U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Research Council (US) Panel on Collecting, Storing, Accessing, and Protecting Biological Specimens and Biodata in Social Surveys; Hauser RM, Weinstein M, Pool R, et al., editors. Conducting Biosocial Surveys: Collecting, Storing, Accessing, and Protecting Biospecimens and Biodata. Washington (DC): National Academies Press (US); 2010.

Cover of Conducting Biosocial Surveys

Conducting Biosocial Surveys: Collecting, Storing, Accessing, and Protecting Biospecimens and Biodata.

Show details

2Collecting, Storing, Using, and Distributing Biospecimens

Social scientists have long experience with population surveys, but the collection of biospecimens as part of a survey protocol requires different technical and logistical skills and introduces complex legal and ethical issues. Additional, and often unforeseen, costs in terms of both money and time frequently must be borne as well. To be sure, epidemiologists have a long tradition of collecting biological data (often mimicking a clinical setting in the field) along with social and demographic data (that are often less rich than social surveys) but typically they are not designed to be representative of a larger population universe. Furthermore, social scientists generally have expectations regarding data sharing and access that differ from those of epidemiologists, clinicians, and medical researchers.

Adding the collection of biospecimens to social surveys provides the opportunity to identify or test biomarkers that can address questions in many areas of interest, from susceptibility to disease to measures of environmental exposure to a wide range of compounds. It is important, however, for proposed bio-markers to be aligned properly with the goals, hypotheses, and concepts of the research. Each team of investigators will need to consider carefully numerous questions that will guide their decisions regarding the collection, use, and storage of biospecimens: What biospecimens will best advance the research questions or hypotheses being investigated? What kinds of specimens will provide the most reliable measures (and be practicable in the field)? What constraints are involved in transporting specimens to a laboratory or tissue repository (in time)? What kind of preparation needs to be done in the field (and is it feasible)? What safety issues need to be addressed? Which assays should be done first? In addition, a host of logistical questions must be considered by the participating scientists and their Institutional Review Boards (IRBs). These include how the specimens will be collected, managed, and stored; how access to them will be regulated and monitored; how they will be used to extract relevant bio-markers; and whether and how the laboratory data and survey information will be made available to scientific collaborators, the broader scientific community, and the public.

There is no single formula for adding the collection of biospecimens to a survey; each investigation will have different requirements that will have to be weighed and balanced against constraints of budget, time, field conditions, and participant burden. Significant advance planning, piloting, and revision will be required, and even then it is wise to expect the unexpected.

These cautions are particularly important given the rapid pace of technology development both with respect to the collection of specimens (e.g., blood, urine, saliva, or hair) as well as the ability to analyze data derived from such specimens (ranging from blood glucose levels to C-reactive protein to mercury levels to DNA). The prime example here is the analysis of genetic markers. Just a few years ago, analysis of large numbers of biological specimens was limited to examining a small number of candidate genes or, at best, a few thousand genetic markers. But the advent of microarrays that allow the measurement of expression levels of unprecedented numbers of human genes in a single experiment or the profiling of a million single nucleotide polymorphisms (SNPs) across the genome began to change the way scientists and the IRBs that oversee their projects conduct their work. For example, pooled data from genome-wide association studies (GWAS), representing the genomes of multiple individuals, were viewed for some time as acceptable for public release. But when recent work on forensic analysis of DNA samples showed that the presence of a single individual could be detected in a large pool of such samples (Homer et al., 2008), researchers and policy makers at the National Institutes of Health (NIH) reconsidered and changed data release policies.

Concerns about data release and protected health information have been compounded by the rapid pace of development of next-generation DNA sequencing technologies. Sequencing the first human genome was a 15-year project that cost billions of dollars; new technologies, however, allow sequencing of human genomes in times that are on the order of 1 month at a cost of less than $100,000, and technology advances may reduce the cost to $1,000 or less. This capability raises the prospect of having to deal with unprecedented amounts of personal information—the entire genome sequence from large numbers of individuals. While such data may have great potential for the discovery of biomarkers and functional studies, dealing with the data and their implications for privacy and protection of human subjects will require addressing many as yet unanswered questions.

The focus in this chapter is on the specimens themselves, not the data derived from them (discussed in Chapter 3), although clearly the kinds of data that ultimately can be obtained are constrained by the specimens that are collected, the quality of the collection and storage procedures, the timing of the assays, and the integrity of the procedures. The chapter begins by describing some documents that provide guidance regarding best practices for investigators. It then reviews cross-cutting considerations for studies that include bio-specimens. The subsequent sections offer a detailed discussion of issues related to collection, use, storage, custodial responsibility and ownership, and access and distribution.


Fortunately, researchers undertaking the collection and analysis of biospecimens for the first time do not have to reinvent the wheel. First, an increasing number of social science investigators have hands-on experience with collecting, storing, using, and distributing biospecimens. A simple piece of practical advice is to talk with them. Second, documents that describe recommended procedures and laboratory practices are already available. Although they were not developed with social surveys specifically in mind, these protocols have been field tested and approved by numerous IRBs and ethical oversight committees. These best-practice documents are updated frequently to reflect the growing knowledge and changing opinions about the best ways to collect, store, handle, and distribute biological specimens. They are useful places to begin.

