National Academies Press: OpenBook

Monitoring Human Tissues for Toxic Substances (1991)

Chapter: 7. Program Design and Management Issues

« Previous: 6. Chemical Assay of Specimens
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

7
Program Design and Management Issues

INTRODUCTION

This chapter deals with the design and management of a new program. Effective management of a newly planned and expanded program for monitoring human tissues requires careful attention to administrative issues. The program must be properly located within the Environmental Protection Agency (EPA), funding must be adequate to accommodate program design and implementation, advisory and peer-review committees must be established for program evaluation and accountability, the professional and administrative capacity must be expanded. For proper operation and implementation of the program, its design must allow changes and flexibility in goals and scope, and some philosophic issues will require specific policy decisions by EPA (e.g., on managing the legal and ethical implications of collecting tissues and on reporting findings of abnormal values).

This chapter concludes with discussion of the importance of timely data analysis and reporting; the importance of cooperation and information transfer with other organizations; timely production and dissemination of reports to enable researchers, policy-makers, etc.; access to and descriptive analysis of the data; and the need for an orderly transition to prevent further deterioration and loss of institutional memory.

ADMINISTRATIVE AND AGENCY ISSUES

Administrative Location

The organizational location of the program is critical. The selection of an

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

agency to spearhead the national monitoring of human tissues for chemical exposures is not simple; indeed, the multiagency history of the NHMP and the current ambivalence within its parent agency regarding its future are clear indications that the match of program goals, potential benefits, and EPA mandates is not perfect. A successful monitoring program must be highly relevant to regulatory needs, but could and should—indeed must—serve a wide range of client programs and must not be dominated by any one of them. Critical aspects of institutional support should not depend on the necessarily changing priorities, requirements, and resources of small subprograms. The NHMP, unlike many other EPA programs, seems to have both a life of its own and a rationale that transcends any of EPA’s individual regulatory objectives. And some outputs of a human-tissue monitoring program should be of major interest at the highest levels of the host agency. However, the multi-purpose aspect of a national human-tissue monitoring program has meant that, to some extent, the NHMP is not an integrated part of any specific EPA regulatory program and hence is not at the top of any major agenda.

The committee has specific concerns about possible untoward effects of placing a monitoring program in any subunit with direct, major regulatory responsibilities. It firmly recommends that monitoring be kept strictly independent of regulation itself; both fact and appearance are important. Whether the present and past shortcomings in NHMP direction and funding are unavoidable results of a mismatch between agency and program is unclear. The committee concludes that considerations of input to policy, impact, visibility, and independence argue for a location at the highest feasible organizational level. From our consideration of alternative host agencies (based on general concepts of agency mission and resources, with explicit disregard of political and organizational barriers), it is clear that several logical possibilities exist.

Environmental Protection Agency

Advantages of this choice include continuity and a smooth implementation of whatever program changes are required; ease of communication with EPA environmental monitoring efforts; and direct input to a major class of regulatory users of the NHMP products. Such an activity might be placed in the Office of the Assistant Administrator for Research and Development. EPA staff have argued that the NHMP is primarily an environmental monitoring effort and that EPA has both the most appropriate mission and the greatest capability to manage such a program. If a strong and sufficiently independent Bureau of Environmental Statistics (or equivalent) is established, it might also be a suitable home for a monitoring program. Possible disadvantages of EPA

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

as the host agency include its demonstrated difficulty in providing direction and achieving intra-agency and federal support for the NHMP, systematic failure to approach research needs of the program, and failure to “market” results to other users and otherwise to stimulate and support efforts to develop nonregulatory benefits from the program. Other programmatic disadvantages include the lack of an in-house research focus that deals with the biomedical issues of tissue monitoring and a modest (although growing) orientation of present staff to those issues. The ideal host agency must provide technical leadership, as well as funding and administrative direction; that is best achieved with substantial and closely related internal research activity and expertise.

Agency for Toxic Substances and Disease Registry

This alternative would have the advantage that ATSDR is already committed to develop disease and exposure registries, and its mission is more oriented to the biomedical aspects of monitoring than is EPA’s. However, ATSDR’s activities focus on Superfund concerns, whereas national monitoring will continue to emphasize broad aspects of chemical use and exposure, rather than localized chemical pollution. Comparison data from national surveys are clearly needed by ATSDR for use in interpreting exposures to hazardous waste, but a major effort aimed at national exposure characterization might be too far removed from its mandate. The committee is also concerned that ATSDR is still rather new, growing and evolving rapidly, and perhaps suffering from the “growing pains” inherent in newness and growth.

Centers for Disease Control

Advantages of this choice include strength in epidemiologic research, as well as biomedical and chemical surveillance; demonstrated ability in supporting national monitoring efforts such as NHANES and blood-lead surveys; inhouse capability for chemical analyses; and a substantial research focus. Disadvantages include CDC’s distance from responsibilities tied to results of monitoring and the fact that CDC is not likely to be a major user of the findings.

Summary

After considering the most likely government units to house a human-tissue

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

monitoring program and hearing testimony at the January 24–25 workshop, the committee recommends that a national program for monitoring human tissues remain within EPA. However, we recommend that its administrative location within EPA be reconsidered and that it be moved to an organizational unit with EPA-wide responsibilities. A location that is geographically close to other programs and laboratories active in the relevant technical disciplines would facilitate important exchanges about methods, as well as followup of findings.

