Recommendation 10. IRBs should operate on the presumption that research on coded samples is of minimal risk to the human subject if
- the study adequately protects the confidentiality of personally identifiable information obtained in the course of research,
- the study does not involve the inappropriate release of information to third parties, and
- the study design incorporates an appropriate plan for whether and how to reveal findings to the sources or their physicians should the findings merit such disclosure.
Recommendation 11. In determining whether a waiver of consent would adversely affect subjects’ rights and welfare, IRBs should be certain to consider
- whether the waiver would violate any state or federal statute or customary practice regarding entitlement to privacy or confidentiality,
- whether the study will examine traits commonly considered to have political, cultural, or economic significance to the study subjects, and
- whether the study’s results might adversely affect the welfare of the subject’s community.
Recommendation 12. If research using coded or identified human biological materials is determined to present minimal risk, IRBs may presume that it would be impracticable to meet the consent requirement (45 CFR 46.116(d)(3)). This interpretation of the regulations applies only to the use of human biological materials collected before the adoption of the recommendations contained in this report (specifically Recommendations 6 through 9 regarding informed consent). Materials collected after that point must be obtained according to the recommended informed consent process and, therefore, IRBs should apply their usual standards for the practicability requirement.
Recommendation 13. OPRR should make clear to investigators and IRBs that the fourth criterion for waiver, that “whenever appropriate, the subjects will be provided with additional pertinent information after participation” (45 CFR 46.116(d)(4)), usually does not apply to research using human biological materials.
Collectively, recommendations 10, 11, 12 and 13 address the four criteria set forth in 45 CFR 46.116(d) under which a waiver of consent may be granted. The Working Group agrees in principle with these four recommendations and finds that they clarify and simplify the process by which IRBs can determine whether or not to grant a waiver of consent for research involving human biological materials. In addition, the Working Group agrees that these four recommendations provide a logical, sensible, and appropriately flexible approach to interpreting and applying the existing criteria for waiver of consent to research. Since all four criteria must be satisfied before a waiver of consent is granted, the Working Group elected to consider these four recommendations as a group.
The Working Group emphasizes that the criteria for waiver of informed consent for research undertaken under FDA regulations differ from those described in 45 CFR 46.116(d). FDA’s regulations 21 CFR 50.23 and 50.24 permit a waiver of informed consent only in very limited circumstances.30
There are two issues that appear throughout the discussion of these four recommendations and require explanation here. First, disclosure of research results to subjects may involve either information pertinent to an individual or general information that does not relate specifically to an identifiable individual. For clarity, the Working Group refers to research results that pertain to an individual as “individual research results.” Second, the Working Group distinguishes between disclosure of research results that is initiated by the researcher, who offers unsolicited information to a subject, and the disclosure of results in response to a direct request from the study subject.
The first criterion to be met for granting a waiver of consent is that subjects should not be exposed to more than minimal risk. Recommendation 10 addresses this criterion, focusing specifically on individually identifiable information. The Working Group agrees that this is the most critical issue for determination of minimal risk in studies that involve human biological materials. The Working Group agrees with Recommendation 10 (a) and (b) that a study using appropriate strategies to protect such information may be presumed to be of minimal risk.
The Working Group is persuaded that studies for which consent can be waived will generally be those for which it is not expected that research results will merit disclosure to subjects. The Working Group appreciates the caution and sensitivity signaled by NBAC in 10(c) with respect to the possibility that individual research results may be so compelling that the investigator may feel obligated to initiate disclosure of such findings even when such disclosure had previously been considered unlikely. In certain types of research, the likelihood is small that research findings could be so significant as to merit investigator-initiated disclosure to an individual who is not aware of having been a study subject. For some of these studies, it will be difficult to predict what kind of results might merit such disclosure. Consequently, it will be difficult for the researcher to develop - and for the IRB to assess - an appropriate plan for whether and how to disclose results. For such studies, the researcher could make a commitment that (1) if individual research results appear to merit disclosure, the researcher would then develop and submit for IRB approval a disclosure plan, and (2) no investigator-initiated disclosure would take place until IRB approval was granted. For other studies for which it may be feasible to predict results that might merit disclosure, the investigator should develop an appropriate plan for such a contingency at the time of the initial IRB review. When it is unlikely that results will merit disclosure and when there is an appropriate plan in place to address that unlikely event, the proposal would be eligible for a waiver of consent. In sum, the Working group agrees with NBAC that a high threshold should be applied when determining whether an investigator should offer individual research results to a subject of a study conducted under a waiver of consent.