The panel identified three documents that it believes are most relevant to the concerns of newcomers to biospecimen collection. Although these documents focus on guidelines for biospecimen repositories, they also provide more general guidance and important reminders for consideration during the planning process. They will be most useful for identifying factors that need to be considered in choosing a repository or in establishing even a small biospecimen archive.

First is 2008 Best Practices for Repositories: Collection, Storage, Retrieval, and Distribution of Biological Materials for Research, prepared by the International Society for Biological and Environmental Repositories (ISBER) (International Society for Biological and Environmental Repositories, 2008). ISBER was formed in part to develop effective strategies for the long-term storage of biological specimens; it “fosters education and research and promotes quality and safety in all activities relating to specimen collection, storage and dissemination” (International Society for Biological and Environmental Repositories, 2005, p. 5). The document covers repository organization, management, and facilities; storage equipment and environment; quality assurance and quality control; safety; training; tracking of biological materials; packaging and shipping; specimen collection, processing, and retrieval; and legal and ethical issues for human specimens.

Second, in 2002 the National Cancer Institute (NCI) set out to understand the quality and characteristics of biospecimens used in cancer research and to develop a set of principles for their collection and handling. A 5-year process of research, workshops, and feedback led to the publication in 2007 of National Cancer Institute Best Practices for Biospecimen Resources (National Cancer Institute, 2007). Less detailed than the ISBER publication, the NCI guidelines do not offer specifics, such as the temperature at which tissue samples should be stored or the precise sorts of labels that should be used. Rather, they provide “salient guiding principles that define state-of-the-science biospecimen resource practices, promote biospecimen and data quality, and support adherence to ethical and legal requirements” (National Cancer Institute, 2007, p. 1).

Third, in 2007 the Organisation for Economic Co-operation and Development (OECD) released OECD Best Practice Guidelines for Biological Resource Centres (Organisation for Economic Co-operation and Development, 2007). These guidelines are broader than those of NCI in that they address more types of biological specimens, including viruses, bacteria, and other microorganisms, as well as samples taken from human subjects. They are also broader in the sense that they represent best practices from 30 countries rather than being drawn primarily from the best practices of U.S. institutions. However, they take the same general approach as the NCI guidelines in that they provide guiding principles rather than specific suggestions.1

Collectively, these documents provide excellent advice on how to store biospecimens in ways that preserve them,2 ensure adequate documentation, and protect confidentiality. However, they do not address questions that are most closely related to the design of the research itself, such as choice of biospecimen (e.g., blood, urine, saliva), choice of biomarker, and choice of assay.3 Another important distinction for biosocial researchers is that the data archive (i.e., the collection of data derived from the specimens, as well as the data from the survey) is likely to be maintained separately from the specimens themselves, while documentation about the specimen collection and survey protocols may be archived in yet another location. Unlike the researchers addressed in the best-practice documents, therefore, social scientists are likely to use a biorepository as a facility for the storage and distribution of specimens, not as a partner in the substantive or intellectual concerns of the study or as an entity with a claim to ownership of the specimens. These differences translate to a different set of challenges for data security and the protection of confidentiality from those covered in the best-practice documents. That said, the documents provide useful guidelines for the choice and evaluation of a repository for biospecimens: even if the biorepository is not an intellectual partner, it must track and protect information in its custody.


Ethical, Legal, and Policy Issues

Safeguarding the rights and safety of study participants is of paramount importance. Honoring promises regarding the privacy and confidentiality of participants and ensuring that all procedures conform to what has been established during the informed consent process are fundamental to ethical research. The damage that can be caused by breaches of confidentiality—possible at many stages of research using biospecimens—can be devastating. Policies and protocols for using biospecimens properly in scientifically sound research and for ensuring adherence to federal, local, state, and international laws governing the collection, storage, and use of biospecimens by all members of the research team and its collaborators and contractors are fundamental. Chapter 3 examines issues related to confidentiality and the sharing of biological and social data, while Chapter 4 contains a more detailed discussion of issues related to informed consent.

Informed consent, privacy, confidentiality, and identifiability in genomic research are all closely interrelated (Lowrance, 2006a). All are relevant to decisions about deidentification and coding of data (and thus to protection, which in turn allows for dissemination). Privacy refers to the protection of an individual from unwanted intrusions, including the acquisition of information about that individual. It is a general protection against other individuals, groups, and society as a whole. By contrast, confidentiality refers to the protection of information about an individual that has already been provided willingly to one party; if the information is confidential, the recipient is responsible for ensuring that it is not released to others without the donor’s permission. Identifiability is the ability to associate data with a particular person. The identifiability of data can be thought of in terms of a spectrum from data that are impossible to identify, to those that can be identified as possibly being linked to a given individual, to those for which that linkage is known with certainty.

