VOCs) and tap water concentrations of metals. The sample weighting procedures used the data collected in the screening interviews for all household members to fit logistic models for nonresponse in the later phases of the study. Moreover, the statistical analysis weights were poststratified to 1994 State population projections obtained from the Bureau of the Census to ensure consistency with other statistics for the Region.
Journal of Exposure Analysis and Environmental Epidemiology 01/1999; 9(5):369-80. · 2.72 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The National Human Exposure Assessment Survey (NHEXAS) Phase I field study conducted in EPA Region 5 provides extensive exposure data on approximately 250 study participants selected via probability sampling. Associated environmental media and biomarker (blood, urine) concentration data were also obtained to aid in the understanding of relationships of the exposures to both contaminant sources and doses. Distributional parameters for arsenic (As), lead (Pb), and four volatile organic compounds (VOCs)--benzene, chloroform, tetrachloroethylene, and trichloroethylene--were estimated for each of the relevant media using weighted data analysis techniques. Inter-media associations were investigated through correlation analysis, and longitudinal correlations and models were used to investigate longitudinal patterns. Solid food appeared to be a major contributor to urine As levels, while Pb levels in household (HH) dust, personal air, and beverages all were significantly associated with blood Pb levels. Relatively high (>0.50) longitudinal correlations were observed for tap water Pb and As, as compared to only moderate longitudinal correlations for the personal air VOCs.
Journal of Exposure Analysis and Environmental Epidemiology 01/1999; 9(5):381-92. · 2.72 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The National Human Exposure Assessment Survey (NHEXAS) provides a rich database of exposure and environmental measurements for persons living in EPA Region V (Great Lakes). Demographics (e.g., gender, minority status, age, income, and year home built) between U.S. Census data and the overall Region V sample were compared and showed good agreement. This representative sample was used to conduct an exploratory investigation of selected subpopulations that might exhibit higher exposures, on average, to volatile organic chemicals (VOCs) such as benzene, chloroform, etc.; inspirable particles; and metals (e.g., lead, arsenic, etc.) than the general population in Region V. Means and medians were the metrics of comparison. Personal air exposures for p-dichlorobenzene were significantly higher in adults (> 21 years old) than in children (1-14 years old) (median: below detection limit vs. 0.87 microgram/m3, p = 0.0005), while a trend toward higher levels of arsenic exposure in children than adults was observed (median: 1.13 vs. 0.8 ng/m3, p = 0.083). A trend towards higher personal air exposure to lead for minorities vs. nonminorities was evident (median: 26 vs. 12 ng/m3, p = 0.066), but personal exposure to 1,1,1-trichloroethane tended to be higher in nonminorities (mean: 22 vs. 3.7 micrograms/m3, p = 0.081). Dietary exposure to arsenic from solid foods was significantly higher in adults than children (mean: 21 vs. 7.1 micrograms/kg, p = 0.0001; median: 10 vs. 5.6 micrograms/kg, p = < 0.001), and for cadmium it was higher for nonminorities than minorities (median: 18 vs. 15 micrograms/kg, p = 0.023). In contrast, the dietary intake for arsenic, which is based on body weight, was significantly higher in children than adults (mean: 1.72 vs 1.38 micrograms/kg-1 day-1, p = < 0.0001; median 1.02 vs. 0.83, p = < 0.0001). Dietary exposure to chromium in beverages tended to be higher in minorities than nonminorities (median: 16 vs. 13 micrograms/kg, p = 0.017). Lead levels in surface dust wipes tended to increase with the age of the home (mean: 128 micrograms/g in homes built since 1980 to 1075 micrograms/g in homes built before 1940; median: 93 to 236 micrograms/g, respectively). These findings were consistent with the observation that for persons living in older homes personal air exposures to lead are elevated compared to persons living in recently built homes (median: 12 ng/m3 in homes built since 1980, vs. 24 ng/m3 in homes built before 1940, p = 0.043).
