Human Biomonitoring for Environmental Chemicals (2006)

With the growth of biomonitoring to include more investigations, more chemicals, and expanded population sampling, the challenges of interpretation are increasingly evident. What does science tell the nursing mother when she learns that she has a large number of chemicals in her breast milk, all at relatively low concentrations? How should the findings of low concentrations of persistent chemicals in umbilical-cord blood be reported and interpreted? And, perhaps most important, what are the implications of biomonitoring results for public health and environmental policy? Those questions epitomize the challenges of biomonitoring with respect to its use, interpretation, and limitations, and the ethical and communication issues that it presents. This chapter provides a research agenda for addressing those issues to advance the application of biomonitoring.

In several instances, biomonitoring data have confirmed health effects of environmental exposures and have validated public-health policies. For example, population data on blood lead concentrations that were associated with adverse health effects provided the impetus for the U.S. Environmental Protection Agency (EPA) regulations reducing lead in gasoline. Methylmercury concentrations in blood and hair that were correlated with neurodevelopmental effects provided the rationale for EPA’s revision of the oral reference dose. In those examples, the biomonitored concentrations of chemicals could be shown to be related to adverse health effects because of the body of epidemiologic, toxicologic, and clinical

data. In another example, data on serum cotinine, a biomarker of exposure to second-hand smoke, showed that serum cotinine in U.S. children and adults declined by more than 50% among nonsmokers from 1998 to 2002, demonstrating the effectiveness of smoking cessation efforts in the United States.

For population-based studies, biomonitoring data can help to identify chemicals that are found in the environment and in human tissues, can be used to monitor changes in exposure, and can be used to establish the distribution of exposure among the general population. Biomonitoring provides a measurement of exposure that—when used with available epidemiology, toxicology, and pharmacokinetic modeling data—can help to estimate how much has been absorbed into the body and estimate potential health risk. Biomonitoring can also be a very efficient means of assessing exposure and can provide a context for understanding environmental exposures on an international level.

In spite of the potential of biomonitoring, tremendous challenges surround its use. They include improving our ability to design biomontoring studies, interpreting what biomonitoring data mean for health risk and public health, addressing ethical uses of the data, and communicating results to policy-makers and the public.

To realize its potential, an investment in biomonitoring research is needed to address the critical knowledge gaps that hinder our ability to use and interpret the biomonitoring data. The committee’s research recommendations focus not on specific chemicals but rather on methods that can be applied to a broad array of chemicals. Implementation of the research recommendations will benefit from enhancement of some parts of our nation’s research infrastructure.

To address the challenge of improving the interpretation and use of biomonitoring data, the committee has developed four major findings and corresponding research recommendations. The committee considered these recommendations to be of the highest priority in advancing the field of biomonitoring. Addressing the knowledge gaps will require a broader vision of biomonitoring, including a coordinated scientific approach to setting priorities for biomarker development; better integration of epidemiology, toxicology, pharmacokinetic modeling, and exposure assessment to put biomonitoring results into a meaningful risk context; improved reporting of biomonitoring results; and understanding of the ethical issues that constrain the advancement of biomonitoring. Other research recommendations, not addressed below, are found in Chapters 3-6.

Biomonitoring offers great promise as an effective technique for identifying chemicals of potential public-health significance. The committee finds that broad population screening for a large number of chemical biomarkers has provided useful and at times surprising evidence of human exposure. That type of screening should continue. However, it can be improved. The current biomonitoring strategy relies on the emergence of biomarkers from various research avenues (such as epidemiology, analytic chemistry, and workplace monitoring), but the uncoordinated fashion in which this has occurred has allowed widespread exposures to go undetected for many years—for example, exposures to polybrominated diphenyl ethers (PBDEs) and perfluorooctanoic acid. In addition, susceptible subpopulations, including infants and children, are generally omitted from large-scale biomonitoring studies because of difficulty in sample collection.

The committee recommends that a coordinated scientific research strategy be developed to ensure that selection of chemicals for development of biomarkers and for biomonitoring focus, first and foremost, on the potential of chemicals to cause harm, and consider the likelihood of widespread exposure, including exposure of susceptible subpopulations. The biomonitoring-research strategy needs to set priorities among chemicals on the basis of one or more of the following: evidence of widespread exposure in the general population, biomonitoring data or exposure-analysis information that indicates exposure of susceptible subpopulations, toxicology data indicating that a chemical is capable of causing effects of public concern, and environmental persistence or use-pattern information that indicates that exposure will probably continue or increase in the future. The strategy should include a systematic analysis of chemicals to which there is widespread exposure so that priorities can be set for biomarker development. The biomarkers can be combined with biomarkers already developed via other avenues, such as epidemiologic research and workplace monitoring, to construct a comprehensive biomonitoring program.