Funding

The critical resources in a program of monitoring human tissues include funding and expertise in appropriate scientific fields.

Sufficient funding (and a reasonable assurance that it will remain adequate over the next few years) is essential. The final budget for a program of this type should be determined after the specifications for the program have been formulated. Major factors that will affect funding requirements are the annual sample size, the set of chemical assays to be performed, the type of tissue to be collected, the size of the in-house staff necessary to monitor the program and analyze results, and requirements for research and development. We do not include in the budget for this program the cost of special studies that might be generated to explore some of the NHMP results in more detail; such studies should, in general, have separately identified funding. Although there is some flexibility in the funding—obviously the greater the funding, the more detailed the analyses that can be carried out—there is a minimum below which it is not worth while to have any program at all. We discuss below alternative budgets that can be considered for this program. The lowest level should be considered an absolute minimum. Furthermore, there needs to be a commitment within EPA to request at least the minimal funding through the indefinite future. The amounts discussed below represent the committee’s best judgment of what the various levels could support. These figures are not the committee’s judgment of what might be feasible, nor are they recommendations, per se.

An appropriate level of funding for a human-tissue monitoring program is difficult to determine, because it will depend so heavily on the specific program plan developed within the agency. Some guidance is available however, from EPA experience and from similar programs elsewhere. During its public session in January 1989, the committee heard several speakers suggest budgets as high as $50 million per year. Within the United States, a program at NIST for banking liver specimens alone is funded at $150,000. The National Ocean-

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

ic and Atmospheric Administration National Status and Trends program (consisting of both benthic surveillance and a study of mussels) archives specimens in support of those monitoring programs (human health is not a consideration) and is funded at $100,000 for the banking component only.

Support of the entire EPA National Human Monitoring Program has been $641,000 in FY `87 and $900,000 in FY 1988, FY 1989, and FY 1990. These recent figures for EPA are for external costs only and do not include salaries and overhead of EPA staff members, which in recent years seem to have averaged about two full-time equivalents (EPA, personal communication, June 18, 1990). This level of support at EPA has simply not been enough to sustain a program. The committee has found no one who believes that activity that remains at this level can produce useful results.

The committee firmly concludes that funding should either be increased enough to support a useful program or be eliminated. A clear decision to end the program would be preferable to seriously inadequate support. It would not encourage hopes in the community of present and future users about data that cannot be produced, it would be a clear statement that EPA does not accord human-tissue monitoring a high priority, and it would transfer institutional responsibility out of EPA and perhaps to other federal agencies. However, serious drawbacks to such a decision include the likelihood that no comprehensive, coordinated program would be developed elsewhere, the loss of skilled staff and institutional memory, and perhaps the destruction of specimens that have been banked and saved. The committee sees no likelihood that the information needs documented in Chapter 2 can be filled under this option, unless EPA can find some other willing sponsor. The chance of that seems remote, except possibly for ATSDR, but it is not clear that ATSDR in its present phase of rapid growth, development, and consolidation can spare the top management attention needed for a successful monitoring program. The committee urges that EPA consider termination of human-tissue monitoring only under the most compelling circumstances, and even then only after the fullest exploration of ways to ensure an orderly transfer to some other appropriate agency in less straitened circumstances. Each prior move of the program—from the Public Health Service at CDC to the EPA Office of Pesticides Program (OPP) in 1970 and from OPP to the EPA Office of Toxic Substances (OTS) in 1981 has caused serious disruption, and another move would almost certainly be equally disruptive, even with best efforts to reduce the damage.

It appears to the committee that support at a level of $3 million per year could be barely adequate to sustain the minimal activity needed to keep a program in long-term existence. That is approximately the level of support that was provided in the early 1980s, adjusted for inflation. It was enough

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

then to collect a statistically minimal number of specimens, perform batteries of the most critical chemical tests, maintain the preservation of samples saved from earlier years, and write occasional reports. It was not enough to expand sampling to important geographic areas or population segments not covered, to test for some important chemicals not in the basic panel, to undertake very much in the way of special studies, to prepare records and specimens in a form suitable for use by others, or to undertake the publicity and outreach programs needed to promote widespread use of the collected specimens and data. The committee recommends against a return to the $3-million level, unless careful study shows that circumstances have changed enough for a program of this size to avoid the problems of recent years. That seems unlikely, and the committee is concerned that support at $3 million per year would lead to another decline into unproductive scrabbling and mediocrity, followed in a few years by a new need to decide whether to revive or kill a moribund program. A minimal program might generate some useful data, especially if it were tailored to produce a few important annual reports for a Bureau of Environmental Statistics, or equivalent, but it could not rise to any level of distinction.

Two other critical, specific issues should be addressed by EPA before any decision to implement a minimal program: Is this level of support adequate to attract and maintain the “critical mass” of scientists and technical support staff needed to ensure sampling, testing, and analytic interpretation of the quality needed? Can a program of this size be adequately protected in times of financial exigency, so that it does not again start down the slippery slope on which the effects of decreased resources seem to justify still further decreases? We suspect that the answer to each question is no. Moreover, it would afford no opportunity to collect blood as well as fat specimens. Hence, we do not recommend long-term support at the minimal level, even if it would at first appear to meet the most critical data needs. A larger and broader view is necessary.

