Early childhood and maternal exposure to pollutants are important aspects of the NCS. The long-term goals of exposure characterization for the NCS are to understand and characterize “true” exposures using a combination of multimethod assessments and to use the metrics to test various NCS hypotheses. , adapted from the NCS Research Plan (NCS 2007c
), shows three components that are necessary for estimating exposure or dose: environmental measurements, biological measurements, and questionnaires. These components are greatly aided by computational models for exposures and biological doses. Further, three levels of data are associated with exposure characterization and estimation within the scope of NCS: community-, household-, and individual-level data, dependent on time, locations, and activities. At each level, measurements need to be taken or values estimated for appropriate variables (NCS 2007b
). Thus, for each child in the study, the collection of environmental data will be associated with “fields of influence,” for example, adjacent to the child, the household of the child, and the surrounding community. Data from all levels can be combined and used to establish the exposure or dose for each participating child.
Figure 2 (A) The three components that are necessary to estimate “true” exposures or doses to selected subjects (based on the NCS Research Plan). (B) Increase in uncertainties when data at a coarse level are used to supplement segment- or individual-specific (more ...)
For many persistent organic chemicals and some metals (i.e., chemicals with long biological half-lives), validated biomarkers will be used to determine the levels of persistent compounds in blood, urine, and the like, before, during, and after pregnancy (Barr et al. 2005
). The biospecimen sample collection will switch from the mother to the child after birth. The results from biomarker measurements can indicate whether or not an exposure has occurred for this suite of compounds, but for most agents they cannot identify the source or route of exposure. Eventually, exposure reconstruction modeling tools may employ these data to test NCS hypotheses in conjunction with other exposure-related information (e.g., activity patterns, home measurements) (Georgopoulos et al. 2009
). However, one must remember that exposure reconstruction is more feasible for chemicals with a long half-life, a condition that does not represent the full range of agents that can affect exposure–response relationships within NCS populations.
One component of the NCS that provides information to support environmental exposure characterizations will be the examination of the mother and child’s home using observation and questionnaires/diaries. Information will be collected at an initial home visit and updated periodically throughout the study. The observations will provide both qualitative and quantitative information and guidance on the types of sources or environmental insults that might actually be experienced by the children from contact with contaminant source emissions or from other specific types of contaminant contact related to their activities. The questionnaires and details on their potential use with biological and chemical markers of exposure are found in the NCS design document (NCS 2007b
Actual environmental measurements in the home will include settled dust and wipe samples, specific water samples from a subset of homes, and possibly air samples within sub-studies. These data will be used to help estimate a child’s individual exposure over time. The current list of measurements to be completed in the households is summarized in the NCS Research Plan (NCS 2007a
) and includes multiple compounds, multiple media, and multiple exposure pathways [see Supplemental Material, Table S1 (doi:10.1289/ehp.0900623.S1 via http://dx.doi.org
For the communitywide environmental measurements, a number of plausible measurement techniques can be implemented during the NCS. However, given the complexity of the study and the fact that resources needed to conduct environmental field measurements are quite extensive, the NCS has determined that most environmental data must be retrieved primarily from available databases.
The cost and the human resources needed to establish new community monitoring networks within the NCS data collection plan, which would include each segment and aim to characterize all contaminants of importance over 2 decades, are considerable, and the associated effort would be daunting. Thus, a coherent set of local ambient environmental data for Queens is not expected to be available to test all study hypotheses using the monitoring data [see Supplemental Material, Table S1 (doi:10.1289/ehp.0900623.S1)]. Short-term sampling will be implemented and will provide critical information on contaminant and exposure variability within Queens and the other NCS locations. These data alone, however, will not provide a comprehensive environmental and microenvironmental characterization for cumulative exposure estimates of each NCS participant. Therefore, it is necessary to identify the overall multiexposure route processes based upon a segment’s location and the proximity of outdoor sources to the individual households, as well as upon the presence of indoor sources.
Characterizing the impact of local environmental factors requires thoughtful and careful application of national, regional, and local data (), which need to be linked in a consistent manner, even though different data collection approaches and protocols may have been used for each medium, pollutant, or scale. For example, when national-level data sets are used to estimate exposures at an individual or segment level, the uncertainties associated with application of such data are high, as shown in . Having an understanding of local demographic, geographic, and contextual information about neighborhoods is critical in order to adapt or for adopting national sampling plans to reflect local circumstances.
The QVC has developed an approach for obtaining the information and identifying environmental databases that can be used to establish baseline or background levels in the county for specific pollutants and media; this approach will eventually be available for use in comprehensive models for exposure–dose characterization within the segments (Georgopoulos and Lioy 2006
). The above databases would provide a path forward for use of data collected by long-term measurement programs available from a variety of organizations. The goal would be to incorporate the data into long-term characterizations of exposures for the participating children.