A primary consideration is that the privacy of study participants should be respected and protected. Sound scientific research involving biospecimens depends on protecting the privacy of the participants who contribute them and the confidentiality of their data. Yet many of the very scientific and technological advances that make the collection of biological specimens so valuable—progress‐ in genomic and proteomic science, the sequencing of the human genome, the culture of rapid release of genomic data, and the increasing use of Internet-based searchable databases containing those data—increase the risk of potential breaches of confidentiality and make it increasingly important to anticipate and prevent such breaches (National Cancer Institute, 2007). Advances in digital computing, communication, and storage technologies also are linked with a new world of research characterized by immense data sets, new forms of interdisciplinary collaboration, and unprecedented levels of data sharing among researchers, all of which raise the risk of breaches throughout the data collection, analysis, publication, and distribution process (see National Academy of Sciences, National Academy of Engineering, and Institute of Medicine, 2009).

While summary-level data are of great value, biosocial research often requires that individual-level survey data be linked to each biological specimen to facilitate the analysis of biodata derived from those specimens. However, the unrestricted release of both the individual survey data and the biodata poses potential risks for breaches of confidentiality. Although the public release of nominally deidentified data used to be widely accepted, the concept of deidentification needs to be reconsidered when the data can represent an individual’s entire genome sequence or the state of a million or more variant positions along the genome, possibly together with various social, economic, psychological, or physiological characteristics. There is as yet no way of knowing what the frequency and consequences of breaches in data security might be, but it appears clear that they have the potential to lead to individual or group discrimination or stigmatization, as well as to diminished public trust and reduced participation in and support for research. It is crucial to have policies and protocols in place to prevent the unauthorized release of sensitive information and to respond effectively if it occurs.

At the federal level, constitutional, legislative, and administration protections of privacy apply to situations in which biospecimens are collected, stored, analyzed, or disseminated. In at least one case, the U.S. Supreme Court implied that the Constitution might guarantee individuals some level of “informational privacy” (Whalen v. Roe, 429 U.S. 589 [1977]). The Court has not developed that right further, however, and in any event, it would apply only to actions by governments and not private organizations, such as most health plans (Jones and Sarata, 2008, p. 17). The federal court finding that is perhaps most relevant to the issue of genetic privacy is Norman-Bloodsaw v. Lawrence Berkeley Laboratory. A group of workers at Lawrence Berkeley Laboratory (LBL) complained that their privacy rights had been violated when the laboratory used their stored blood samples to test for pregnancy, sickle-cell anemia, and sexually transmitted diseases without their consent. The district court sided with LBL, but the U.S. Court of Appeals for the Ninth Circuit disagreed and sent the case (later settled out of court) back to the district court. The Court of Appeals found that, although the employees had originally agreed to provide blood samples, “the ensuing chemical analysis of the samples to obtain physiological data is a further intrusion of the tested employees’ privacy interests.” The court added, “One can think of few subject areas more personal and more likely to implicate privacy interests than that of one’s health or genetic make-up,” and suggested that the tests “may also be viewed as searches in violation of Fourth Amendment rights.”4

Among the federal statutes that may apply to genetic privacy are the Privacy Act of 1974 (5 U.S.C. § 552a) and the Freedom of Information Act (FOIA) (5 U.S.C. §§ 552 et seq.). The Privacy Act makes it illegal for federal agencies to disclose information (including medical information) from records maintained on individuals except under certain conditions. FOIA was intended to make much of the information maintained by federal agencies available to the public, but the act specifically exempts “personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy.”5

Additional federal privacy protections flow from the Health Insurance Portability and Accountability Act (HIPAA) of 1996 and the Genetic Information Nondiscrimination Act (GINA) of 2008. Pursuant to HIPAA, the Secretary of Health and Human Services (HHS) put forth a set of regulations designed to protect the privacy of health information. The regulations cover all types of health information in any form—electronic, on paper, or oral—and they define health information very broadly so that it encompasses genetic information as well as such details as disease status or family medical history; however, they apply only to “covered entities,” that is, institutions and providers that are subject to HIPAA. According to HHS regulations, the entities covered by HIPAA are health care providers that carry out certain types of electronic transactions, health plans, and health care clearinghouses (U.S. Department of Health and Human Services, 2005a). Many biosocial surveys will not be conducted in HIPAA-covered entities, but following the guidelines is a good step toward protecting privacy, so the spirit of that law can perhaps serve as a useful guiding principle. In cases where HIPAA does apply, researchers “are required to have in place reasonable safeguards to protect the privacy of patient information and limit the information used or disclosed to the minimum amount necessary to accomplish the intended purpose of the disclosure” (Jones and Sarata, 2008, p. 19; but see also Institute of Medicine, 2009). Serious violations may be punished by up to 10 years in prison and fines of up to $250,000. The consequences of breaches of confidentiality can be so dire that the penalties should be proportionately large; some would argue that $250,000 for an institution might not be enough. In addition to HIPAA, GINA protects people from being discriminated against by health insurers or employers as a consequence of differences in their DNA that could potentially affect their health, say, by increasing their chances of getting a particular disease. The law enables people to participate in research studies or obtain medical tests without fear of having their DNA information subsequently used against them by health insurers or in the workplace.