The next larger support level the committee considered is $5 million per year, exclusive of staff salaries and overhead. That should be adequate for collection and analysis of at least a statistically representative number of samples per year (see discussion in Chapter 4) plus associated support, research, and analytic activities. We have not undertaken detailed cost analyses, but study of EPA’s own history and the operations of other tissue monitoring programs suggests that $5 million per year could support a substantial flow of high-quality, policy-relevant information about chemical burdens in human tissue and that the two special problems just mentioned—maintaining a critical mass of scientific talent and ensuring stability of essential core funding during hard times for the agency—could be solved. It is still not munificent support.

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

Samples would continue to be small and spotty, so most resulting data would necessarily refer to the whole U.S. population and not more than four to six major geographic segments or population groups; data on say, individual states would not be available, nor would data on small groups of special interest, such as persons living near hazardous waste dumps or working in specific occupational categories (e.g., farmers). Resources for special analyses would be limited, there would be little or no surplus for new chemical tests on stored specimens, and the outreach activities necessary to promote the use of the program outside EPA would be severely constrained. But it might be enough to serve EPA’s policy needs and even to bring some critical distinction to the agency. Furthermore, it could be used to develop a solid base of competence, experience, and relevance to promise expansion in the future.

Even greater financial support—even up to the $25–50 million per year suggested by some heads of other agencies (as discussed at the committee’s workshop in, January 1989)—could be put to good use, given appropriate planning and the organizational setting and mission described elsewhere in this report. However, such allocations do not seem feasible now, so their implications are not explored here.

Whatever budgetary level is chosen, agency staff, in consultation with the scientific advisory body, should pay continuing attention to the competing demands of adequate sample size (within a context of periodic redesign of the sample to meet changing conditions and needs), protecting and measuring quality at all levels from the selection of subjects to the publication of reports, timely testing for chemicals of current interest, and outreach. A special, critical category will be research within the program—research on improving human-tissue monitoring itself (e.g., on methods of tissue preservation), on improving both the utility and the use of specimens and data, etc. (those matters are discussed in Chapters 46).

The committee recommends the prompt allocation of a full-time program manager and other staff with funds adequate for the planning and full design of a new program, concurrent with preparations for the absorption of the NHATS. A support level of at least $3 million per year will be needed for the continuation of current program activities and planning. That part of the program might take 12 months or more after release of this committee report.

We recommend that after the transition period, additional full-time staff be assigned and support be increased to at least $4 million per year for at least 2 years of consolidation. Further growth should be expected, but in a context of competition for funds based on successful implementation of the plan to that time.

Money is clearly essential, but two other kinds of in-house resources are necessary. The first is technical staff with expertise in statistics (primarily

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

survey methods) and in toxicology and related scientific fields. It is acceptable to use an outside contractor for data collection, as now for the NHATS, but the federal agency responsible for the program should be involved in major aspects of the program. It should develop the general plans for data collection, it should monitor the contractor’s work, it should plan and oversee data analysis, and it should be prepared to introduce modifications in the program that reflect changes in environmental concerns. That requires staff with deep knowledge of the subject matter and survey methods. (We are not implying that EPA does not have such resources; rather we are describing, in general terms, what an agency needs to carry out a program of the type suggested.)

The second kind of in-house resource that is needed is support at high administrative levels in the agency. That is, of course, necessary to ensure the required funding, but other critical factors are involved. Administrative support informs both the agency staff and the public that the agency has a firm commitment to the program.

The program will require a substantial and continuing research effort in support of improved monitoring, and a large fraction of that should be conducted intramurally or with much more intimate staff involvement than in NHATS contracts. Research activities should be developed in close consultation with other monitoring programs. Examples of subjects that will need research are improved methods for specimen collection and storage, and chemical analytic methods.

Finally, a means should be established to enable the agency to receive advice from outside experts in the pertinent scientific fields. Advice should be sought on such subjects as the performance of the survey (including both the quality of the data and the efficiency of operations), the need to add substances to the analysis because of new environmental concerns, methods of data analysis, and types of reports to be published and their frequency (are discussed below).

Science Advisory Committee

The committee has considered the needs for continuing oversight of a human-tissue monitoring program and concludes that such needs are substantial. The present EPA program has suffered seriously from lack of long-term attention from an outside advisory panel; it might otherwise now be in a far stronger position. Indeed, the committee senses that it is itself serving in some ways as a substitute for oversight that might have been provided in a more timely, effective, and useful way if the highest management levels of EPA had provided sufficient stimulus and internal support.

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

Such expert advice can be obtained in several ways. One is to have a formal advisory committee on monitoring of human tissue. Another is to engage the service of consultants to make periodic reviews of the progress of the work or of specific issues. A third method is to rely on a broad-based scientific committee advisory to EPA, making sure that the progress of the human-tissue monitoring program is often on its agenda. Other ways are ad hoc reviews by an outside body (such as the present committee), internal program reviews, and continuing attention from some existing advisory body. We are dubious of relying on a general advisory committee; such a committee probably could not devote enough time to a special topic, such as monitoring of human tissues, to make a useful contribution to the program.