Developing the coordinated scientific strategy will require input from various agencies involved in biomonitoring and supporting disciplines, in-

cluding the Centers for Disease Control and Prevention (CDC), EPA, the National Institute of Environmental Health Sciences (NIEHS), the National Toxicology Program (NTP), the Food and Drug Administration, and the U.S. Department of Agriculture. Coordinated input from those agencies would ensure that population-based biomonitoring studies would target chemicals of increasing public-health concern, with chemicals that are prudent to monitor for broad screening or status and trends purposes. The scientific strategy would need to be transparent to the public. Such a coordinated approach might eliminate redundancies in research efforts among agencies and help to leverage additional funds for the most pressing public-health questions.

In addition to developing a strategy for identifying specifically targeted chemicals, the committee recommends that population-based biomonitoring studies include a subset of samples to screen specific populations for untargeted analytes and to identify and quantify these chemicals. Such an approach is feasible with current analytic techniques that provide for the tentative identification of unknown analytes.

To address sensitive subpopulations better, including infants and children, the committee recommends the development of additional matrices— including cord blood, saliva, meconium, and breast milk—in concert with expanded analyses of infant blood or urine to enhance our ability to detect exposure during early life stages.

Our ability to interpret the risks associated with biomonitoring findings depends on our knowledge of exposure, toxicologic, pharmacokinetic, and epidemiologic data of particular chemicals, as illustrated by the framework in Table 3-1. To interpret biomonitoring data better in the context of

this framework and to understand the public-health implications of the data, coordinated research is needed to increase the use of biomonitoring in epidemiologic studies, to expand toxicologic studies to incorporate the collection of biomonitoring data and foster the development of pharmacokinetic models, and to incorporate exposure assessment in biomonitoring studies. Each is discussed below.

Development of the application of biomarkers in epidemiology is needed to improve the understanding of the relationships between biomonitoring data and health effects. As illustrated by our knowledge of blood lead that is based predominantly on epidemiologic studies that examined the relationship between blood lead and IQ in study subjects, epidemiologic studies may provide the optimal database for linking biomonitoring data to health effects. However, such information, as in the case of lead, is accumulated over many years and at high cost. The key question is when and how to use epidemiology more efficiently to attain such understanding of other chemicals.

Quantitative epidemiologic research is expensive, but the emphasis should be on cost effectiveness, not on cost. The most scientifically desirable scenario is the funding of a major epidemiologic study specifically targeted at interpreting biomonitoring data; however, funding realities limit the feasibility of that. The federal budget for epidemiology related to chemical exposures is significant and should be further leveraged. Almost all epidemiologic studies have some metric of exposure (such as questionnaires, environmental measurements, and biomonitoring), and each approach has different degrees of value and cost. With careful attention to study design, supplemental funding for planned or impending epidemiologic studies that take tissue samples and analyze them for biomarkers could contribute substantially to interpretation of the data and assist in our understanding of the data gaps in the continuum from exposure to disease.

Consideration should be given to increasing the number of biosamples collected and stored in epidemiologic studies to provide future research opportunities for assessing associations between biomarkers and outcomes within existing study designs. Examples include recent analyses of organochlorine in biologic samples collected and stored decades earlier as part of the 1959-1965 National Collaborative Perinatal Project (Borrell et al. 2004; Gray et al. 2005; Longnecker et al. 2005). As new biomonitoring results have become available, spinoff studies have been initiated within the NIEHS-EPA Center for Children’s Environmental Health, for example, to evaluate effects of prenatal phthalate and PBDE exposures by using banked samples within the existing research designs (Eskenazi et al. 2005). The National Children’s Study (NCS), whose funding is currently being debated, has the potential to offer many opportunities to assess relationships

between high-priority biomarkers and health outcomes, both prospectively and retrospectively. The committee also recognizes that population-based biomonitoring studies complement the national strategy of public-health tracking and the links between environmental exposures and public health. The applications of biomonitoring in those studies are important examples for future epidemiologic investigations.