An alternative to promising confidentiality is obtaining consent from participants in advance with the understanding of full disclosure (Church, 2005; Lunshof et al., 2008) and acknowledging that biological samples are convertible into identifiable genomic and trait data. Large security gaps are often social in nature; for example, even though high-security defense access requires psychosocial security checks (well beyond any employed by health research organizations), classified materials still slip outside of secure environments (e.g., via theft, mislabeling, cross-contamination, willful sharing of data despite consequences, or unanticipated reidentification algorithms). To some, it may appear disingenuous to continue to promise confidentiality of samples in light of numerous recent examples to the contrary. To help ensure informed consent (rather than merely obtaining legal signatures on consent forms), some groups require 100 percent scores on tests of comprehension of the potential risks and benefits for the subjects, families, and society (Church, 2005). This check has the additional benefit of educating participants before the study, rather than after they need to be informed of some alarming result.

The NCI recommendations regarding privacy and confidentiality (National Cancer Institute, 2007) include a detailed section designed to help guide bio-repositories in their role as honest brokers of sensitive information. Although the recommendations refer to biorepositories, they are applicable to multiple stages of research, analysis, storage, and distribution processes:

Biospecimen resources should establish clear policies for protecting the privacy of identifiable information. These policies may include data encryption, coding, establishing limited access or varying levels of access to data by bio-specimen resource employees, use of nondisclosure agreements, and use of an honest broker system. (National Cancer Institute, 2007, p. 22)

Researchers (and biorepositories) should consider obtaining a certificate of confidentiality (National Cancer Institute, 2007). NIH is authorized to issue certificates to researchers involved in clinical, biomedical, or other research that would allow them to refuse to disclose identifiable information in federal, state, local, civil, criminal, administrative, or other proceedings (see also National Institutes of Health, 2009). If such a certificate is obtained, the participants should be informed about it during the informed consent process. The informed consent process should also include explaining what such certification means and the limits to the protection it can afford (National Cancer Institute,‐2007).

Technical and Logistical Issues

Most biospecimens are classified as biohazardous materials. Personnel who handle them at any stage of a project should be trained in their safe use and handling and take adequate precautions. Protection of those who collect and handle the specimens, including laboratory personnel, should be ensured. Potential threats include not only exposure to infectious agents, but also, for example, exposure to toxic chemicals used in processing specimens, cuts from broken glass or shards, exposure to dry ice during the shipping process, and irritation from adhesives. Protocols for collection need to be established that will ensure the safety of both research participants and the study staff, and protocols for processing, shipping, storage, and use must take account of the safety of each person who may be exposed. Research institutions may have specific requirements to ensure the safety of biological materials.

Documentation is needed at all stages of a study. This requirement may be especially complex for social scientists who, in general, will be archiving not only specimens but also the more familiar kinds of self-reported or interviewer-assessed data that are typically included in social science surveys. As noted earlier, biosocial survey data and specimens usually will be archived in more than one location or type of facility. The data requirements associated with the biospecimens themselves include information about the administrative and operational “trail” of the specimens—collection site, date and time of collection, freeze/thaw occurrences, assay procedures, and laboratory quality control (QC) and quality assurance (QA), to name but a few that will be new to most social survey investigators. Researchers must also establish protocols for linking the different archives in a manner that protects individual confidentiality while having a minimal effect on research. Excessively restrictive rules for linking the data and specimens may hinder the originally envisioned research, the kind of research to which the participant consented in the first place.

The specimens themselves will generally be stored in a biorepository. Detailed information on all specimen collection, processing, and storage procedures should be recorded and tracked by the biorepository’s information and database systems. The NCI guidance discussed earlier includes detail on the necessary systems: “An informatics system should support all aspects of bio-specimen resource operations, including (but not limited to) research participant enrollment and consent; biospecimen collection, processing, storage, and dissemination; QA/QC; collection of research participant data; data security; validation documentation; and management reporting functions. In addition, the system should manage clinical annotations to the biospecimens” (National Cancer Institute, 2007, pp. 11–12). The NCI guidelines cover such areas as identification of biospecimens, integration with local systems, interoperability, and ethical and legal issues pertaining to informatics systems.

Pretesting is important at all stages of a study. Those new to working with biospecimens will find it necessary to test processes and equipment that social scientists usually do not have to consider. The adhesive on specimen labels, for example, should be checked to ensure that the labels will not come off in the freezer or if they get wet; in a similar vein, the ink must be checked to make sure it will not smudge or run. It is wise to use ice water or “scrap” specimens to test shipment protocols for frozen specimens and ensure that the packaging includes adequate dry ice. It is wise as well to anticipate possible delays at the shipping docks, at customs inspection stations, in traffic, and at airports; establishing fall-back procedures for misadventures is advisable. In short, every step of the protocol should be pretested with the equipment and processes to be used at that step. Sufficient training of survey staff (data and specimen collectors and handlers) is also critical to eliminate the human error that may occur in the collection and processing of biospecimens.


As discussed above, the design of the protocol for collecting biospecimens must ensure the safety of both participants and staff. Interviewers or other survey staff who collect biological specimens from human subjects should take precautions to avoid infection from any pathogens that may be present in the specimens, especially when blood samples are being collected (Twitchell, 2003). Again, these issues should be addressed during the training of survey staff. Well-trained interviewers also can facilitate the process of obtaining participants’ informed consent, decrease data collection errors by ensuring that the correct protocols are followed, and minimize the risk to study participants. In addition, there has been a movement recently to develop measures and methods that can be administered by nonclinicians, in some cases yielding immediate information that can be imparted to participants by an interviewer who is not competent to offer any clinical or diagnostic interpretation of the information. Thus it is important to emphasize in interviewer training and supervision that, while interviewers may tell respondents what they have just measured (e.g., blood pressure), they should not offer any interpretation of such information.