The committee concludes that numerous criteria—including cost, breadth and depth of review, objectivity, program stability, timeliness of response, and a continuing need for fresh ideas—point strongly to a standing outside scientific advisory body to provide advice and program oversight to management levels of the program and EPA. Each italicized word requires comment.

“Outside” means outside the program and entirely or almost entirely outside EPA, aside from the services of an executive secretary and other support. Staff members from “client” government agencies might be members, as might various persons from academic institutions, industry, and public-interest groups. However, it should be made clear that each member is appointed as an individual and not to represent any organization, bloc, or set of interests.

“Scientific” means that nearly all members should be knowledgeable about one or more scientific and technical disciplines important to the work of the program, such as toxicology, biostatistics, survey statistics, biochemistry, and pathology. Any nonscience members should be selected on the basis of specific, major program needs, such as detailed knowledge about other current programs (surveys of the National Center for Health Statistics come to mind), knowledge about public dissemination of results, or a deep understanding of the policy implications of program findings.

“Advisory” implies that staff should pay substantial attention to recommendations, but make its own evaluations and, for good and clearly stated cause, follow other courses as necessary. The advice should include oversight of the following:

  • Program content.

  • Program management.

  • Program planning.

  • Resource needs.

  • Technical operations.

  • Timeliness and appropriate dissemination of results.

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

An outside scientific advisory body should also be constituted to facilitate a substantial two-way flow of information and ideas. A role in providing advice to the program is obvious, but role in conveying knowledge about the program and its products back to the range of communities represented on the body is also important.

However the charge is drafted, we believe that the outside scientific advisory body should have no other major responsibilities related to the program. We envision a schedule that initially calls for quarterly meetings, but tapers rapidly to perhaps annual meetings, so that committee assignment will not be seriously burdensome. Members must not, however, be diverted from their primary responsibility for oversight of the human-tissue monitoring program.

The present committee understands the limitations on appointment of new advisory bodies imposed by law, policy, and costs. A standing subcommittee of an existing advisory body might be an appropriate response to the need for outside oversight if most of its members (including the chair) were not members of the parent body.

Other Administrative Issues

Details of program structure and organization will necessarily depend heavily on a host of management decisions that the committee cannot foresee. This section presents some general issues that should be considered and expresses judgments about several broad aspects of structure and organization, but does not reach into details.

An important issue is the role of inhouse competence, both scientific and managerial. The committee recognizes the general competence and expertise of personnel who have been associated with EPA’s present program, but sees evidence that such general competence must be backed up within the program itself by substantial first-hand knowledge (training amplified by experience) in each of the technical disciplines relevant to a monitoring program, such as toxicology, biostatistics, pathology, and biochemistry. Contract support, consultants, and borrowed experts can be helpful, but they are not enough. The program requires technically competent staff who are intimately familiar with both the general program needs of EPA (and perhaps other agencies) and emergent problems and who are in close day-to-day contact with potential users of the collected data or stored samples. We are not arguing against appropriate use of contract support, but rather against contracting out so much of the management, technical competence, and programmatic understanding that critical organizational roots are damaged.

A second general issue is the need for some professional staff members to

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

be fully dedicated to human-tissue monitoring. The committee cannot specify full-time scientific disciplines or organizational capabilities, but reaches its conclusion from its perception that EPA’s present program has suffered severely from the diversion of time and talent when competing responsibilities have been perceived to be more acute (if not more important). Again, consultants and contract support can be important in program development and support, but they must not be used as substitutes for a core of fully dedicated professionals who have no other, competing duties.

The committee recommends that the program be designed in a modular fashion insofar as is feasible. It is not possible to foretell future strains on resources, trends in the availability of tissues, or needs for program output. It therefore seems prudent to organize a human-tissue monitoring program in such a way that new activities can be smoothly added and so that the most critical core activities can be maintained even when other activities are, for good cause, curtailed, suspended, or even terminated. For example, some panels of chemical tests might be identified as critical for collection every other year, but less critical for annual collection. Oversampling of some population segments might be suspended without reducing efforts to maintain basic coverage of the population as a whole. Preservation of a bank of saved specimens might be assigned higher or lower priority than other important program activities. Judgments about such matters will, of course, be important on initiation of a new monitoring program, but the recommendation here has to do with possible future contingencies and the design of activities so that desirable expansion or necessary retrenchment will cause the least possible disruption to core activities. This point could affect a wide range of initial and continuing program decisions, including such matters as the timing of chemical testing of new samples, the storage of preserved specimens (in case some must be discarded), and the design of new series of measurements or reports when projected followup efforts do not come to pass. The committee recognizes the fine line between prudent planning for contingencies and the unintended invitation of inappropriate budgetary cuts, but still recommends that this matter be given some attention from the outset.

It is most important for prudent, frequently up-dated planning to take advantage of resources that become available on short notice. Such planning might best be done, in a widely visible form, in annual budgetary submissions, even when it seems likely that resources will be tightly constrained.

