Contact Us Search: All EPA This Area

• You are here: EPA Home
• Research & Development
• National Center for Environmental Research
• Research Project Search
• Community Based Intervention to Reduce Pesticide Exposures to Young Children
• 2002 Annual Report

Research Project Search

 Enter Search Term:
   
 NCER Advanced Search

2002 Progress Report: Community Based Intervention to Reduce Pesticide Exposures to Young Children

EPA Grant Number: R826709C001
Subproject: this is subproject number 001 , established and managed by the Center Director under grant R826709
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
Center: CECEHDPR - University of California at Berkeley
Center Director: Eskenazi, Brenda
Title: Community Based Intervention to Reduce Pesticide Exposures to Young Children
Investigators: Eskenazi, Brenda
Institution:
EPA Project Officer: Fields, Nigel
Project Period: January 1, 1998 through January 1, 2002
Project Period Covered by this Report: January 1, 2001 through January 1, 2002
Project Amount: Refer to main center abstract for funding details.
RFA: Centers for Children's Environmental Health and Disease Prevention Research (1998)
Research Category: Children's Health , Health Effects

Description:

Objective:

The objectives are to: (1) determine whether exposure to pesticides is associated with poorer neurodevelopmental functioning and behavioral problems, delayed growth, and increased respiratory symptoms and disease; and (2) determine whether exposure to environmental allergens and respiratory irritants is associated with increased respiratory symptoms and disease.

Progress Summary:

Quantitative Pesticide Exposure Assessment (QEA)

Sampling protocols and study instruments were developed in collaboration with Don Whitaker of the U.S. Environmental Protection Agency (EPA) to conduct intensive environmental sampling for pesticides in the homes of 20 children. Media sampled included air, food, dust, surfaces, spot urine, and overnight diaper urine. Sampling was completed at the end of September. Samples will be analyzed by Battelle Laboratories, EPA in Cincinnati, and the Centers for Disease Control and Prevention (CDC) beginning in October 2002. Information on Battelle method development progress is presented below.

Method Development and Validation for Analytes in House Dust and Other Media

Battelle Laboratories has successfully developed GC/MS and LC/MS/MS methods to measure all of the CHAMACOS target pesticides. This accomplishment resolves earlier problems with methods that were unable to measure key pesticides with heavy use in the Salinas Valley. GC/MS laboratory methods utilize 1:1 hexane:acetone, solid phase extraction (SPE) cleanup on a silica cartridge. Laboratory QA/QC includes an internal standard and a 7-point calibration curve for each analyte. For the validation of the analysis method, four 0.5 g aliquots of a reference house dust were analyzed at each of the following fortification levels: 0 ng added, 50 ng added, 250 ng added, and 2500 ng added. LC/MS/MS methods were developed for thermally unstable analytes, including oxydemeton-methyl, a highly toxic organophosphate (OP) that is widely used in the Salinas Valley.

Overall, the recovery (accuracy) and precision of the analytical methods are within acceptable limits that may be the case here.

Application of Method to Pyrethroid Insecticides. On a subsequent method development and validation program for EPA, the analytical method that was developed was modified slightly at the extraction step for analysis of air polyurethane foam, surface wipes, soil, and cotton socks for diverse organophosphate and pyrethroid insecticides. On this EPA program, we determined that the above-described GC/MS-based method provides recoveries of about 65-110 percent for pyrethroids such as allethrin, bifenthrin, cyfluthrin, cypermethrin, deltamethrin, esfenvalerate, lambda-cyhalothrin, natural pyrethrin I, tetramethrin, resmethrin, sumithrin and tralomethrin. Due to increased interest in exposures of children to pyrethroid pesticides in the home, as well as to tracked-in pesticides from farm fields, we chose to acquire data on the pyrethroid pesticides in these dust samples, as well as the contractually-agreed upon analytes listed in Table 1. We will explore options to obtain final concentration results for these widely used pesticides.

