When Participants in Genomic Research Grow Up: Contact and Consent at the Age of Majority

The landscape of pediatric research is becoming more complex. Contemporary research studies, and genome studies in particular, frequently involve a range of research activities. For example, the Clinical Sequencing Exploratory Research (CSER) Consortium studies evaluate whether genomic testing can be useful in clinical settings.(1) These studies include the storage of biosamples for future research, observations about the way providers interact with results reported in electronic health records, and the contribution of data to national databases like ClinVar and the Database of Genotypes and Phenotypes (dbGaP).(2, 3)

Given this complexity, pediatric investigators and Institutional Review Boards (IRBs) working on research projects of this type may find it challenging to develop an appropriate plan for addressing consent when pediatric research participants reach adulthood. After all, the research activities that comprise a complex study frequently differ with respect to timing, the risk involved, and the opportunities for interaction between investigators and participants. Any plan to readdress consent when a participant reaches the age of majority must account for all of these considerations.

In order to address this changing landscape, we provide a framework that genomics researchers and IRB members can use to address consent when enrolled children reach adulthood. Based on both ethical and regulatory perspectives, we argue that the importance of replacing parental permission with a young adult's own informed consent depends on the nature of the research activities that will take place following a participant's transition to adulthood. Research activities that either confer substantial risk, or provide opportunities to obtain consent through ongoing contact, should be continued only if explicit informed consent is obtained from young adult participants. However, we argue that it is ethically permissible and consistent with current regulations to continue certain types of research activities under a waiver of participant consent, and that IRBs should utilize the full range of options provided by existing regulations. In making this argument, we work from the understanding that IRBs have significant latitude to interpret regulatory frameworks and policies in light of local priorities. We then apply this framework to a range of more specific questions, including how investigators leading genomic studies should manage the issues that arise when attempts to contact a young adult for consent are unsuccessful.

Case Vignettes

Consider a hypothetical genomic research study in which pediatric research participants are taking part in a series of research activities. After children are enrolled with the permission of their parents, their genome sequence is analyzed to identify genetic targets to inform therapeutic interventions for a specific medical condition. Treatment for the medical condition is modified based on sequencing results, and evaluated through frequent follow-up appointments over a three-year period. Once this interventional stage is completed, the investigators intend to conduct additional analyses using both genomic data and other collected data. If they discover additional genomic results that might be clinically relevant to a participant, they intend to disclose these results to the family. They also intend to contact all participants and parents once a year to remind them they may receive genomic results in the future, to provide updates on the study, and to update participants' contact information. Consider three pediatric participants whose involvement in these research activities differed at the time of their transition to adulthood, and the implications of these differences for the approach to contact and consent.

The first participant is Anna. She enrolled when she was ten years old, and provided her assent to participate in the study. She completed the interventional portion when she was thirteen. When she reached the age of majority, the investigators were still analyzing her identified sample. They had remained in contact with her and her parents every year, and kept her contact information up-to-date. Now that Anna is a legal adult, do investigators need to obtain her consent to continue using her sample for research purposes, and before reviewing additional results that might be returned to her?

Bryon, the second participant, was also ten years old when he assented to enrollment. He completed the interventional portion when he was thirteen years old, and then moved with his parents to another state. When he turned fourteen, research coordinators tried unsuccessfully to update his contact information. By the time he reached the age of majority, it had been nearly five years since the study investigators had been in contact with him. Do the investigators need to attempt to contact Bryon before using his sample for further research? What should they do if they are unable to reach him? Do they need to de-identify or even destroy his data and samples?

Christopher enrolled in and assented to study participation when he was sixteen years of age. He was still an active participant in the interventional portion of the study when he reached the age of majority. Do the investigators need to obtain Christopher's consent in order to keep him enrolled in the interventional portion of the study?

Even though all three adolescents participated in the same study, their involvement in the study when they became legal adults was quite different. These differences can affect both the feasibility of reaching a participant and the approaches to consent that might be appropriate.

Research Activities

Studies often include research activities that differ in their implications for considering consent when participants reach the age of majority. First, they differ with respect to timing. Some research activities last only a few months and will have been completed at the time an adolescent reaches the age of majority. Others may continue indefinitely. Thus, depending on the age when the child enrolled, the activities that are ongoing when a participant reaches adulthood will vary.

Second, the risks vary. Research activities that examine the effect of modifying clinical practice based on genomic test results may place participants at risk for physical harm. There is also a concern that providing genomic research results to participants may cause psychological harm,(4) although this risk appears to be very low.(5, 6) The storage of research data can confer informational risk. This is the harm, including emotional harm and adverse effects on employment, that may result if private information about participants is released without consent.(7)

Third, research activities differ with respect to opportunities for investigators and research participants to interact. Clinical trials that incorporate genomic testing into clinical care involve frequent face-to-face contact, providing an opportunity for participants to reaffirm or withdraw their agreement to participate. On the other hand, research activities that involve only analyses of previously banked DNA samples do not typically provide opportunities for investigators to interact with participants. Thus, the feasibility of renewing consent is affected by the nature of the research activities taking place when a research participant reaches the age of majority.