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

IMPLEMENTATION AND OPERATIONAL ISSUES

Flexibility

Any broad program of human-tissue monitoring should be developed with a view to the need for developmental change, even given the most optimistic expectations regarding the adequacy of initial design. Several factors will make continuing or scheduled periodic redesign essential. As was discussed in some detail in preceding sections covering statistical design, sample collection, chemical analysis, and data analysis, the conditions that govern design choices will inevitably change. For example, demographic and environmental changes can affect the appropriateness of the sampling network, the list of chemical agents included or potentially measurable in the analytic scheme used, the possible choices of tissues based on analytic constraints, and costs. New chemicals will be introduced into use or will be newly recognized as environmental contaminants or risk factors. Advances in analytic technology might permit the recognition of new groups of agents or metabolic transformation products. Reduced assay requirements brought about by improved analytic technologies might permit alternative tissue-collection strategies that are not now feasible. Most monitoring programs can benefit from occasional updating, but the present program will almost certainly require well-managed evolutionary change, because of the open-ended nature of the scientific issues linking environmental contamination with tissue markers of exposure and health effects. Progress in such diverse fields as toxicology, statistical methods for sampling design and data analysis, exposure assessment, and environmental chemistry can be expected to affect the potential uses and value of human-tissue monitoring data.

Although developmental change is essential to the continued value of a long-term human-tissue exposure surveillance program, it is equally important that analytic stability and comparability of data be maintained. Established monitoring procedures (frequency of analysis, sampling design, and assay methods) must not be changed casually. Formal demonstration of comparability of results should precede any alteration in monitoring methods. The use of “probationary” methods in parallel with established methods is one way of managing the implementation of new methods in the monitoring program.

Some strategies for planning for program development are discussed in the preceding chapters. They can be summarized as follows:

  • Technical planning needs to be formalized for each year’s effort in sample design, sample collection, archiving, chemical assays for monitoring, chemical analysis for program development, and routine and exploratory data analy-

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

sis. Planning documents that define and resolve design issues should be issued regularly for review by scientific advisers and other agencies.

  • Specific budget items to support internal development leading to improved methods and to support external research directed at program needs should be identified.

  • Strong external technical input, arising from a scientific advisory panel and based on good communication with related programs and projects, will amplify technical planning by program staff. Equally important is the communication of program needs and findings to outside researchers (whether in other programs or agencies, academe, or the private sector) to stimulate followup.

  • Annual reports to Congress and reports for public use should discuss developmental needs and issues.

Legal and Ethical Issues

In addition to the legal and regulatory requirements associated with obtaining human specimens for environmental monitoring, many ethical issues should be addressed. General ethical concerns regarding environmental testing have been outlined elsewhere; some of them depend on the types of tests done to monitor exposures (Belmont Report, 1979; NRC, 1987). One of the important ethical issues to be considered in a monitoring program that uses human tissues is patient confidentiality with respect to the collected and reported data. Patient confidentiality must be safeguarded by handling and analyzing all data on the basis of coded numbers that are used to identify each specimen and related data. The key to identification codes might be broken if additional patient information or samples were needed and if it were clear that this use of identifying information would entail no material risk of further loss of confidentiality.

The codes might also be broken if it were decided to inform patients or their families of extremely high or otherwise atypical concentrations of chemicals detected on routine monitoring. Whether to provide such information is in itself an important ethical issue. In many cases, no therapy will be indicated or even possible for high concentrations of specific chemicals. In others, therapy of family members might be helpful if the conditions of the subjects exposure are such that family members might also have been exposed. Because interpretation of individual measurements of chemicals present in tissue would usually be difficult, a monitoring program must have a clear policy about whether and how such information is to be provided to patients, collectors, or patients’ physicians. That issue has been addressed in the NHANES

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

program and has been evaluated by the National Institute for Occupational Safety and Health (NIOSH) and discussed elsewhere (NIOSH, 1988). The NIOSH report demonstrates the difficulty related to disclosure of information under the circumstances of occupational exposures.

The National Institutes of Health is the government agency with primary responsibility for developing and maintaining ethical standards in scientific research that uses humans or human tissues. Living patients must furnish informed consent for the removal of blood or other specimens or tissues primarily for research or monitoring. The consent forms should specify why the samples are being collected, point out that saved (archived) specimens might be used for future projects, and indicate whether results will routinely be provided to the subject, the subject’s family, or the family physician. The persons from whom samples are taken should also be informed about the potential complications of specimen collection, including pain associated with obtaining samples (for example, the discomfort of needle biopsies used to obtain samples of fat or other tissue).

Except for blood, most monitoring will use tissues that are not removed specifically for research, but are remnants from diagnostic specimens, including autopsies and surgical procedures. Specific patient permission or permission from the next of kin is not required to perform research with such tissues. Such use, along with review of medical records, appears to be in an “exempt” category (OPRR Report 45CFR46, March 8, 1983). Most research institutions include a general permission clause in the surgical or autopsy consent forms to permit use of tissues and nonidentifying patient information in research. Use of such specimens requires strict adherence to both basic tenets of patient confidentiality and specific terms of the informed consent.

Those requirements and guidelines for human research not only apply to research supported by NIH, but also are used generally by institutional review boards (human-use committees) at academic institutions and in industry when such organizations review experimental protocols dealing with humans or with human tissues. In addition, a government committee is preparing uniform requirements for informed consent that will apply to all research supported by any agency of the federal government. European countries tend to be much stricter in their requirements for informed consent from patients whose tissues or medical records are used for medical research (Luepke, 1979).