Table 1. Analyte Detection in First 164 Dust Samples

Method Testing for Analytes in Cotton Union Suit and Sock Matrices. In support of the QEA, we are testing cotton matrices for child pesticide transfer studies. Candidate extraction techniques for cotton union suit sections and socks included accelerated solvent extraction (at elevated temperature and pressure in a stainless steel cell), shake extraction in an Erlenmeyer flask, and Soxhlet extraction. Preliminary data indicate the cleaning and extraction procedures will work for these matrices.

Results. The frequency of analyte detection in the first 164 samples, the mean and median values for selected analytes in these dust samples, and the range of levels found are listed in Table 1.

The analyses of the dust samples will be completed in October. The data will be reduced to Excel spreadsheets, and the senior-scientist reviews will be completed. The initial shipping logs will be compared with final sample data to check for entry errors and lost samples. The QA Office will review the data before it is transmitted to the University of California (UC) at Berkeley in October 2002.

A final validation exercise will be conducted for recovery of analytes from the union suits. This will entail multiple analyses at lower and higher fortification levels than that used for the initial evaluation of the method. The union suits and socks will be extracted and analyzed using GC/MS, and the data will be reviewed and transmitted to UC Berkeley using the protocols developed and implemented for the dust samples.

Vehicle Dust Sub-Study

In collaboration with Ray Chavira (University of California at Los Angeles doctoral student and EPA Region 9 employee) we assessed dust pesticide levels in motor vehicles used by both farm and non-farm worker families in the Salinas Valley. Dust was collected from 36 vehicles. The pesticides dacthal (herbicide) and cis-permethrin (insecticide) were found in the dust of all vehicles, while residues of the pesticides methimidophos, disulfoton, dichlorvos, naled, fonofos, dimethoate, phosmet, and chlorpyrifos oxon were not detected in any vehicles. Farm worker vehicles had higher iprodione levels than non-farm worker vehicles (p = 0.008). (There are no iprodione-containing pesticide products registered for residential uses in California.) Geographically, vehicles from participants residing in rural agricultural areas had higher dacthal and bensulide levels (p < 0.001) than urban vehicles. However, vehicles from the more urban Salinas area had higher trans-permethrin residue levels (p = 0.040) than their rural counterparts. These preliminary findings suggest the translocation of pesticides into motor vehicles should be considered as a source of pesticide exposure.

CDC Urine Analysis

Table 2 summarizes total diethyl and dimethyl urine levels at three points during pregnancy. The levels rise significantly during the pregnancy (p < 0.01). The enrollment levels are slightly lower, and the delivery levels are slightly higher, than levels reported for children in Washington State. Through the CDC Research Data Center, we have obtained percentiles for women of child-bearing age (18-35) samples as part of the National Health and Nutrition Examination Survey (NHANES) IV and published in the CDC Exposure Report. When comparing individual metabolites, CHAMACOS metabolite levels are in some cases significantly higher than the NHANES reference data (Figure 1). Statistical analyses evaluating possible risk factors for CHAMACOS exposures, including job type, number of farmworkers in the home, wearing of work shoes and clothes into the home, and proximity to fields are not significant, except, possibly, proximity to fields for diethyl pesticides.

Table 2. Organophosphate Urinary Metabolite Levels at Three Time Points During Pregnancy

	Geometric Mean	Range	Detection Frequency	Percentiles
Dimethyl Metabolites (nmol/L)				10	25	50	75	90
Baseline (n=587)	89.9	14-34363	86%	15	23	77	258	690
26 Weeks (n=500)	76.4	4.8-4058	100%	17	35	76	155	339
Delivery (n=494)	170.1	4.8-21857	95%	19	65	163	445	1321
Diethyl Metabolites (nmol/L)
Baseline (n=590)	17.6	1.7-1319	83%	6	6	14	33	75
26 Weeks (n=503)	20.6	1.7-735	100%	4	9	23	44	91
Delivery (n=493)	24.8	1.7-666	96%	6	11	26	57	122

Percentile

Figure 1. CHAMACOS (n = 587) and CDC (n = 110) Total Urinary Pesticide Levels. Preliminary CDC percentiles for women of child-bearing age (18-35 years) sampled during NHANES IV. CDC anticipates final public release of NHANES data for approximately 2000 people in early 2003.