Toxicologic studies need to be expanded to incorporate collection of biomonitoring data in animals that can be related to humans. Much of the dose-response information used in risk assessments is derived from animal toxicologic studies, and these do not collect information on internal dose. Therefore, dose-response relationships can be expressed only in terms of external dose (such as milligrams per kilogram per day). However, to interpret biomonitoring data, the relationship between internal dose (biomarker concentration) and effect must be understood.

Expansion of animal toxicologic study designs to include collection of biomarker data (for example, concentrations of the parent chemical in blood or of key metabolites in urine) will facilitate the development of biomarker-response relationships that can be extrapolated across species to interpret human biomonitoring results. That involves adding pharmacokinetic groups to a toxicologic study design or recreating a toxicologic study (with identical species and dose groups) that provided key dose-response information. Ideally, sufficient pharmacokinetic data will be collected from the study to facilitate the development of a physiologically based pharmacokinetic (PBPK) model that can be used to predict biomarker concentrations in animals across a wide range of doses and whose results may be extrapolated to humans. For example, NTP toxicologic protocols have added a pharmacokinetic component that has the potential to assist in the development of PBPK models and estimation of biomarker concentrations in toxicologic studies (Buchanan et al. 1997).

The incorporation of biomonitoring data in toxicologic studies will maximize their utility for interpreting biomonitoring data. Characterizing a biomarker-response relationship in animals could lead to development of reference doses or cancer slope factors based on biomarker concentration rather than external dose. Biomarker-based toxicity values would be directly applicable to interpreting human biomonitoring studies. Fostering development of PBPK models that can be used for internal dose reconstruction in critical toxicologic studies will facilitate extrapolations across species, dose routes, and doses.

Exposure assessment should be a component of population-based biomonitoring studies to facilitate interpretation of the data. Typically, large-scale biomonitoring studies do not evaluate potential sources of exposure. That often leads to the question, Where is the exposure coming from? For some chemicals, exposure pathways may be well defined from previous

studies (for example, mercury in the general population comes primarily from fish ingestion); for others, however, it may be largely unknown.

The committee recommends the inclusion of a detailed and accurate exposure analysis for a subset of the biomonitored population in large-scale biomonitoring studies that includes analyses of environmental media in the residence and uses a survey instrument to obtain information on diet, consumer product use, occupational exposures, and other factors relevant to the chemical exposure pathways that are being examined. The exposure assessment can be patterned on protocols used in other exposure analyses, such as the National Human Exposure Assessment Survey (NHEXAS), the Minnesota Children’s Pesticide Exposure Study, and Children’s Total Exposure to Pesticides and Other Persistent Organic Pollutants.

In addition, existing databases where environmental media and biomonitoring data are collected (such as NHEXAS) could be further studied to estimate exposure and explore the relationships between biomarker concentration and exposure. That information can be used to apportion chemical intake into the different exposure pathways to assist in interpreting population variability, to calculate exposure by combining environmental measurements with survey information to verify estimates of exposure from pharmacokinetic models, and to identify research needs on the basis of discrepancies between estimates obtained from the exposure-pathways analysis and biomonitoring results.

Several other kinds of research would enhance our interpretation of the biomonitoring data. Such research includes understanding and characterizing the effect of human variability on biomonitoring results, including sampling time and population variability in metabolism, creatinine clearance, and other pharmacokinetic factors influenced by age and sex. Specific research includes dose-simulation modeling techniques that combine behavioral and pharmacokinetic factors to characterize how population variability in exposure can affect biomonitoring results in different age groups (Zartarian et al. 2000; Rigas et al. 2001) and the development of pharmacokinetic models for various susceptible subpopulations to predict biomonitoring results.

The committee recommends that efforts be made to develop human pharmacokinetic models early in the study-design process to understand the influence of such factors as metabolism, sampling time, and population variability, that are critical to interpretation of the biomonitoring data.

Evaluating the extent of exposure to mixtures and developing methods to assess public-health effects by using emerging technologies (such as -omics) are important research needs. Because population-based biomonitoring studies report on human exposures to a large number of chemicals, understanding exposures to mixtures is critical. Monte Carlo simulation modeling techniques are needed to estimate the number and concentrations

of chemicals to which the population in the CDC biomonitoring study is exposed. Animal bioassays should be developed to assess the combined effects of chemical mixtures that are commonly found in human tissues (e.g., NTP 1993a; NTP 1993b; Yang 1994). In addition, modeling techniques need to be developed to simulate the pharmacokinetic and pharmacodynamic interactions of multiple chemicals (Poulin et al. 2001; Poulin and Theil 2002; van de Waterbeemd and Gifford 2003; Yang et al. 2004; Reddy et al. 2005; Yang et al. 2005; Yang et al., unpublished material, 2006).