Some aspects of the collection and use of human tissues for research or monitoring have been complicated by the commercialization of biotechnologic products that were developed from human tissues without specific permission. One commercial use resulted in a major lawsuit [Moore versus the Regents at the University of California et al. (88 C.D.O.S. 5320)] that has been considered by the California supreme court. The fundamental issue in the lawsuit

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

was whether human tissue removed from the body is the property of the person from whom it was obtained. The ruling determined that the tissue did not belong to the patient. Because of this and similar lawsuits, the Office of Technology Assessment recently reviewed the issue of ownership of human tissues (OTA, 1987) and proposed several alternatives for Congress to consider in future legislation. The legal issues associated with use of human tissues in research and transplantation have also been recently reviewed (Swerdlow, 1985).

For the short term, it is important that tissues used for biomedical research or environmental monitoring be separated from those used for the development of commercial products (Grizzle, 1985). Some organizations that supply human tissues for research require that the tissues not be used to make commercial products (Clausen et al., 1989), and a similar proviso should be applied to all tissues collected for environmental monitoring or supplied through the tissue archive of the monitoring program. The issues associated with the commercial use of human tissues are extremely complex and not easily solved, but may in the long run inhibit the availability of human tissues for noncommercial research or environmental monitoring, whether because of donor reluctance to provide tissues that might be used commercially or because of legislation that might inhibit the collection of tissues.

ANALYSIS AND REPORTING OF DATA

Types of Data Analysis and Reporting

The production of data sets that characterize nationally representative average tissue concentrations of specific chemicals is a necessary goal, but not in itself sufficient for a national tissue-based monitoring program. It is highly desirable that the monitoring program develop a plan for making maximal use of raw analytic data to pose and answer questions related to exposures of the U.S. population. However, such an open-ended charge may have adverse consequences; for example, the reporting of results may be delayed while exploratory data analyses are conducted, or the more sophisticated, nonroutine data analyses may be deferred, possibly indefinitely. The plan for each year’s monitoring effort should include data-analysis, analytic subprojects (i.e., what questions are to be addressed and what approaches and strategies would be used), and timelines for each activity. Analyses that are well defined should result in early reporting of findings; analyses that require method development or input data beyond the basic chemical-analysis data set might be reported less rapidly or less often. At a minimum, however, descriptive summary analy-

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

ses of the data should appear with the same frequency as the chemical analyses.

Descriptive analysis should include characterization of central tendency, confidence intervals, and the frequency or prevalence and magnitude of extreme values for the collection set as a whole and for target groups of samples defined by such variables as geography, age, and sex. Results should be compared with those of previous years to identify statistically significant trends. Such comparisons might be difficult or yield uncertain results if sampling methods or analytic procedures change. This most basic (“level 1”) data analysis should be presented with information on sample collection, routine chemical assay, and quality control in a report that can form the basis for an annual report of activities. Examples of additional analyses not likely to be completed on the same schedule as level 1 analyses are those that use extramural data (such as toxicologic data or environmental-contamination data) to address questions of dose or routes of exposures and multivariate discriminant functions or factor analyses to assess combinations of results that might be more sensitive than results on individual chemicals (“level 2”). Still more remote in time would be data analyses that require a multiyear data set or input data not currently available (“level 3”). For level 2 analyses, it would be possible to develop explicit plans and schedules, whereas level 3 efforts would depend on the outcome of efforts not part of the NHMP itself and therefore might not be subject to definite schedules. Program planning carried out continuously would include review of data-analysis plans and status and would thereby differentiate realizable goals from goals that are impractical within the constraints of the program.

Integration of Data Sets with Data from Other Sources

Data from human-tissue monitoring cannot in general be fully interpreted without other information regarding tissue concentrations, patterns of exposures, and the metabolism and toxicology of individual chemicals. Reporting of those should be systematized, but this will require close collaboration of the statistical analysts with persons providing the other information. Comparison data, especially reports of tissue concentrations of chemical agents, should be sought continuously in the scientific literature and in technical reports from complementary programs, such as NHANES. Environmental-survey results produced by other EPA programs will also merit detailed evaluation, particularly if NHMP findings indicate exposures at variance with predictions from environmental data. Information that would be useful in establishing quantita-

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

tive links between tissue concentrations and exposures is another critical category of comparison data, because risk assessments are generally based on exposure magnitudes. Where data are needed for NHMP purposes and are unavailable, the needs should be identified in annual reports, and other efforts should be made to stimulate research. Strong interagency and intra-agency collaborations in planning and exchange of data will be required for addressing measurement objectives, will facilitate the analysis of monitoring-program results, and will result in a program that is more useful to cooperating agencies.

Reporting Frequency

It is a key requirement that the program produce timely reports on a regular basis. At a minimum, an annual report of level 1 analyses should be produced within a year of completion of the collection of samples. Additional reports addressing more specific questions regarding exposures or trends among chemicals or population groups, or more fundamental issues of exposure assessment, should also appear frequently, but perhaps with a little more lead time.