Cumulative Risk Assessment of Organophosphate Pesticide Exposures to Pregnant Women

This work is one of the first case studies applying proposed EPA guidelines for cumulative risk assessment of mixed exposures to multiple organophosphate pesticides.

Method 1—Dialkylphosphate Metabolites. Based on dialkylphosphate metabolite levels derived from multiple urine samples collected from 455 pregnant women, we estimated cumulative OP pesticide doses with steady-state models. We used pesticide use reporting data for the Salinas Valley to quantify the likely mixture to which the women were exposed. Findings suggest that approximately 14 percent of pregnant CHAMACOS women had organophosphate exposures that exceed a health-based reference dose estimated as the index chemical’s inhibition benchmark dose (BMD10) divided by a 100-fold safety factor. The BMD10 is based on the most recent EPA Office of Pesticide Program’s benchmark dose for chlorpyrifos. The risk assessment methods convert all exposures into chlorpyrifos equivalents. The uncertainty factor of 100 is based on standard recommendations to account for population and species variability. Including additional safety factors of 3x or 10x, which has been proposed as part of implementation of the Food Quality Protection Act (FQPA), will substantially shift the estimated proportion of women exceeding a given health-based reference value. See Figure 2.

Method 2—Chemical Specific Metabolites. We collected urine samples from 50 randomly selected CHAMACOS participants and analyzed them for 10 organophosphate-pesticide specific metabolites. Based on these metabolite concentrations, we will estimate cumulative 24-hour organophosphate pesticide dose equivalents by adapting EPA’s new guidelines for cumulative risk assessment of pesticides that share a common mechanism of toxicity. This will enable us to estimate cumulative dose and risk from exposure to a quantifiable mixture of organophosphate pesticides. Our aim is to determine whether pregnant women living in this region are potentially exposed to organophosphate pesticides in excess of health-based reference values and thus whether their fetuses may be at an increased risk of adverse health outcomes.

Time Activity Assessment (TAA)

Stanford collaborators have completed the videotaping portion of the project. Twenty-three children have been taped, including 11 during the summer of 2002. Our original goal was to tape 20 children. However, extra children were recruited to try to balance the gender and age structure, to make up for shorter taping time in some cases, and to take advantage of equipment and staff already in the field. Thirteen girls and 10 boys were videotaped, for a total 93 hours of footage. Participants included residents of Salinas and smaller towns in the southern Salinas Valley. Tapes are being copied to prepare them for data translation, which will begin this fall. We are also making copies and delivering them to the families. It will take approximately 270 hours to complete the translation of the mouth and left and right hand movements. The time-sequenced micro-activity data obtained from the videotapes will be used as input data in several exposure modeling efforts ongoing in the Stanford University laboratory. The data will also be used in several modeling efforts by other team members of the CHAMACOS group.

Figure 2. Distributions of Average Cumulative OP Pesticide Dose Estimates for Pregnant Women in an Agricultural Community (n = 455)

Pesticide Use Report (PUR)

Unique to California, 100 percent of all agricultural pesticide use is reported to the California Department of Pesticide Regulation. The PUR system contains data on all individual agricultural pesticide applications, organized by geographic unit (geocoded to 1 square mile), dates of pesticide applications, pounds of active ingredient applied, crop, and number of acres treated. In collaboration with research partners at the California Department of Health Services, we generated PUR tables for all pesticides applied in 1999 and 2000 (year 2000 PUR became available in early 2002) in the study catchment area. We developed methods and computer coding to identify outliers and remove them from the database. Geographic coordinates of residences determined in home visits have been linked to the PUR data and nearby pesticide use indices have been developed. See Figure 3 for an example of mapped pesticide use data.