Emerging technologies, including toxicogenomics, can be used to develop methods to assess public-health effects better. For instance, toxicogenomics provides an opportunity to move beyond the traditional approaches of exposure assessment—based on one exposure to one chemical in one environmental medium—to an approach involving multiple exposures and via multiple biologic-response pathways (Weis et al. 2005; Wild 2005). A recent molecular epidemiologic study by Vermeulen et al. (2005) that used array-based proteomics to develop potential biomarkers of exposure and early biologic effect demonstrated the potential for -omics biomonitoring in chemical-exposed populations. Emerging technologies will also allow for the intergration of biomarkers of exposure with those of effect to determine whether the markers of effect track with individual chemicals or with exposure to mixtures.

Given the central role of communication in interpreting and using biomonitoring data, developing research must have high priority for biomonitoring investigators and funders. To that end, the committee proposes three communication research recommendations (detailed in Chapter 6). Understanding how laypeople and scientists conceive of the causal links between external dose, biomarker concentrations, and biologic effects and their views about exposure reduction and risk managers, will reveal their agreements and disagreements and will suggest bases of messages that could reduce disagreements. Assessing the content of current biomonitoring education and com-

munication materials will help to evaluate their efficacy, and determine the extent to which beliefs about causal linkages are accurately reflected in them. Alternative messages about biomonitoring, especially those concerning the deep uncertainties in the field, should be informed by research on people’s beliefs and on how people communicate. Testing responses to alternative messages on individual, community, and population-based aspects of biomonitoring will advance communication strategies.

Participants in public-health studies that measure hundreds of biomarkers might give “informed consent” only with respect to the general objectives of the study on the grounds that detailed discussion of each biomarker is prohibitive. However, failing to make available more detailed information, no matter how many chemicals are involved in a study, raises ethical questions.

Because of the challenges posed by informed consent for studies that use high-output, high-throughput technologies, the committee recommends research that develops, evaluates, and disseminates methods that ethically and practically inform study subjects during recruitment and during later communication of study results.

There is a concern that blanket consent (for example, for future testing of tissue samples with genomic or metabonomic assays that are not available at the time of study recruitment) has the potential to result in abuse. It is a highly contentious issue and particularly pertinent to sample collection from children and other susceptible subpopulations. As a result, it has led to increased difficulty in obtaining institutional review board approval for some kinds of biomonitoring studies.

The committee is sympathetic to such concerns but is also aware of the ethical (and practical) problems of undue replication of tissue sampling that the implicit ban forces, as each new sample application is imagined. Therefore, the committee recommends that research be conducted to develop new approaches for obtaining consent for future uses of biomonitoring data.

The current infrastructure to support research recommendations discussed previously is severely limited. Improvements in the research-related infrastructure are needed to support these recommendations and to enhance the value of the biomonitoring activities described in preceding chapters. In many cases, the recommendations for infrastructure needs are cost-effective in that they rely on expansion of structures and activities that are already in place. The infrastructure needs encompass laboratory issues, expanding the scope and utility of CDC’s National Health and Nutrition Examination Surveys (NHANES) data, maximizing the utility of collected human samples, and fostering international biomonitoring collaboration.

Further investments in federal, state, and university laboratories are needed to create the national capacity to exploit biomonitoring fully as a public-health tool. Analysis of human specimens for trace concentrations of environmental chemicals poses serious challenges to the analytic chemistry laboratory. Specimen sizes are often small, the required detection limit is often low, and many interferences are typically present in a complex matrix, such as blood or urine. The recent growth in biomonitoring applications has been made possible by commercial availability of a new generation of more sensitive and selective instrumentation for chemical identification and quantification, of isotopically labeled internal standards, and of robotic systems capable of automated sample preparation. The costs associated with those items and the specialized skills needed to perform the tests have limited the number of laboratories capable of biomonitoring measurements. In recognition of the national deficiencies in the biomonitoring-laboratory capacity, CDC funded 33 states to identify local public-health problems and to develop plans to create the biomonitoring-laboratory capacity needed to address them. Because of fiscal constraints, only three grants to provide the needed laboratory capacity were ultimately awarded—and at substantially decreased funding levels (APHL 2006; CDC 2005). Funds should be appropriated to expand CDC support of state public-health laboratories and to allow implementation of their already-developed plans to address local exposure-related problems.

