Identifying and Filling Data Gaps
This section defines and explains the concept of data gaps and whether and how health assessors should attempt to address them.
During the PHA process, you may find that information about the community’s exposure to a site’s hazardous contaminants is incomplete. You should evaluate if the lack of information is critical to reaching a public health conclusion and therefore represents a data gap that should be addressed in your documents.
In some cases, health assessors will find that the available sampling data are insufficient for characterizing exposures in pathways of interest. Data gaps are incomplete information that prevent health assessors from reaching conclusions about potential and completed exposure pathways. Examples include:
- Data do not include all potentially affected media. For instance, the surface soil at a site of an unplanned release where the public has access was never sampled, or a drinking water well downgradient from a leaking underground storage tank was never sampled.
- Data do not include the potential contaminants of concern. For example, a drinking water well was sampled but not for the potential contaminant of concern.
- The amount of data was not sufficient to characterize exposure. For instance, the spatial and temporal extent of the sampling did not represent potential site exposures in an exposure unit.
When you conclude that data gaps need to be filled, ATSDR will typically recommend that other agencies or organizations, such as EPA, tribal groups, state agencies, or site owners, conduct sampling. The recommendation should be precise about exactly what is needed and what gaps must be filled. For instance, see the example recommendations to fill a data gap in the box below.
To address data gaps, you can seek additional data from all appropriate entities (e.g., federal and state agencies, site owners and their contractors) or recommend any of the following:
- A continued or new sampling effort.
- Modeling studies that predict contamination levels.
- An EI to collect biological or environmental data (see ATSDR’s PHAT EI Mini-Module pdf icon[PDF – 1,034 KB]).
Poor example: “Monitoring of trihalomethanes in drinking water is needed to complete the evaluation.”
Good example: “ATSDR recommends that the state environmental agency collect tap water samples from every household on Main Street that is connected to the municipal drinking water supply, and all of these samples should be analyzed for bromodichloromethane, bromoform, and chloroform using EPA Method 551.1. ATSDR has offered to review the sampling protocol to ensure the results can be used in the public health assessment process.”
In some cases, ATSDR will deem it appropriate to perform an exposure investigation pdf icon[PDF – 1,034 KB] to conduct additional sampling. In addition, state-of-the-art modeling tools may also be appropriate in cases where ATSDR has sufficient reason to believe these tools will help define exposures for certain time intervals (e.g., the past), help determine where sampling should be conducted, or supplement the available data (e.g., better define the range of possible contaminant concentrations) (see the modeling section for more information).
When filling gaps for the PHA process, health assessors may be asked to help design or review the design of environmental or biological sampling programs. This role is different from reviewing sampling data that have already been collected, which is discussed in other parts of this e-manual section.
Health assessors who have the opportunity to help design or review new sampling programs should be mindful of certain components to ensure the sampling program is designed in a manner that will inform the PHA process. Health assessors will want to ensure sampling programs are tailored to meet specific desired goals, consider site conditions, reflect local environmental settings, and lay out other necessary components discussed in this section.
Articulating clear DQOs is particularly important. Whether you are preparing the DQOs or reviewing the DQOs prepared by others, you should reflect on the seven steps of the DQO process described by EPA in its Guidance on Systematic Planning Using the Data Quality Objectives Process [PDF – 1,216 KB]pdf iconexternal icon. The first steps in this process involve specifying the sampling program goals. Consider the following:
- What is the principal study question (or questions) that the sampling will address?
- What public health value will be added by completing the sampling program?
- How much data are needed, over space and time, to answer the principal public health question (or questions)?
- What is the likelihood that the principal study question (or questions) will remain unanswered after completing the sampling?
Details of the sampling program will flow from clearly articulated DQOs. These will include the spatial and temporal domains of the sampling, the contaminants to measure, the sampling and analytical methods to use, and data acceptability criteria. You or the agency with whom you are working will subsequently document these details in a sampling plan, which should be reviewed by all key site stakeholders.
The sampling program should be informed by details of the local environmental settings, which may include hydrogeology for groundwater sampling, geology for soil sampling, and meteorology for air sampling. You can see more details about these and other aspects of the local setting in this e-manual’s Exposure Pathways section.
The sampling plan should document these items:
- Identification of the project’s technical goals, site conditions, expected outcomes of data collection, and acceptance criteria for the results.
- DQOs, which consider elements such as precision, accuracy, completeness, and representativeness.
- Environmental media to be sampled.
- Analytes to be measured within these media.
- Sampling and analytical methods.
- Proposed sampling locations.
- Sampling schedule (frequency and duration).
- QA/QC measures, such as use of duplicate samples and replicate analyses to characterize measurement precision; use of audit samples to assess measurement accuracy; analysis of field blanks to determine whether sampling equipment are contaminated; and specification of sample handling procedures, such as holding times and chain-of-custody requirements; and types of equipment calibration.
- Health and safety considerations for field personnel.
Once the sampling plan is reviewed and approved, the project will move from the planning phase to execution. For sampling conducted specifically to inform the PHA process, health assessors should try to be present during field work, whether to physically collect samples or oversee those who do. This hands-on experience will prove valuable for health assessors when they eventually review the data that were collected to fill data gaps.