Journal of Exposure Analysis and Environmental Epidemiology 01/1999; 9(1):49-55. · 2.72 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Methylcyclopentadienyl manganese tricarbonyl (MMT), a manganese-based gasoline additive, has been used in Canadian gasoline for about 20 yr. Because MMT potentially increases manganese levels in particulate matter resulting from automotive exhausts, a population-based study conducted in Toronto, Canada assessed the levels of personal manganese exposures. Integrated 3-day particulate matter (PM2.5) exposure measurements, obtained for 922 participant periods over the course of a year (September 1995–August 1996), were analyzed for several constituent elements, including Mn. The 922 measurements included 542 participants who provided a single 3-day observation plus 190 participants who provided two observations (in two different months). In addition to characterizing the distributions of 3-day average exposures, which can be estimated directly from the data, including the second observation for some participants enabled us to use a model-based approach to estimate the long-term (i.e. annual) exposure distributions for PM2.5 mass and Mn. The model assumes that individuals’ 3-day average exposure measurements within a given month are lognormally distributed and that the correlation between 3-day log-scale measurements k months apart (after seasonal adjustment) depends only on the lag time, k, and not on the time of year. The approach produces a set of simulated annual exposures from which an annual distribution can be inferred using estimated correlations and monthly means and variances (log scale) as model inputs. The model appeared to perform reasonably well for the overall population distribution of PM2.5 exposures (mean=28 μg m-3). For example, the model predicted the 95th percentile of the annual distribution to be 62.9 μg m-3 while the corresponding percentile estimated for the 3-day data was 86.6 μg m-3. The assumptions of the model did not appear to hold for the overall population of Mn exposures (mean=13.1 ng m-3). Since the population included persons who were potentially occupationally exposed to Mn (in non-vehicle-related jobs), we used responses to questionnaire items to form a subgroup consisting of non-occupationally exposed participants (671 participant periods), for which the model assumptions did appear to hold. For that subpopulation (mean=9.2 ng m-3), the model-predicted 95th percentile of the annual Mn distribution was 16.3-ng m-3, compared with 21.1 ng m-3 estimated for the 3-day data.
[Show abstract][Hide abstract] ABSTRACT: Particle concentrations were measured for a probability-based sample of 178 nonsmoking individuals aged 10 or older residing in Riverside, California, in the fall of 1990. Two 12-hr personal-exposure PM10 samples were obtained for each participant, along with fixed-location PM10 and PM2.5 indoor and outdoor air samples at their residences. The particle samples were also analyzed via X-ray fluorescence (XRF) to determine elemental concentrations for selected elements, including some toxic metals, crustal elements, and combustion- and industrial-source related elements. About 25% of the target population was estimated to have 24-hr personal exposures to PM10 that exceeded the national ambient air concentration standard of 150 micrograms/m3. The daytime personal exposure levels (median of 130 micrograms/m3) tended to exceed both indoor and outdoor levels by about 50%; nighttime personal exposure levels were lower and were only slightly higher than nighttime indoor levels. Several possible reasons for the elevated daytime personal PM10 levels (relative to indoor levels) are considered. Certain activities such as house cleaning and smoking were found to be associated with elevated personal exposure levels.
Journal of Exposure Analysis and Environmental Epidemiology 01/1993; 3(2):227-50. · 2.72 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The Particle Total Exposure Assessment Methodology (PTEAM) study provided the opportunity to test methodologies for measuring personal and microenvironmental PM10 and PM2.5 concentrations in a full-scale probability-based sample of 178 persons and homes in Riverside, California during the fall of 1990. The purpose of the study was to estimate frequency distributions of exposure to PM10, PM2.5, and selected elements in an urban population. Quality control samples and analyses were used to evaluate method performance. These included collocated sample collection, field and lab blank filters, sampler and balance field audits, and intra- and interlaboratory replicate elemental analyses. A portion of the study was also designed to include side-by-side operation of the personal and microenvironmental samplers with reference method (high-volume and dichotomous) samplers to provide an evaluation of method comparability. Over 95% of the approximately 2,900 scheduled samples were collected and analyzed, with very few losses due to equipment failure. The method limit of detection for the personal and microenvironmental monitor PM10 sampling was 8 micrograms/m3. Mean relative standard deviations (RSDs) of 2% to 8% were obtained for collocated personal and microenvironmental samples. Sampler flow rates were within the +/- 10% accuracy criterion during two field audits. Balances operated in a specially designed mobile laboratory were within specified tolerances for precision (+/- 4 micrograms) and accuracy (+/- 50 micrograms). Elemental analysis accuracy was measured with standard reference materials with biases ranging from 2% to 7%. Measurement precision for most elements ranged from 2.5% to 25% mean RSD. Personal and microenvironmental samplers gave median PM10 concentrations that were approximately 9% higher than the dichotomous sampler and 16% higher than the high-volume sampler across 96 monitoring periods at a fixed outdoor location.
Journal of Exposure Analysis and Environmental Epidemiology 01/1993; 3(2):203-26. · 2.72 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The goal of the study was to estimate the frequency distribution of exposure of an urban population to inhalable particles (less than 10 micrometers in diameter). A probability sampling design was used to select 178 nonsmoking residents aged 10 or above in Riverside, CA. Each person was monitored for two consecutive 12-hour periods during the fall of 1990. Concurrent samples were collected in the home and immediately outside the home. The indoor-outdoor samples included both inhalable particles (10 micrometers) and the fine fraction (2.5 micrometers). A central site operated for all 48 days of the study, collecting 96 12-hour samples using reference samplers (dichotomous and hi-vol) side by side with the personal and indoor-outdoor monitors.