Consideration must also be given to the different kinds of preparation that will need to occur in the field. For example, urine may need to be aliquoted or acidified, or blood may need to be centrifuged. Some assays must be performed on freshly collected specimens, and in these cases, it is important to plan for rapid transit or on-site laboratory work.

Once a laboratory has been selected to analyze the biospecimens for the biomarker(s) of interest, the laboratory itself should conduct tests to evaluate assay reliability, and the investigator may want to test for both intra- and inter-laboratory reliability (see Box 2-1 for a discussion on selecting a laboratory). Thus, the protocol may need to include the preparation of duplicate aliquots (for intralaboratory tests) or even triplicate aliquots (to test interlaboratory agreement of assay results). Ideally, the specimens used for these tests should be collected from volunteers outside the target sample so as not to overburden the participants. The specimens to be tested should be indistinguishable in form from those of the participants so the laboratory cannot tell which specimens are intended for reliability testing. Duplicates should be sent to the laboratory at regular intervals to check for consistency of results. If possible, a third specimen should be sent to a different laboratory to validate results. For genomic assays, however, such replication may not be possible, and assay replication, in particular interlaboratory assay replication, may not be possible for assays that require unusually fresh specimens.

Box Icon

BOX 2-1

Selecting a Laboratory for Analysis of Biospecimens in Population Studies. Several issues must be addressed in selecting a laboratory to analyze bio-specimens for biomarker(s) of interest in population studies. Once the decision has been made on which (more...)

The kits a laboratory uses to perform assays may change over time because improvements occur in assay technology or because manufacturers discontinue making them. Investigators may want to purchase in advance all the assay kits needed for all the specimens to ensure consistency throughout the project. More generally, care should be taken to ensure that protocols remain constant in the field as well as in the laboratory and do not drift over the duration of the work. Activities that need to be monitored and checked for reliability or drift include the calibration of measurement instruments, the timing of specimen collections, specimen treatment, shipping methods, transit duration, and field processing (e.g., aliquoting and reagents), as well as practices within the laboratories that perform the assays. In addition to monitoring for consistency over time, the checks should ensure that activities of all staff remain in conformity with the approved informed consent and safety procedures.


Important issues with respect to use and reuse of biospecimens include the question of who uses them and when, costs associated with sharing the specimens, informed consent for reuse, and shipping protocols.

Who Uses Biospecimens and When?

As noted in Chapter 1, biospecimens in a repository are a nonrenewable resource: the amount of blood, tissue, or other material that has been collected from each participant is limited, whereas the data derived from those materials are limitless and can be used as often as desired. Thus an important question becomes: Who will use the specimens and for what purpose? Those who have ownership of the biospecimens will need to develop procedures for choosing among competing demands.

A related question is when analysis of biospecimens and biodata should be undertaken. Costs of many analyses are steadily falling, so waiting to perform analyses can save money or make it possible to perform more analyses for the same cost. But there are also costs to waiting—time wasted by investigators, for example, and delays in uncovering what might be valuable information.

Costs Associated with Sharing Specimens

In addition to the costs of collecting and storing data, there are costs associated with sharing specimens once they are in a biorepository. The process of sharing specimens can be complex and time-consuming and involves a variety of steps, such as internal review and reutilization procedures, with nontrivial costs. These costs must be accounted for and should be reflected in the budget. As discussed later in this chapter, the panel agreed that the data sharing plan submitted with a research proposal should address these issues.

Informed Consent for Reuse

Reuse of specimens must be approved on the original consent form or through a separate IRB approval process. The latter process may require obtaining a new consent from the participants; under certain circumstances, however, some IRBs will consider waiving the reconsent process. Each IRB has its own criteria for deciding on future use, but those criteria generally include the possibility that it may not be feasible to obtain updated consent for every new use, so that any new use should lie within the general scope of the original research. Blanket permission for any and all future (unspecified) uses is generally considered unacceptable by today’s increasingly conservative IRBs. Chapter 4 presents a more detailed discussion of this issue, including some proposed approaches for addressing it, while Chapter 3 reviews salient federal regulations.


All biospecimens should be retrieved and shipped in a way that safeguards their integrity. Concerning retrieval, the NCI best-practice guidelines simply stipulate: “Samples are retrieved from storage according to biospecimen resource SOPs [standard operating procedures] that safeguard sample quality” (National Cancer Institute, 2007, p. 6). For shipping of specimens, however, the guidelines go into much greater detail. They specify, for example, exactly how samples are to be shipped to maintain them at various temperatures, ranging from 8°C to −150°C. They describe the packaging to be used for paraffin blocks and slides, and they suggest that for particularly valuable samples, test packages, such as frozen water samples, be sent out first. They also cover how to maintain the proper paperwork when shipping biospecimens (National Cancer Institute, 2007). It is worth repeating the earlier caution about testing shipment protocols with ice water or “scrap” specimens to check each potential vulnerable point in the protocol.


