October 26, 2011

U.S. Department of Health and Human Services
Office for Human Research Protections
1101 Wootton Parkway
Suite 200
Rockville, MD 20852

Attn: Human Subjects Research Protections: Enhancing Protections for Research Subjects and Reducing Burden, Delay, and Ambiguity for Investigators, HHS–OPHS–2011–0005

Dear Dr. Menikoff:

Genetic Alliance welcomes the opportunity to comment on the Department of Health and Human Services’ advance notice of proposed rulemaking (ANPRM) regarding the “Common Rule” (Subpart A of 45 CFR part 46). Genetic Alliance believes that individuals should be active and full participants in biomedical research. We eschew the word “subject” and know that individuals are more willing to participate, are less vulnerable to harm, and contribute more meaningfully if they are seen as partners in the research enterprise. Most significantly, from our perspective, consumers should be informed participants in the process. One key element of this active role is individuals’ stewardship of their own data. In their ability to share what they wish, where they wish, with whom they wish, lays a solution to many human “subjects” issues. A person’s comfort level with sharing clinical and other information varies throughout his or her life course, and differs from the comfort levels of others.  Individuals should make the choices about sharing their own data in the context of their lives, to meet their needs and the needs of the communities in which they live. Inherent in this relational, rather than transactional, enterprise, is an informed consent process, rather than a form. When individuals are empowered with knowledge and control of their own data, they are more likely to understand risk, consent, data security, and privacy.  This increased knowledge is a boon to the biomedical enterprise and to those who need the solutions the biomedical field seeks. We support changes that will move the system toward fostering informed, empowered consumers.

Distinction Between Types of Risk:
Genetic Alliance supports the proposal in the ANPRM to eliminate the previous distinction between “expedited” and “exempt” review categories, creating a single “excused” category not subject to IRB review. The types of research activities qualifying for this new excused category should include all those formerly under the “expedited” label, and this list should be updated regularly (7). We do not believe this change will discourage individuals from participating in research, as there will still be some form of intra-organizational oversight. Furthermore, we do not think increasing the types of studies qualifying for this excused category will result in any substantive reduction of protection for patients. Excused studies would include surveys, focus groups, certain types of social and behavioral research, and other projects with similar methodology practiced on competent adults (14).

Genetic Alliance also agrees that there may be a better term to label studies falling into this category. The term “excused” may convey that studies in the category have no form of oversight, which is not the case. Instead, a term such as “registered” would be better, as it suggests there will be review of some kind (20). In many regulatory schema used throughout the medical world, a first tier registration is often the first level of oversight. This would be a similar concept.

In general, Genetic Alliance is in favor of streamlining the protocol for submitting paperwork for review. Research studies believed by investigators to involve only minimal risk should be able to submit a short, one-page form detailing the essential elements of the study to a board for local oversight, and begin work immediately (10). Additionally, all studies falling under the new excused status should be exempt from continuing review, as should studies posing greater than minimal risk whose post-experiment activities include only those qualifying for excused status (3). A reviewer should always be able to request continuing review for any study, as the ANPRM states, but we agree with the default regulation not requiring it.

With the purpose of establishing a more efficient system for reviewing research, Genetic Alliance also supports the mandate of one IRB of record for domestic, multi-site research studies. Such a mandate would expedite these studies, which are increasingly common. It would also greatly increase the effectiveness of such studies, and in doing so encourage this type of important collaboration, without detriment to the protection of research subjects (30, 33). We believe a similar centralization should also be implemented in the reporting of adverse events and unanticipated problems during research studies. It would be beneficial to have reports of such events and problems collected in a single database (or a federated indexed solution), accessible to all relevant Federal agencies (69).

Data Security:
At Genetic Alliance, we believe significant gains can be made in protecting research participants through better data security. Because of this, we are in favor of better confidentiality and data security protections, as well as regular updates to what is considered identifiable data by experts in the field of data security (55). We are dubious whether the application of the HIPAA Privacy Rule as a model would be best, as it has been shown to unnecessarily impede research and decrease participation, but a similar set of standards is necessary to guard against the primary danger in minimal risk studies: informational risks (54, 59).