Improvement in the array of chemicals capable of being measured in human specimens is needed. For example, the Government Accountability Office (GAO, previously the General Accounting Office) assembled a non-exhaustive list of 1,456 chemicals considered by the Department of Health and Human Services, EPA, or other federal entities to pose a threat to human health (GAO 2000). Laboratory methods have not been developed and validated to measure most of them. The recent NHANES report, which provided

the most extensive biomonitoring survey of the U.S. population available, lists only 148 analytes, reflecting a partial representation of potential exposures, many of which are not discretely listed in the GAO report.

Laboratory methods in use today need further improvement. Lower detection limits will facilitate studies of background contamination in “un-exposed” populations and determination of reference ranges. Many analytes cannot now be conveniently measured in a large proportion of such people. For example, the 95th percentile lipid-adjusted serum 2,3,7,8-TCDD concentrations were below the limits of detection for all seven population groups tabulated in the most recent NHANES report (CDC 2005). Even when censored data are less predominant, they force assumptions about the distribution of results below the detection limit, and this introduces uncertainty into many statistical calculations and complicates efforts to detect temporal, geographic, or ethnic differences in exposure (Needham 2005). Greater analytic sensitivity can also allow testing of smaller sample aliquots. Smaller aliquots will permit more tests on a single sample, help to identify more clearly chemical exposures that correlate with one another, and help to identify exposure sources.

Laboratories must develop the ability to test specimen types and sample volumes that can be collected with less invasive sampling techniques to facilitate subject participation, especially in studies involving children. More sensitive analytic methods can allow collection of smaller specimens; this is especially important in studies involving newborns and children (Barr et al. 2005). Additional analytic methods must be developed and validated to address specimens less invasively obtained (such as saliva, exhaled breath, and breast milk).

The need for high-throughput, low-cost testing procedures will be increasingly apparent as biomonitoring techniques are more widely applied to large-scale epidemiologic studies; mass-casualty events, such as chemical terrorism and chemical accidents; and efforts to define reference ranges in multiple but narrower segments of the population. Substantial throughput and cost improvements may require innovative approaches quite different from those in use today.

Improving the quality of biomonitoring-laboratory data (especially their accuracy, precision, and interlaboratory comparability) will be important for many of the applications discussed in Chapter 3 and for supporting the research recommendations described previously. The committee recommends that a group of biomonitoring-laboratory experts be assembled to make consensus recommendations to improve the quality of data used for medical and research purposes. The Clinical and Laboratory Standards Institute (CLSI 2006) exemplifies an organization that might be charged with this task. Possible goals include

• Creating consensus recommendations for good laboratory practices associated with clinical-sample analysis for environmental and occupational medicine.

• Making recommendations to increase the availability of isotopically labeled analogues of target chemicals, such as taking advantage of the extensive use of labeled chemicals in CDC’s analytic program.¹

• Developing a plan to increase the available array of biomonitoring-relevant analytic reference materials (human samples containing chemicals at known concentrations appropriate for environmental or occupational medicine); the National Institute of Standards and Technology’s Standard Reference Material Program is a useful model of such an activity.

• Developing a mechanism to expand interlaboratory comparison programs to include a broader array of chemical targets in human specimens; no broadly based program exists in the United States.

• Improving the quality of the biomonitoring laboratories that analyze human samples for the purposes of diagnosis, prevention, treatment, or health assessment. These laboratories are regulated under the Clinical Laboratory Improvement Amendments (CLIA) of 1988 (42 CFR 493 [2000]). However, the focus of this statute is mainstream clinical testing. For example, the 41-page CLIA tabulation of approved proficiency-test providers lists only one environmental chemical (blood lead) (CMS 2005).

The proposed consensus committee should consider whether the CLIA program should become more active in ensuring the quality of environmental and occupational biomonitoring data by, for example, establishing a chemistry subspecialty in environmental and occupational medicine or expanding the array of biomonitoring-relevant proficiency tests available from the Center for Medicare and Medicaid Services-approved providers.