Most social scientists will lack the expertise and resources to establish or maintain their own facilities for the storage of biospecimens. The panel’s best advice is to use experts in this area but to exercise careful oversight. A reasonable starting point might be a cancer or Alzheimer’s disease center if the investigator is at a university that has one. Alternatively, many commercial facilities are available. The information in the best-practice documents described earlier can serve as an important guide for assessing the adequacy of potential sites.

To ensure the maximum value from biospecimens, it is necessary to store them so they do not degrade, to keep careful records on them, and to ship them in a way that preserves their quality and their identity. Given the complexity of the storage and distribution of biospecimens, investigators should have the option of delegating storage and distribution to a specialized institution—a bio-repository—that is available to accept the biospecimens collected from social science surveys (see Chapter 5). Biospecimens need to be kept stabilized at all stages of the process: collection, storage, shipment, aliquoting, in the field, and in the repository. A biorepository should be chosen that can ensure storage of all biospecimens in an appropriate stabilized state to preserve their integrity and allow the maximum number of analyses to be performed. For example, the Australasian Biospecimen Network describes a number of specific recommended practices in its Biorepository Protocols:

Audits should be conducted to check that biospecimen storage locations concur with database records; storage vessels (tubes, cassettes, etc.) should be checked to ensure they have remained intact; storage conditions should be monitored by a central alarm system and/or local alarms . . . ; back-up systems and enough empty freezer space should be allowed in case quick transfer of specimens from malfunctioning freezers is required . . . ; having multiple storage sites (on or off-site) ensures that all specimens will not be destroyed in the case of freezer malfunction or other emergency situations. (Australasian Biospecimen Network, 2007, p. 60)

Automated security systems should monitor all storage equipment, and backup systems should be in place in case of an emergency such as a power failure. For example, the ISBER best-practice guidelines call for using an uninterruptible power supply (UPS) for critical equipment: “Computer systems and electronic systems, such as environmental monitoring systems, safety systems (e.g., oxygen sensors, ventilations systems, etc.) or controllers for liquid nitrogen freezers, should be protected by a UPS. UPSs used in repositories should be tested on an annual basis to ensure their proper backup capabilities” (International Society for Biological and Environmental Repositories, 2008, p. 14).

For quality assurance and quality control purposes, training of personnel and harmonization of protocols throughout different repositories and the laboratories of individual researchers are essential to ensure reproducibility. There are a variety of international standards for laboratories and related facilities, such as ISO9001:2000, developed by the International Organization for Standardization (ISO) for quality management systems (International Society for Biological and Environmental Repositories, 2005, 2008).

Perhaps the most important issue in storing biospecimens is whether to keep them in a central repository or use an alternative approach, such as maintaining them in separate repositories associated with the various institutions involved in their collection or keeping different types of specimens in different repositories. A major argument for using a central repository is that it should be more cost-effective than distributing specimens and data among several repositories, but this argument is based on general principles, not experience, as little is known about the cost of one option versus the other. Other issues to consider include which specimens (and which biodata) to keep and for how long. Storage is not free—there is always an incremental cost for additional specimens and data—so at some point one must decide which specimens or data it makes sense (economically) to maintain and which can be discarded without significant loss. In the case of redundant specimens or data, erroneous data that cannot be corrected, or specimens that have insufficient or no identifiers, it may make sense at some point to destroy them. On the other hand, the costs of data storage are declining, whereas the same cannot be said of the costs of biospecimen storage. Therefore, the issue of what to keep and what to dispose of concerns mainly specimens rather than data. No clear guidelines exist on this issue, but it is one that researchers should be aware of and be prepared to address, preferably at the start of a study.

Fortunately, new technologies are changing approaches to the storage of biological specimens and the associated costs. For example, while fresh-frozen tissue used to be essential for gene expression profiling (because of RNA stability issues), new methods such as Illumina’s DASL6 assay have made it possible to use formalin-fixed, paraffin-embedded (FFPE) tissue. This advance has reduced the need for expensive low-temperature storage and opened up the possibility of using large collections of FFPE samples that have been routinely archived in hospital pathology departments.

Finally, the informed consent process should include anticipating and requesting permission for the storage of specimens, as well as their future use (see above), and the specimen archive should include information that both links use of the specimens back to the informed consent documents and links the specimens to the data derived from them.7 Specimens need to be stored with information that will allow determination of consent approvals. Did the participant agree to use of the specimen only at the time of collection? Was the agreement only for nongenetic tests? Has the participant withdrawn consent (and what does that mean with respect to the disposition of his/her specimens)? (See also Chapter 4.)


Storage of biospecimens raises issues concerning their custodianship and ownership. A number of groups and organizations have examined these issues, and a variety of publications suggest policies and best practices for addressing them.