Review Process for Reports

Draft reports will almost automatically have internal review within contracting organizations, by NHMP program managers, and by EPA quality-assurance coordinators. Additional extra-agency review of draft reports is desirable, especially peer review by individual scientists, by the science advisory group(s) established for the program, and perhaps by other persons with special expertise. Reviewers should be acknowledged in final reports. The program should result in a continual flow of significant findings and summary results suitable for publication in the peer-reviewed scientific literature. Periodic workshops could be convened to obtain wider peer review of the program and its reports.

Scientific reports of findings should be supplemented by documents that establish program goals, plans, and technical strategies and that in other ways summarize the scientific framework of the program. Because application of findings to individual instances of chemical exposure and possible harm is anticipated as an important use, guidelines for interpreting NHMP findings should be formulated. A “user guide” that addresses such issues as representative sampling, population and individual variability, ways of relating dose to

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

tissue concentration and perhaps to health effects, and other concepts necessary for the proper use of the NHMP resource would help to promote the responsible application of results and advise nontechnical users as to the kinds of questions that should and should not be addressed with data from the monitoring program.

Production and Dissemination of Reports

An important role in planning is to lay out a well-defined process for producing a range of outputs with associated responsibilities and a schedule. The schedule should be widely and continually publicized and should be relaxed only under the most compelling circumstances. The full-time scientific staff should bear most of the responsibility for meeting the schedule and should recognize that timely, high-quality reports on important matters are a sine qua non.

The program should have a specific outreach component that is the primary responsibility of at least one person. It is warranted because of the multiuse nature of the program, the wide-ranging interest in the resulting data, and the clear indications that more passive or low-key approaches to publicizing program reports have failed to reach some critical target groups. Technical reports, scientific publications, presentations at scientific meetings, and other normal modes of communication within some segment of the scientific community are necessary but insufficient parts of program “marketing.” Additional steps—such as newsletters, preparation of feature articles for nontechnical publications, and presentations to interested nontechnical groups, including legislators—might all be useful. Some care must be taken to define the larger constituency of the program and to update and expand mailing lists or other tools for reaching potential users of findings.

Lines of communication to academic research investigators must be improved. That group not only is likely to be able to put monitoring-program data to important and innovative uses, but also is a likely source of research findings of value to NHMP projects. National research conferences and even a modest extramural-research funding program might be helpful in heightening awareness of the NHMP in universities, colleges, and other research institutions.

A human-tissue monitoring program should be designed as a multiple-user service activity. That creates substantial obligations for assisting users to understand what the program does and does not provide, for timely analysis and publication of results, for specific and helpful guidance in access to archive specimens, and for active marketing of products. The committee recommends that those matters receive high priority in planning and implementation of a new program.

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

COOPERATION AND INFORMATION TRANSFER WITH OTHER ORGANIZATIONS

A tissue monitoring and archive program must cooperate and communicate with other branches of EPA, other government agencies, the academic and private sectors, and foreign environmental programs. For example, in the design of the program, the sponsoring organization should use information on tissue collection, shipping, and storage, including proper storage containers. Much information on those topics has been developed for the Federal Republic of Germany (FRG) environmental monitoring program, the archive program at NIST, and the Canadian wildlife-tissue archive. In addition, new information on environmental monitoring is being obtained and evaluated in Sweden and Japan. Those programs might also supply valuable information on analytic methods to aid in the design of a new program in EPA.

Not only are such cooperation and information exchange important in the design of human-tissue monitoring, but continuing information exchange is critical to the efficient operation of the new program. Such communication should not just rely on “published” information, but should take place in part through less formal channels. For example, if the FRG program identifies a chemical in human tissue, rapid information exchange will permit an expeditious search of the chemical in human tissues of the U.S. population. Similarly, the program should be a leader in national and international conferences on tissue monitoring and archiving.

Continuing cooperation and communication of the program with other programs and within EPA and other government agencies are critical. Strong interagency coordination among federal exposure data bases will enhance federal monitoring efforts. For example, if EPA is acquiring air data for a particular location, a corresponding tissue program might be coordinated. If FDA has a market basket pesticide survey for a specified area, the resulting data could be calibrated against tissue concentrations. Other branches of EPA and other federal agencies should use data from the monitoring program, and they should have input without introducing delays as to how the data are collected, analyzed, and reported. They also should have access to the data as rapidly as practicable.

The committee thinks that there might be special value in the joint development of a small set of measurements to be made in similar ways across a broad range of public and private programs or a means of establishing comparability among programs that could lead to a worldwide data base for environmental toxicants that migrate across long distances and across national boundaries. This also might involve a Bureau of Environmental Statistics. Some programs that might be involved in such an effort are the National Contami-

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

nant Biomonitoring Program of the Fish and Wildlife Service; the National Pesticides Monitoring Program; various National Oceanographic and Atmospheric Administration programs; and efforts within academia, such as research on organochlorine pesticides in human milk.

TRANSITION

Our main recommendation is that EPA’s present program for human-tissue monitoring be phased out as soon as a program based on blood specimens with limited adipose-tissue specimens is implemented. That leaves open the questions of program activities and support levels while the replacement is being developed. We believe that historical continuity can be valuable; that the longer the past series, the greater the value; and that many of the most critical operational changes (e.g., in freezer temperature and in type of storage containers) can be implemented almost overnight. In addition, samples already collected are scheduled for critical analyses that should not be unduly delayed.