Figure 3. Total Organophosphate Pesticides Used in Salinas Valley, 1998

Pesticide Exposure Modeling

To date we have completed modifications to CalTOX—a regional, fugacity based multimedia exposure model—so that it can be used to quantify the relationship between pesticide body burden and explanatory variables such as time-activity data, contact frequency, pesticide use, etc. We have also worked on methods to improve dermal exposure modeling for complex exposure scenarios.

Multimedia Cumulative Exposure Modeling. The CalTOX model is now in a form that can be used to quantify cumulative and aggregate exposures from pesticides over periods of weeks, months, or year from dermal, ingestion, and inhalation pathways. The model can be used to characterize uncertainty and variability. CalTOX is currently in a form that determines doses attributable to local and household pesticide use within the Salinas valley. Addition of data derived from the EPA Dietary Exposure Potential Model (DEPM) will allow determination of ingestion exposures from the general food supply.

We have built up a database of chemical properties (solubility, vapor pressure, octanol-water partition coefficients, half-life, etc.) for pesticides used in the Salinas Valley including:

• Azinphos-methyl
• Chlorpyrifos
• Diazinon
• Dimethoate (Cygon)
• Disulfoton
• Malathion
• Methidathion
• Methyl parathion
• Naled
• Oxydemeton-methyl
• Phosmet

We have also assembled information on monthly pesticide application loads (kg/m2) and yearly averaged climate and meteorological data and soil-properties for the Salinas Valley. We are still building up our database on household characteristics for the farmworker population in Salinas, on child-activity data, and on child diet information for Hispanic children.

We have used the modified CalTOX model to build a nested structure to account for long-term regional concentrations of pesticides from outdoor uses. This structure is used to evaluate multimedia exposures attributable to agricultural uses as well as suburban outdoor uses. With this nested model, we have determined that the modeling scales of the nested structure are dependent on the characteristic travel distance for each pesticide. A case study with chlorpyrifos indicates that indoor uses and food are the main contributors to exposure. This work was presented at the Society for Environmental Toxicology and Chemistry European Meeting in May 2002. Other findings from this analysis are that multiple pathways contribute to the overall exposure of pesticides, that multiple models should be used to account for all exposure pathways, and that fate models that make use of nested transport models can produce exposure predictions that are within an order-of-magnitude of measured values.

Dermal Exposure Modeling. Because assessing cumulative exposures to pesticides in residential environments requires an evaluation of intermittent dermal contacts, we have developed and evaluated a numerical model of contaminant transport through the epidermis. This model includes characterization of exposure scenarios, skin surface loading and unloading rates, and contaminant movement through the epidermis. With a series of archetypal contact scenarios based on behavioral studies to estimate characteristic loading and unloading rates from dermal transfer studies, we determine cumulative uptake through the epidermis from intermittent contacts. The model can account for variable exposure times, temporal and spatial variations in skin properties, and long-term accumulation. With this model we have determined the accuracy of a macro-activity parameter such as cumulative contact time for predicting long-term integrated uptake of pesticides from complex exposure scenarios. Our results reveal that a simplified model based on chemical-specific skin diffusion and cumulative contact time can be used in place of the more complex model that requires activity data on 1- to 5-minute intervals. For characteristic children’s behavior and hand loading and unloading rates, we have found a power law accounts for the relationship between cumulative contact time and mass transport through the skin.