Responsible custodianship of biospecimens implies certain basic steps, including the development of transparent policies to ensure proper use and storage. It is important, for example, that researchers develop guidelines early on regarding who should receive biospecimens and biodata for use in other studies. These guidelines are particularly important for biospecimens because, as noted earlier, they, unlike data, can be exhausted. Thus researchers must ask themselves such questions as: Should remaining biospecimens be distributed on a first-come, first-served basis? Should there be a formal application process? In studies within the United States, should foreign investigators have access? And in general, what factors will be taken into account in deciding who will be allowed to share the data? These are complex questions. If the proposed analyses lie beyond the competence of the original investigators, for example, projects will need to ensure that they have adequate support for the purpose.

The panel agreed that the data sharing plan of a research proposal should specify policies and implementation plans for archiving and sharing both specimens and the data derived therefrom. In general, there is no one best plan for the use and reuse of specimens, but the plan should include a discussion of the adequacy of the storage and retrieval protocols. It should spell out criteria for allowing other researchers to use (and therefore deplete) the available stock of specimens, as well as to gain access to any derived data. The plan should also specify the procedures for accessing the specimens and data. It should include provision for the storage and retrieval of specimens and clarify how the succession of responsibility for and control of the specimens will be managed at the conclusion of the project. Finally, the plan should contain information on how specimens and data derived from them are to be documented and provide for public access to that documentation.

The NCI best-practices document describes a number of custodianship practices that should be followed to ensure that biospecimens are maintained and used as effectively as possible (National Cancer Institute, 2007):

  • The biorepository should ensure the proper storage of the specimens to maintain their physical integrity and the integrity of study participants’ data linked to the specimens. It is the responsibility of the principal investigator to choose a biorepository that can meet these requirements.
  • Clear and transparent protocols for the distribution of the specimens and the data to other investigators should be in place.
  • Plans for handling, storing, and disposing of the biospecimens and associated data should be in place for such contingencies as the end of a grant, the end of a particular study, biospecimen depletion, and a study participant’s request for discontinuation of participation.

Policies on Custodianship

Although the NCI best-practices document provides extensive guidance on many practices for biospecimen collection, storage, use, and maintenance, it does not specify the custodial roles and rights of biorepositories and their responsibilities to their host institutions or study participants, nor does it provide a functional definition of custodianship. To clarify these responsibilities and to help develop a transparent policy for biorepository governance, the NCI Office of Biorepositories and Biospecimen Research (OBBR) held a workshop in October 2007 (Custodianship and Ownership Issues in Biospecimen Research Symposium-Workshop). The workshop summary is an informative resource that makes a useful distinction between the concepts of “custodian-ship” and “ownership” (Office of Biorepositories and Biospecimen Research, 2008). It also provides recommendations for dealing with issues of financial and other types of conflict of interest (COI), intellectual property (IP), and access to products and benefits. Another complexity is the relationship between the principal investigator and the biorepository. Most social scientists will use a biorepository to store and distribute specimens, but they will not wish to delegate scientific decisions to the biorepository. Thus, it is important to specify at the outset the roles and responsibilities of the principal investigator and the repository. The OBBR workshop summary offers the following guidance on these issues:

  • One OBBR recommendation is that “the custodian of biospecimens should be someone other than the investigator . . . investigators with small biospecimen collections should be encouraged to establish or join an existing IRB-approved regulated biospecimen resource” (p. 8).
  • Decisions regarding the distribution and reuse of biospecimens have the potential to create a conflict of interest. An expert panel might be appointed to provide technical/scientific advice and to make recommendations to the investigators regarding the disposition and use of the specimens in accordance with all ethical and IRB guidelines.
  • Conflicts of interest should be clearly identified. A starting point would be the Code of Federal Regulations (CFR) Part 50, Subpart F,8 along with other NIH and university guidelines. All existing and possible institutional, NIH, and other COIs regarding biospecimens should be reviewed to determine whether they have been sufficiently addressed. All individuals responsible for biospecimen distribution should report COIs, and financial COIs should be disclosed to the public whenever possible.
  • Biorepository personnel and staff, as custodians of biospecimens, are not considered inventors under patent law for any inventions resulting from the research on specimens housed in the repository. Biorepositories have no inherent rights to future IP associated with inventions developed by researchers using the biorepository’s collection (see also National Institutes of Health, 2007).
  • Educational materials on IP issues related to biospecimen research should be developed and made available to study participants. Possible financial and nonfinancial benefits and commercial products resulting from such research should be discussed in the informed consent document, including possible supplemental material.
  • The existence of biospecimens should be made public when research data resulting from the use of those specimens are published, even if the biospecimens themselves are not available to the research community.

Court Decisions Regarding Ownership

There are few legal precedents related to the ownership of biospecimens, and they tend to be fact- and jurisdiction-specific (Office of Biorepositories and Biospecimen Research, 2008). In general, the courts have denied claims to ownership by research participants. In one recent prominent case, Washington University v. Catalona, the Federal District Court for the Eastern District of Missouri ruled that the research participants (human subjects) retained no rights to control their biospecimens after donation.9 The court stated that in that particular case, the donation of biospecimens was an inter vivos gift. The informed consent document and the accompanying brochure were the key documents in this case, and the court used them as evidence that research subjects retain no rights to repossess or transfer their biospecimens after donation or to direct any future use of the specimens for research.