Consent Forms:
Genetic Alliance firmly believes that consent should be a process, which culminates in explicit consent. As such, we support a number of the changes to consent forms outlined in the ANPRM, including (37):

• More concise and understandable writing. These forms should undergo a literacy level review, just as any good public outreach documents do. The increasing litigious climate creates a scenario where institutions rely on boilerplate language to cover all possible risk to the institution, but this is not the purpose of a consent process.  Excessive use of boilerplate language has made consent forms too verbose and jargoned (35).
• Inclusion of specific risks and details of study in clear language, along with a statement explaining there will be no detrimental effects of non-participation (36).
• Standardization of the basic elements of consent forms, including reconsent and recontact.
• Oral consent sufficient for competent adults in situations where regulations would permit such activities. Investigators should be able to abbreviate the elements of informed consent in 45 CFR 46.116 to some predetermined degree, but there should be some standard for what constitutes a competent adult (38, 42). Special attention also should be paid to participants to makes sure they are not in a position to be coerced.

In general, Genetic Alliance is in favor of consent systems that allow participants to have meaningful ownership of their data and open communication with the investigators using that data. Because of this, we tend to prefer a tiered method of consent (a method with reconsent options for future usage of samples/data) to blanket methods (which might allow an individual to consent to all future research on their data and/or sample). Although there must be some equilibrium reached between control of the data by the participant and the investigator’s time and efforts in obtaining consent for various usages, we believe the more ownership a participant can take the more willing they will be to contribute, both in the present and future (50). Therefore, we would like to see all consent forms include a provision that grants the data back to the individual. At the very least, some basic language could allow that all individuals are stewards of their own data, and as such may chose to share it with other investigators, studies, institutions and so on. In its best form, this would entail a granular, dynamic consenting system that has been operationalized by a number of systems in medicine and other industries. Having this standardized process and language would reduce redundancies in studies and accelerate large population associations in the form of genotype-phenotype correlations, adverse events, and stratification of populations based on response.

Biospecimens:
The identifiability of biospecimens raises difficult issues. With increasing ability to genotype with deeper coverage comes an ultimate identifier. However, for the purposes of biomedical research, there has to be another data set to create the opportunity to identify a person in a social context. It is not authentic to treat associated data from biological samples that have been stripped of identifiers as non-human and nonidentifiable. It is also untenable to have every sample treated as though it has typical identifiers associated with it. Genetic Alliance would like recognition that this question cannot be resolved with a simple declaration that these samples are not human. In most cases, more facile, electronic, and contextual consenting systems would alleviate some of the issues associated with identifiable samples. Research participants—engaged in an informed consent process and given stewardship of their own data and samples—will be better able to avoid harm. In most instances, when this issue has been studied, individuals said they wanted to be asked to have information or samples used, and they would usually give consent. In all cases, including studies generating particularly large datasets, a community of trust can play a key role in mitigating harm that might come from the perception of identification of samples. Special consideration should be given to communities who have experienced harm in the past (Native Americans, African Americans, First Nations, and vulnerable communities). These communities are particularly hesitant to accept the notion that as long as the specimen remains de-identified within organizations with Federal oversight the potential harm resulting from re-identification is minimal (50). They will require more active involvement with the research community – participation.

We hope that efforts will be made to recontact participants when possible and to use new data security and consenting systems. We agree that reducing unacceptable burden on investigators for minimal protection to participants (52, 53, 58) is not optimal. We would like to see a dynamic balance between offering participants meaningful engagement and reducing burdens on investigators to accelerate biomedical research. The questions are difficult ones, and the solutions will not be simple. In the long term, the right process will create a robust research enterprise that alleviates burdens for all, especially those suffering from disease today.

Best Regards

Sharon F. Terry
President & CEO