As noted in Chapter 2, NHANES and the associated National Report on Human Exposure to Environmental Chemicals provide the most comprehensive summary of biomonitoring data on a representative sample of

the U.S. population. The committee concurs with CDC that the current effort provides data essential for identifying chemicals and concentrations, establishing reference ranges, tracking temporal exposure trends, assessing the effectiveness of interventions to reduce exposure, and setting priorities for research on human health effects (CDC 2005). The achievements of NHANES argue for its expansion and for procedural changes that will enhance the utility of the resulting data.

NHANES reports results by age group, sex, and racial group (Mexican American, non-Hispanic black, and non-Hispanic white). The dataset is insufficient to address other ethnic groups or to determine exposures by locality, state, or region. The committee considers that the missing data are important for setting priorities among groups and geographic areas for intervention, and obtaining them may lead to an improved understanding of exposure pathways. Additional ethnic groups and susceptible subpopulations might be incorporated by program expansion or oversampling within the NHANES program design. Production of location-specific data, especially if temporal trends are sought, may require substantial program redesign, because samples are now collected only from a small number of sites each year (Schober 2005). Alternatively, such data could be produced by supporting state- or city-based health and nutrition examination survey (HANES) projects. The New York City HANES project is a useful example of such a program (Gwynn and Thorpe 2004).

CDC staff have noted that NHANES collects only a small amount of biomonitoring data relevant to exposures in the fetus, infant, toddler, and preschooler (Needham 2005). The committee recognizes that constraints intrinsic to the existing NHANES protocol may make addressing those groups difficult and therefore recommends continued and expanded federal support of the NCS (NCS 2005).

The committee noted that the data presented in the printed Third National Report on Human Exposure to Environmental Chemicals are incomplete and that the content of the larger dataset available on the CDC Web site (NCHS 2005) is not optimized to benefit the entire scientific and medical community. For example, the report’s tables and charts do not include data below the 50th percentile. The publicly available dataset should be sortable by sample type, chemical, region or location, age group, race, and socioeconomic status to facilitate interpretation and identification of groups at higher risk. Some applications of the dataset, such as analysis of exposure to mixtures and chemical interactions, require knowledge of the pattern of chemicals present in individual subjects. CDC should endeavor, where sample volumes permit, to perform the broadest array of tests possible on each subject to maximize the opportunity to detect correlations among chemicals measured in different analyses.

Chapter 4 outlines the steps associated with designing and executing a biomonitoring study. The costs associated with assembling and characterizing a population for study, securing informed consent, and collecting human specimens are substantial. Properly collected and stored specimens remain valuable after completion of the initial study, especially if informed consent was or can be obtained for future studies. When already collected and characterized samples are available, costs of future studies are decreased; samples are readily available for pilot studies, for laboratory-method validation, and for testing with newly developed biomarkers (Holland et al. 2005). For some applications (such as tracking time trends of exposure), specimen banks are essential. The committee considers that future progress toward the research goals described in this chapter will be accelerated and study costs lessened by the increased availability of already collected and characterized samples. However, the difficulties associated with long-term sample storage are substantial. Each sample needs readily accessible, but secure, records related to chain of custody, processing, location, and temperature stability. Costs of equipment (for example, freezers, cryotanks, automated sample handling and tracking equipment and software, and back-up power supplies), space costs, and personnel costs can be high. The committee therefore recommends expanded long-term funding for existing biorepositories and the creation of new biorepositories for support of biomonitoring studies.

Because biomonitoring is conducted on an international level by numerous organizations and there is much knowledge to be gained from understanding worldwide patterns of exposure, the committee encourages the global exchange of biomonitoring information and expertise. That would include sharing of biomonitoring data, study approaches, and tracking of trends. To that end, the committee encourages the development of such information exchanges between EPA and the Organisation for Economic Co-operation and Development (OECD).

This chapter presents the major research recommendations required to improve the use and interpretation of biomonitoring data for improving public health. Implementation of the research recommendations will require expansion of the biomonitoring infrastructure. Table 7-1 summarizes the committee’s research recommendations and infrastructure needs.

APHL (Association of Public Health Laboratories). 2006. APHL Minute Newsletter 1:12-14 [online]. Available: http://www.aphl.org/docs/newsletter/january_february_3.pdf [accessed Feb. 1, 2006].

Barr, D.B., R.Y. Wang, and L.L. Needham. 2005. Biologic monitoring of exposure to environmental chemicals throughout the life stages: Requirements and issues for consideration for the National Children’s Study. Environ. Health Perspect. 113(8):1083-1091.

