Biobanking and Broad Consent

Biobanks store biological samples and associated data for research that has not been designed yet, which breaks the classical consent model. Traditional informed consent describes a specific study with specific risks; a biobank asks permission for studies nobody can describe. How that gap is bridged, and what a participant can still control afterward, is the substance of this area.

• Broad consent permits storage and future unspecified research within stated limits, rather than consent to one protocol. The 2018 Common Rule revisions formally recognized broad consent as a pathway for secondary research use of identifiable specimens and data.
• What broad consent must still disclose: the types of research that may be conducted, the types of institutions that may receive the material, how long it will be stored, whether and how results might be returned, and whether commercial entities may use it and whether the participant shares in any resulting profit. That last point is a common source of anger when it surfaces later rather than at consent.
• Identifiability is the axis that determines protection. Fully de-identified specimens fall outside much of the human subjects framework, while identifiable or re-identifiable material remains protected. Genomic data complicates this: a genome is inherently identifiable in a way a random chemistry value is not, so "de-identified genomic data" is a weaker claim than it sounds.
• Withdrawal is partial in practice. A participant can generally stop future use of their sample and request its destruction, but data already distributed, analyzed, or published cannot be recalled. Consent should say this plainly rather than implying a clean exit.
• Tiered consent offers participants choices rather than one switch: agreeing to storage but not to commercial use, or to some disease areas and not others. It respects autonomy at the cost of operational complexity.
• Return of results from biobank research is often not offered at all, because the link back to the participant may be deliberately severed. Participants should know this before contributing.
• Ownership and benefit sharing remain contested. The general legal position in the United States is that donors do not retain property rights in donated tissue, a position shaped by Moore v. Regents and by the Henrietta Lacks history, which is why transparency about commercial use is now treated as an ethical baseline rather than an optional courtesy.
• Group harm is specific to genomic biobanking. Findings about a population can stigmatize a community whose individual members each consented, which is why community engagement is standard practice for research involving identifiable populations. The Havasupai case, where samples collected for diabetes research were used for studies on schizophrenia, inbreeding, and population origins, is the reference example.
• Pediatric samples raise the question of re-consent at the age of majority, since a parent consented to indefinite storage of a child's material.

• A family consents to store their child's leftover diagnostic sample in an institutional biobank. The counselor covers what broad consent means, that specific future studies cannot be named, that commercial partners may have access, and that results will generally not be returned. The family agrees to storage but declines commercial use, which the tiered consent form permits.
• A participant asks to withdraw ten years later. The biobank destroys the remaining sample and stops future distribution, but explains that data already shared with three research groups and included in two publications cannot be retrieved. Because this was stated at consent, it is a disappointment rather than a betrayal.
• A researcher proposes using stored samples from a specific community for a study on a stigmatizing trait. The IRB requires community engagement before approval, on the basis that individual consent does not address harm to the group.
• A young adult whose sample was banked in infancy on parental consent is recontacted at 18 and asked to consent in her own right. She declines, and her sample is withdrawn.

• Do not describe withdrawal as complete. Data already distributed cannot be recalled, and saying otherwise creates a promise the biobank cannot keep.
• Do not treat genomic data as reliably de-identified. Re-identification from genomic data is demonstrated, and consent language should reflect that rather than overstating anonymity.
• Do not bury commercial use. Participants who discover it later frequently feel deceived even where the consent form technically disclosed it, which is an argument for saying it clearly rather than including it.
• Do not skip the return-of-results expectation. Participants routinely assume they will hear something. Most biobank research returns nothing.
• Do not ignore group-level risk in individually consented research. Consent from every individual does not address harm to the community they belong to.