Breastmilk Sub-Study

In collaboration with CDC, Rosana Hernandez, a doctoral student with CHAMACOS, spent 9 weeks this summer at CDC in Atlanta developing methods to extract, analyze, and quantify pesticide and other environmental chemical residues in breast milk samples. This project stemmed from an EPA Science To Achieve Results (STAR) Fellowship proposal that was granted to Ms. Hernandez in 2001. New technologies, including Stir Bar Sorptive Extraction (SBSE), were employed for method development using Gerstel Twister stir bars and the Gerstel Thermal Desorption System attached to an Agilent GC/MS. Initially, extraction of 40 environmental chemicals including non-persistent pesticides, persistent pesticides, polychlorinated biphenyls (PCBs), and phthalate monoesters was attempted using this technology. Nineteen compounds had a relatively linear response (R2 > 0.85) over three orders of magnitude in concentration. Seven of these compounds were OP pesticides, a primary exposure of interest for the CHAMACOS study. At the end of the 9-week internship, researchers at CDC concluded that SBSE is a valuable technique for quantifying environmental chemicals in biological fluids such as breast milk. This research will be continued at the CDC in collaboration with Ms. Hernandez. In addition to the laboratory work performed at CDC, Ms. Hernandez is conducting a small study to collect breast milk samples from women in the San Francisco Bay Area. These samples are needed to conduct method development and method validation work. Recruitment flyers were placed in research facilities; doctors’ offices; and Women, Infants, and Children (WIC) offices. Overall participant response has been positive with approximately 20 participants contributing a total of 2.1 L of breast milk over the 3 months of collection. All types of samples were accepted including older frozen samples, refrigerated samples, as well as freshly pumped samples. Collection will continue until 3 L of breast milk is collected and available for method development work.

Association of Pesticide Air Monitoring and Pesticide Use Data

In support of the CHAMACOS modeling component, a small statistical analysis evaluating the association of chlorpyrifos and diazinon air monitoring with state PUR data was completed. The state Air Resources Board has monitored these pesticides in agricultural areas in California (Fresno and Tulare counties), offering a unique opportunity to evaluate these associations. For chlorpyrifos and chlorpyrifos oxon, nearby pesticide use significantly predicted air concentrations. The R-squared improved as the three lagged previous day use variables were added. For the full model, the R-square, adjusted for the number of variables in the model, was 0.15 and 0.36, for chlorpyrifos and chlorpyrifos oxon, respectively, p < 0.01. The higher R-square for chlorpyrifos oxon is consistent with the sunny conditions (chlorpyrifos in air is quickly photo degraded) and the longer estimated half-life of chlorpyrifos oxon (12 hours) compared to chlorpyrifos (4 hours). Similar trends were found for diazinon. For the full model, the R-square, adjusted for number of variables in the model was 0.32, p < 0.01. Average daily temperature and wind speed improved the model significantly. Our results suggest that agricultural applications of chlorpyrifos and diazinon are associated with outdoor air concentrations and that the PUR is a valid indicator of inhalation exposures.

Amniotic Fluid Pilot Study

A pilot study of pesticides in amniotic fluid has been completed. Funding was provided by a pilot grant from the National Institute of Environmental Health Sciences (NIEHS) Environmental Health Sciences Center at Berkeley. OP, carbamate, wood preservative, disinfectants, and mothball pesticides were detected in the amniotic fluid samples. A journal article will be submitted for publication at the end of this year.

Housing Quality

Of the women who enrolled in CHAMACOS, home environmental assessments were completed for 513 during pregnancy and 385 when the child was 6 months old; these assessments represent 645 distinct homes. The home visits included visual and hands-on inspections and a questionnaire about the home environment.

Most of the homes that were visited had significant housing quality problems, including peeling paint (58%), moderate to severe mold on at least one wall (42%), water damage (25%), plumbing leaks under sinks (16%), or rotting wood (14%). Twenty percent of sleeping areas for the pregnant women or children had severe mold infestations. Sixty percent of homes had cockroach infestations, and 32 percent had rodent problems. When compared to national Hispanic reference data from The American Housing Survey conducted by the U.S. Census and the Department of Housing and Urban Development, our study population had substantially more housing quality problems. Peeling paint, rodent infestations, and bad odors were 12, 1.8, and 1.4 times more common, respectively. Leaks found under sinks were 1.4 times more common than all leaks reported nationally. Housing density in our populations (calculated as the number of people living in a home divided by the number of rooms) was also higher, with nearly half (49%) of the homes having 1.5 or more persons per room; this is compared to 2.8 percent of Hispanic homes and 0.4 percent of all occupied units nationally.