Borrell, L.N., P. Factor-Litvak, M.S. Wolff, E. Susser, and T.D. Matte. 2004. Effect of socioeconomic status on exposures to polychlorinated biphenyls (PCBs) and dichlordiphenyl-dichloroethylene (DDE) among pregnant African-American women. Arch. Environ. Health 59(5):250-255.

Buchanan, J.R., L.T. Burka, and R.L. Melnick. 1997. Purpose and guidelines for toxicokinetic studies within the National Toxicology Program. Environ. Health Perspect. 105(5):468-471.

CDC (Centers for Disease Control and Prevention). 2005. Third National Report on Human Exposure to Environmental Chemicals. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, Atlanta, GA [online]. Available: http://www.cdc.gov/exposurereport/3rd/pdf/thirdreport.pdf [accessed Sept. 26, 2005].

CMS (Center for Medicare and Medicaid Services). 2005. CLIA Approved Proficiency Testing Programs-2005 [online]. Available: http://new.cms.hhs.gov/CLIA/downloadsptlist.pdf [accessed Feb. 1, 2006].

CLSI (Clinical Laboratory Standards Institute). 2006. CLSI Standards Development [online]. Available: http://www.clsi.org/Content/NavigationMenu/StandardsDevelopment/CLSIStandardsDevelopment/default.htm [accessed Feb. 16, 2006].

Eskenazi, B., E. Gladstone, G. Berkowitz, C. Drew, E. Faustman, N.T. Holland, S. Lanphear, S. Meilsel, F. Perera, V. Rauh, A. Sweeney, R. Whyatt, and K. Yolton. 2005. Methodologic and logistic issues in conducting longitudinal birth cohort studies: Lessons learned from the Center of Children’s Environmental Health and Disease Prevention Research. Environ. Health Perspect. 113(10):1419-1429.

GAO (U.S. General Accounting Office). 2000. Toxic Chemicals: Long-Term Coordinated Strategy Needed to Measure Exposures in Humans. GAO/HEHS-00-80. U.S. General Accounting Office, Washington, DC. May 2000 [online]. Available: http://www.gao.gov/new.items/he00080.pdf [accessed June 3, 2005].

Gray, K.A., M.A. Klebanoff, J.W. Brock, H. Zhou, R. Darden, L. Needham, and M.P. Longnecker. 2005. In utero exposure to background levels of polychlorinated biphenyls and cognitive functioning among school-age children. Am. J. Epidemiol. 162(1):17-26.

Gwynn, C., and L. Thorpe. 2004. New York City HANES: Advancing Local Public Health and Nutrition Surveillance. Abstract 92319 in Program and Abstracts of the American Public Health Association 132nd Annual Meeting; November 6-10, 2004; Washington, DC [online]. Available: http://apha.confex.com/apha/132am/techprogram/paper_92319.htm [accessed Feb. 1, 2006].

Holland, N.T., L. Pfleger, E. Berger, A. Ho, and M. Bastaki. 2005. Molecular epidemiology biomarkers: Sample collection and processing considerations. Toxicol. Appl. Pharmacol. 206(2):261-268.

Longnecker, M.P., M.A. Klebanoff, J.W. Brock, and X. Guo. 2005. Maternal levels of polychlorinated biphenyls in relation to preterm and small-for-gestational-age birth. Epidemiology 16(5):641-647.

NCHS (National Center for Health Statistics). 2005. National Health and Nutrition Examination Survey. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics, Hyattsville, MD [online]. Available: http://www.cdc.gov/nchs/nhanes.htm [accessed May 2, 2006].

NCS (National Children’s Study). 2005. The NCS Herald Study—An NCS Pilot Cohort [online]. Available: http://www.nationalchildrensstudy.gov/events/ncsac/2005_April/loader.cfm?url=/commonspot/securitygetfile.cfm&PageID=10473&CFID=4135254&CFTOKEN=15157050 [accessed June 3, 2005].

Needham, L.L. 2005. Biomonitoring in NHANES and Other Programs. Presented at the Second Meeting on Human Biomonitoring for Environmental Toxicants, April 28, 2005, Washington, DC.