The Working Group also appreciates NBAC’s suggestion that the researcher may work through the subject’s physician to offer unsolicited research results to the subject. However, involving the study subject’s health care provider is not a substitution for the IRB’s responsibility to determine whether the value of disclosing results outweighs risk of such disclosure. In addition, the Working Group cautions that most health care providers are not trained to critically evaluate all relevant aspects of research findings. Health care providers who are asked to provide their patients with unexpected information will need education and support regarding best approaches for disclosure, including approaches to maximize potential benefit to the study subject and minimize damage to the ongoing relationship between health care provider and patient.
The second criterion for waiver of consent emphasizes that consent may not be waived if such action would adversely affect subjects’ rights and welfare.
Recommendation 11(a) refers to the question of whether a waiver of consent for a proposed study might violate any federal or applicable state statutes or customary practices regarding entitlement to privacy or confidentiality. Current DHHS regulations (45 CFR 46.107(a)), supported by existing guidance,31 require IRBs to be knowledgeable about relevant statutes and practices. In addition, 45 CFR 46.116 (d) (2) does not allow an IRB to grant a waiver of consent if the proposed research would adversely affect the study subjects’ rights.32 Application of recommendation 11(a) therefore requires a factual response: granting a waiver of consent for the proposed study either violates or does not violate state or federal laws regarding entitlement to privacy or confidentiality. If it does, a waiver of consent may not be granted.
The Working Group believes that 11(b) and (c) should be regarded as conditions that will require a higher level of scrutiny of a proposal by the IRB but which, even if they existed in a given research proposal, do not preclude granting of a waiver of consent. As long as the IRB determines that a study examining traits considered to have political, cultural or economic significance to the study subjects does not adversely affect the subjects’ rights and welfare, such a study should be eligible for a waiver of consent. Finally, in 11(c) NBAC focuses on the risk that a study might adversely affect the welfare of a subject’s community. Federal regulations governing waiver of consent address the rights and welfare of the individual (emphasis added) study subject. The IRB would therefore need to consider whether, if there were reason to believe that a study could cause harm to a group, such group harm would also adversely affect the rights and welfare of individual study subjects. The Working Group notes that the subject of group harm is addressed more fully in Recommendations 18 and 19.
Recommendation 12 addresses the question of whether or not it is impracticable to obtain consent for research. While the Working Group believes that on occasion it may be practical to obtain consent for new uses of existing samples, the Working Group also agrees that in most cases it is impracticable. Thus, the Working Group agrees in principle with Recommendation 12 and considers it a reasonable approach to interpreting 45 CFR 46.116(d)(3) in a manner that is consonant with the principle of protecting human subjects while allowing investigators access to existing human biological materials. However, the Working Group is concerned about NBAC’s suggestion on implementation of this recommendation at a specific point in time.
Use of improved strategies for obtaining informed consent before collecting human biological samples (such as those in Recommendations 6-9) does not guarantee that, in the future, subjects’ wishes regarding new uses of existing samples will always be unequivocal. In addition, the Working Group notes that currently some IRBs do not consider the procedural or financial burdens associated with seeking consent to be evidence of impracticability; instead, they require proof that it is impossible to contact subjects before granting a waiver of consent. The Working Group shares NBAC’s concern that some IRBs may set an unrealistically high threshold for determining whether or not it is impracticable to seek consent. The Working Group believes that IRBs need additional guidance regarding definition of impracticability to clarify the conditions under which presumption of impracticability may be made. Such guidance would also address the issue of when and how the definition should be applied.
Recommendation 13 recognizes that the fourth criterion under which a waiver of consent may be granted will usually not be applicable to research involving human biological materials. The Working Group notes that NBAC’s conclusion is consistent with the history and original purpose of the fourth criterion for waiver of consent. 45 CFR 46.116(d)(4) is particularly important for certain types of behavioral research in which the research results would be invalid if subjects knew that they were being studied and in which subjects are informed of their participation and the results following completion of the research.
The Working Group notes that a research project that meets the third criterion for waiver of consent (45 CFR 46.116(d)(3), addressed in Recommendation 12) is one in which it is impractical to contact subjects for consent. Therefore, one could argue that it would be equally impractical to contact subjects to provide information. This does not, however, preclude an investigator from attempting, as provided for in NBAC’s recommendation 10(c), to contact subjects in order to offer information, in the event that individual research results merit disclosure. In addition, notwithstanding the Working Group’s agreement with Recommendation 13, subjects have a right to request and receive information about themselves. This issue is addressed further in the response to Recommendation 14.
Proposed Action for Recommendations 10-13
The Working Group proposes that OHRP, in consultation with relevant DHHS agencies and appropriate stakeholders, prepare and disseminate guidance to support implementation of Recommendations 10-13.