Taken together, those points indicate that the present program will be of sufficient value to its successor to merit temporary continued support at a level consistent with its functions of several years ago. The presence of a complete gap of a year or longer—even several months—would risk major damage to public and legislative support, to internal support (if funds, once reallocated, cannot be recaptured), to whatever parts of the collection reporting network are to be preserved, to the maintenance of critical core technical capabilities (including contractors), and to the integrity of data and specimens already in hand. We are concerned, in short, that a too-prompt implementation of the first part of our primary recommendation—the orderly termination of the present program—will create avoidable and expensive impediments to implementation of the second part, a new program. In our view, the two parts are not separable, and to attempt to separate them is explicitly contrary to our recommendation.

However, if interim activities should return to the level of several years ago, budgets must keep pace. We recommend continued support at an annual level of at least $3 million until the present program can be dissolved and its talents and resources absorbed into a new program. The budget during the transition period can probably be different from the funding necessary to sustain an operating program. During the first year (or possibly two) of the transition, the main costs will be for planning and testing of the new program and for continuing the NHATS. As the new program is implemented, its costs will become the dominant part of the budget, and the planning costs will shift to oversight of the program and research and development.

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

SUMMARY AND RECOMMENDATIONS

The committee’s primary recommendation is that the present pro gram for human-tissue monitoring be phased out as soon as a blood-specimen program can be implemented.

Design and management of the proposed program will require careful attention to many aspects, including administrative issues, funding, and data analysis and dissemination.

Administrative Location

NHMP’s rationale and objectives transcend any of EPA’s individual regulatory objectives. Considerations of input to policy, impact, visibility, and independence argue for a location at the highest feasible organizational level.

After considering the most likely government units to house a hu man-tissue monitoring program, the committee recommends that a national program remain within EPA. Its administrative location within EPA should be reconsidered, and it should be moved to an organizational unit with EPA-wide responsibilities.

The program would benefit from geographic proximity to other relevant programs and laboratories.

Funding

Sufficient funding (and assurance that it will continue) is essential.

Funding either should be increased enough to support a useful pro gram or be eliminated.

Elimination of the program would be preferable to seriously inadequate support. However, such an action would risk the likelihood that no comprehensive, coordinated program would be developed elsewhere, the skilled staff would be lost along with institutional memory, and specimens that have been banked and saved might be destroyed.

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

The committee recommends the prompt allocation of a full-time program manager and other staff with funds adequate for the plan ning and full design of a new program.

This should be concurrent with preparations for the absorption of the NHATS. At least $3 million per year will be needed to continue the current program activities and planning.

After the transition period, additional full-time staff should be as signed and support increased.

At least $4 million per year will be needed. Further growth should be expected in a context of competition for funds based on successful implementation. Technical staff with specific expertise will be needed, as will support at high administrative levels in EPA. The program also will require substantial and continuing research efforts in support of improved monitoring.

Scientific Advisory Committee

Numerous criteria, including cost, breadth and depth of review, ob jectivity, program stability, timeliness of response, and a continuing need for new ideas, point to a standing outside scientific advisory committee.

The body should be outside EPA; nearly all members should be knowledgeable about one or more scientific and technical disciplines important to the program; and the body should be constituted to facilitate a substantial two-way flow of information and ideas. The body should have no other major responsibilities related to the program.

Other Issues

Some professional staff members should be fully dedicated to the program, without competing duties.

The program should be designed in a modular fashion.

This will permit new activities to be added smoothly and core activities to be maintained even if other activities are curtailed.

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×

Analysis and Reporting of Data

At a minimum, descriptive summary analysis of data should appear with the same frequency as chemical analyses. Descriptive analysis should include characterization of central tendency, confidence intervals, and frequency or prevalence and magnitude of extreme values for the collection set as a whole and for target groups of samples defined by variables.

Data reporting should be systematic and conducted with close collaboration between statistical analysts and persons providing other relevant information. Comparison data should be sought continuously from scientific literature and complementary programs.

Continuing cooperation and communication among EPA programs and other agencies are critical.

The program must produce timely reports regularly. At a minimum, an annual report of level 1 analyses should be produced within a year of completion of sample collection.

In addition to internal review, peer review and review of reports by the science advisory group is desirable. Reports should be supplemented by documents that establish program goals, plans, and technical strategies.

Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
This page in the original is blank.
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 143
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 144
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 145
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 146
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 147
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 148
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 149
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 150
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 151
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 152
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 153
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 154
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 155
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 156
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 157
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 158
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 159
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 160
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 161
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 162
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 163
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 164
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 165
Suggested Citation:"7. Program Design and Management Issues." National Research Council. 1991. Monitoring Human Tissues for Toxic Substances. Washington, DC: The National Academies Press. doi: 10.17226/1787.
×
Page 166
Next: References »
Monitoring Human Tissues for Toxic Substances Get This Book
×
Buy Paperback | $50.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

The National Human Monitoring Program (NHMP) identifies concentrations of specific chemicals in human tissues, including toxicologic testing and risk assessment determinations. This volume evaluates the current activities of the NHMP; identifies important scientific, technical, and programmatic issues; and makes recommendations regarding the design of the program and use of its products.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!