NTP (National Toxicology Program). 1993a. NTP Technical Report on Toxicity Studies of a Chemical Mixture of 25 Groundwater Contaminants Administered in Drinking Water to F344/N Rats and B6C3F₁ Mice. NIH Publication 93-3384. U.S. Department of Health and Human Services, National Institutes of Health, Research Triangle Park, NC. August 1993.

NTP (National Toxicology Program). 1993b. NTP Technical Report on Toxicity Studies of Pesticide/Fertilizer Mixtures Administered in Drinking Water to F344/N Rats and B6C3F₁ Mice. NIH Publication 93-3385. U.S. Department of Health and Human Services, National Institutes of Health, Research Triangle Park, NC. July 1993.

Poulin, P., and F.P. Theil. 2002. Prediction of pharmacokinetics prior to in vivo studies. II. Generic physiologically based pharmacokinetic models of drug disposition. J. Pharm. Sci. 91(5):1358-1370.

Poulin, P., K. Schoenlein, and F.P. Theil. 2001. Prediction of adipose tissue: Plasma partition coefficient for structurally unrelated drugs. J. Pharm. Sci. 90(4):436-447.

Reddy, M.B., R.S.H. Yang, H.J. Clewell III, and M.E. Andersen. 2005. Physiologically Based Pharmacokinetics: Science and Applications. Hoboken, NJ: Wiley-Interscience.

Rigas, M.L., M.S. Okino, and J.J. Quackenboss. 2001. Use of a pharmacokinetic model to assess chlorpyrifos exposure and dose in children, based on urinary biomarker measurements. Toxicol. Sci. 61(2):374-381.

Schober, S. 2005. National Health and Nutrition Examination Survey (NHANES): Environmental Biomonitoring Measures, Interpretation of Results. Presented at the Second Meeting on Human Biomonitoring for Environmental Toxicants, April 28, 2005, Washington, DC.

van de Waterbeemd, H., and E. Gifford. 2003. ADMET in silico modeling: Towards prediction paradise? Nat. Rev. Drug Discov. 2(3):192-204.

Vermeulen R., Q. Lan, L. Zhang, L. Gunn, D. McCarthy, R.L. Woodbury, M. McGuire, V.N. Podust, G. Li, N. Chatterjee, R. Mu, S. Yin, N. Rothman, and M.T. Smith. 2005. Decreased levels of CXC-chemokines in serum of benzene-exposed workers identified by array-based proteomics. Proc. Natl. Acad. Sci. USA 102(47):17041-17046.

Weis, B.K., D. Balshawl, J.R. Barr, D. Brown, M. Ellisman, P. Lioy, G. Omenn, J.D. Potter, M.T. Smith, L. Sohn, W.A. Suk, S. Sumner, J. Swenberg, D.R. Walt, S. Watkins, C. Thompson, and S.H. Wilson. 2005. Personalized exposure assessment: Promising approaches for human environmental health research. Environ. Health Perspect. 113(7): 840-848.

Wild, C.P. 2005. Complementing the genome with an “exposome”: The outstanding challenge of environmental exposure measurement in molecular epidemiology. Cancer Epidemiol. Biomarkers Prev. 14(8):1847-1850.

Yang, R.S.H. 1994. Toxicology of chemical mixtures derived from hazardous waste sites or application of pesticides and fertilizers. Pp. 99-117 in Toxicology of Chemical Mixtures: Case Studies, Mechanisms, and Novel Approaches. San Diego, CA: Academic Press.

Yang, R.S.H., H.A. El-Masri, R.S. Thomas, I.D. Dobrev, J.E. Dennison, Jr., D.S. Bae, J.A. Campain, K.H. Liao, B. Reisfeld, M.E. Andersen, and M. Mumtaz. 2004. Chemical mixture toxicology: From descriptive to mechanistic, and going on to in silico toxicology. Environ. Toxicol. Pharmacol. 18(2):65-81.

Yang, R.S.H., A.N. Mayeno, K.H. Liao, K.F. Reardon, and B. Reisfeld. 2005. Integration of PBPK and reaction network modeling: Predictive xenobiotic metabolomics. ALTEX 22(Spec. 2):328-334.

Zartarian, V.G., H. Ozkaynak, J.M. Burke, M.J. Zufall, M.L. Rigas, and E.J. Furtaw Jr. 2000. A modeling framework for estimating children’s residential exposure and dose to chlorpyrifos via dermal residue contact and nondietary ingest. Environ. Health Perspect. 108(6):505-514.

This page intentionally left blank.