On appeal, the U.S. Court of Appeals for the Eighth District stated that research participants do retain the right to discontinue participation in research by doing one of the following: not answering additional questions, ceasing donating additional tissue, or disallowing the use of their tissue in future research. The ability not to allow the use of donated specimens in future research is a key right of participants and is closely related to each participant’s right to withdraw from a study at any time.

The decisions of the district court and the appeals court reflect a public policy that medical research can advance only with continuing access to bio-specimens. Nevertheless, it is possible that in another set of circumstances, the informed consent document could be interpreted as giving study participants the right to withdraw and physically repossess their biospecimens.10


Experience has led many IRBs to insist that biospecimens be destroyed when a clearly stated, focused use of the specimens has been completed, thus avoiding even the possibility of new and unapproved uses of the specimens. However, such stipulations impede the use of specimens to reproduce, challenge, or extend scientific findings. The panel finds that the latter goal outweighs the former. Consequently, the panel recommends the creation of at least one central facility for the storage and distribution of biospecimens, provided adequate procedural safeguards are in place (see Chapter 5 for a full discussion of the panel’s recommendations).

Expanding access to biospecimens promotes further research and makes it possible to explore new questions without having to collect new specimens. It creates economies of scale, allowing different researchers to pursue a variety of investigations with specimens originally collected for a single purpose. Sharing biospecimens “fosters an open research community and reinforces transparent scientific inquiry” (National Research Council, 2005, p. 39). Thus ensuring that researchers have broad access to biospecimens is essential to maintaining and improving the quality of the research enterprise. Access to biospecimens by a wide variety of researchers also enables a feedback system that can reveal problems with the specimens or their collection and suggest methods for improvement (National Research Council, 2005). Eventually, moreover, people die, at which point they are no longer “human subjects,” and the associated restrictions on the use of their biospecimens no longer apply (see the discussion of this issue in Chapter 3).

For all of the above reasons, it is important to provide the widest possible access to biospecimens. Decisions about who obtains access can be based on various considerations, including “the scientific merit and potential impact of the proposed research, whether the research use is appropriate to the nature and purpose of the repository, adequacy of the research design and funding, public health benefits and risks of the proposed research, legal and ethical considerations, and the qualifications of the research team and research environment” (International Society for Biological and Environmental Repositories, 2008, p. 49). But these considerations should not be used as a way of artificially limiting access.

To best serve the various research communities that can benefit from their biospecimens, principal investigators should develop clear and ethical access policies to facilitate sharing in ways that minimize disclosure risks and comply with all federal and state regulations. Policies should be flexible enough to respond to new technologies when they appear and general enough that they can be adapted to different kinds of biorepositories. The data sharing plan should include detailed information regarding these policies and their proposed implementation. To avoid any appearance of self-interest, a project might empower an external advisory board to make decisions about access to its data.

When one institution transfers materials to another for research purposes—for example, a biorepository making biospecimens available to an investigator at an institution not affiliated with the repository—a material transfer agreement (MTA) sets forth the conditions under which the transfer is to be made and delineates the rights of the two parties. If a discovery made using the material leads to a commercial application, for example, the MTA describes how any receipts or profits from the application will be allocated.

In the case of MTAs governing biospecimens from repositories and the data derived therefrom, investigators need to ensure that the agreement is specific about what can and cannot be analyzed. If a clear understanding is not reached, and an investigator analyzes and publishes data without the full informed consent of those who have the rights to those data, serious problems can arise. One example is the experience of researchers at Arizona State University who analyzed genetic data from the Havasupai tribe in an effort to understand its high prevalence of diabetes. Through a series of misunderstandings, the tribe came to mistrust the researchers performing the analysis, and demanded that all analysis stop and its biological samples be returned. Because the agreement had not been specific enough, the Arizona State researchers ended up with no data and no analysis after several years of work (Dalton, 2004).



Two organizations in the United Kingdom (UK) have also produced valuable best-practice guides. The Wellcome Trust, the UK-based charity that funds innovative biomedical research, published Sharing Data from Large-Scale Biological Research Projects: A System of Tripartite Responsibility (Wellcome Trust, 2003), while the UK Biobank, a long-term biobank study investigating the genetic and environmental components of disease, produced UK Biobank Ethics and Governance Framework (UK Biobank, 2007).


Blood spots are not addressed.


For discussion of some of these issues, see National Research Council (2008).


Norman-Bloodsaw v. Lawrence Berkeley Laboratory, 135 F.3d 1260, 1269 (N.D. Cal. 1998).


5 U.S.C. § 552(b)(6).


DASL = cDNA-mediated Annealing, Selection, Extension, and Ligation Assay.


To protect research participants, IDs for the biorepository and the archive of the derived data should be different, and access to the link between them should be limited.


Code of Federal Regulations Index, see http://www.gpoaccess.gov/CFR/INDEX.HTML (accessed January 26, 2010).


Washington University v. Catalona, Case No. 4:03CV01065-SNL, Document 152 (March 31, 2006).


See also the discussion in Chapter 4 on withdrawing consent.

Copyright © 2010, National Academy of Sciences.
Bookshelf ID: NBK50729


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (549K)
  • Disable Glossary Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...