We also computed a housing quality index that was based on the presence or absence of peeling paint, mold, water damage, or rotting wood and included a binary housing density variable with 1 person/room as the cutoff point. Logistic regression analysis revealed that the housing quality index was moderately associated with cockroach (OR = 1.6; 95% CI = 1.4-1.9) and rodent (OR = 1.4; 95% CI = 1.2-1.6) infestations. A matched pair analysis that included the 121 families that moved between the prenatal and 6-month home visits was carried out in order to control for personal behaviors that might contribute to attracting pests. The results of this analysis suggest that peeling paint and water damage are the primary predictors of rodent presence, even when the same family lives in two different homes.

We found stored pesticides in the homes of 297 (46.1%) participants. Of these pesticides, 62 percent were pyrethroids, 14 percent were carbamates, 12 percent were OPs, and 21 percent included piperonyl butoxide. Pest infestation was moderately associated with pesticides use (OR = 1.5; 95% CI = 1.1-2.1 for rodents and OR = 2.2; 95% CI = 1.5-3.0 for cockroaches) and with pesticides stored in the homes (OR = 1.5; 95% CI = 1.1-2.1 for rodents and OR = 1.9; 95% CI = 1.4-2.6 for cockroaches).

Research Changes

As discussed in last year’s report, the QEA was redesigned to focus intensive environmental sampling on a smaller group of children (n = 20) while freeing up resources to analyze dust samples across the cohort. The changes followed advice from our Scientific Advisory Committee. The intensive QEA sampling was completed in September. Laboratory analysis of 510 dust samples, 170 collected each at the prenatal, 6-month, and 12-month home visits, was completed in September. Battelle Laboratories is completing final quality assurance reviews and will report final data in October. Analysis of the QEA samples will start in October 2002. To avoid additional burdens on CHAMACOS participants, families participating in the QEA and the Time Activity Assessment were non-participants recruited from similar populations served by our clinical and community partners.

We are conducting a 24-month home visit, which was not originally proposed, in order to collect GPS latitude and longitude coordinates and also to complete a home environmental assessment for this age group. This home visit is scheduled with the 24-month blood collection so that home visit staff can assist the CHAMACOS phlebotomist in collection of child blood samples. No environmental samples are collected during this visit (none were planned).

Journal Articles:

No journal articles submitted with this report: View all 15 publications for this subproject

Supplemental Keywords:

Toxics, Air, Scientific Discipline, Health, RFA, Susceptibility/Sensitive Population/Genetic Susceptibility, indoor air, Risk Assessments, genetic susceptability, Health Risk Assessment, Children's Health, Biochemistry, pesticides, Environmental Chemistry, environmental hazard exposures, health effects, indoor air quality, intervention, air quality, assessment of exposure, farmworkers, outreach and education, agricultural community, community-based intervention, exposure pathways, home, infants, pesticide exposure, sensitive populations, children, age dependent response, disease, household, children's vulnerablity, environmental health hazard, human exposure
Relevant Websites:

http://www.chamacos.org

Progress and Final Reports:
1999 Progress Report
2000 Progress Report
2001 Progress Report
Original Abstract

Main Center Abstract and Reports:
R826709    CECEHDPR - University of California at Berkeley

Subprojects under this Center: (EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
R826709C001 Community Based Intervention to Reduce Pesticide Exposures to Young Children
R826709C002 The Epidemiological Investigation of the Effects of Pesticide Exposure on Neurodevelopmental, Growth, and Respiratory Health of Farmworker Children
R826709C003 A Comprehensive Assessment of Sources of Pesticide Contamination, Concentrations in Pathways, and Exposure-prone Behavior

Top of page

The perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.

Local Navigation

• NCER Home
• Basic Information
• Grants & Funding
• Guidance & Frequent Questions
• STAR Grant Recipients
• Fellowships
• Small Business
• Research Centers
• Other Programs
  • Biodiversity
  • CNS
  • EPSCoR
  • ETOP
  • Nanotechnology
  • P3
  • STAA
• Research Results
• Science Topics
  • CBRA
  • Childrens Centers
  • Drinking Water
  • Economics
  • Endocrine Disruptors
  • Global Change
  • Particulate Matter
  • TSE
• Publications & Proceedings
• Calendar