Regulatory Framework

The U.S. regulations that prescribe how federally funded research with human subjects should be conducted, also known as the Common Rule, do not directly address the question of how consent should be handled when a participant reaches the age of majority. However, the Office of Human Research Protections (OHRP) provides direction on this issue in a set of Frequently Asked Questions (FAQs). The responses to these questions about adolescent participants reaching the age of majority offer three regulatory alternatives.(8)

First, research activities that involve “greater than minimal risk” cannot continue unless the research participant replaces the previous parental permission with his or her own informed consent. A research activity involves minimal risk when “the probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests.”(9) When a research activity that exceeds this standard continues after the participant reaches the age of majority, the authorization for participation must be updated. This is because the person legally empowered to authorize participation has changed, from the parent to the young adult participant.

Second, IRBs have latitude to grant a waiver of consent to allow certain research activities to continue. Under the Common Rule, a waiver of consent can only be granted when the following four criteria are met: (1) the research involves no more than minimal risk to the subjects; (2) the waiver or alteration will not adversely affect the rights and welfare of the subjects; (3) the research could not practicably be carried out without the waiver or alteration; and (4) whenever appropriate, the subjects will be provided with additional pertinent information after participation.(10)

These criteria explicitly mention factors that vary widely among genomic research activities, including both the risk conferred and the feasibility of contacting participants.

Third, informed consent may not be required if the research falls outside the Common Rule's definition of “human subjects research.” Specifically, the Common Rule defines a human subject as a person from whom an investigator obtains data through intervention or interaction, or from whom identifiable private information is obtained.(11) Building on this definition, the OHRP has clarified that research using de-identified biosamples and health information is not considered human subjects research whenever the following criteria are met: (1) the private information or specimens were not collected specifically for the currently proposed research project through an interaction or intervention with living individuals; and (2) the investigator(s) cannot readily ascertain the identity of the individual(s) to whom the coded private information or specimens pertain.(12, 13)

In this regulatory context, such research on de-identified biosamples and health information is typically referred to as “non-human subjects research.”(14)

Ethical Framework

The ethical principle of respect for persons plays an important role in the Common Rule. The Belmont Report, the document that conceptually grounds the Common Rule, emphasizes a connection between respect for persons and recognition of autonomy. Respect for persons, the Report argues, “demands that subjects enter into the research voluntarily and with adequate information.”(15) This principle, then, is the basis for the three regulatory alternatives provided by OHRP.

Respect for persons can be interpreted more broadly, however. Many participants state that they expect the return of individual genomic results when they participate in a genomic study,(16) and thus investigators reasonably consider the return of this information to be an expression of respect for persons. Investigators may also regard communication with participants through birthday cards, newsletters, and updates through social media as demonstrating respect for persons. Such efforts can help participants feel included in the research they have contributed to, and may provide participants with opportunities to learn more about research.(17) In addition, these types of “engaged” approaches frequently serve the larger research enterprise by increasing participants' willingness to provide data, facilitating future research recruitment efforts, and improving relationships between the research institution and its community.

Despite the recent popularity of “highly engaged” research, these approaches should not be construed as required by the principle of respect for persons. For many studies, the use of highly engaged approaches requires significant tradeoffs. The maintenance of long-term relationships and the return of research results require significant investments of time and money. When investigators are not funded for these activities, or when important scientific aims can best be attained using less engaged approaches, respect for persons can be demonstrated in other ways. Investigators and institutions should, for example, exercise responsible stewardship of biosamples and carefully protect research data.(18)

Selecting a Consent Approach

Interventional research activities

Breaking a research study into its component activities can facilitate decisions about which of the three regulatory alternatives proposed by the OHRP is appropriate in a given situation. Investigators who conduct interventional research activities interact directly with research participants. Genomic interventional research activities include those that provide participants with genomic research results, and studies in which genomic information is used to modify clinical care. A pharmacogenomics study comparing the effectiveness of genome-guided prescribing with conventional practice exemplifies the latter.

Interventional research activities should continue into adulthood only if the participant provides his or her explicit consent. This is usually feasible, because interventional research activities typically provide ample opportunities for investigators and participants to interact. Because of this, they rarely meet the criterion that the research could not practicably be carried out without a waiver of consent.

Non-Interventional Research Activities

Many non-interventional research activities do not involve direct interaction between investigators and participants. In genomic research, the most common non-interventional activity is the analysis of banked biosamples and data. The most appropriate consent model for this type of research activity depends on two related factors: the type of information retained, and the level of engagement that investigators maintain with participants.

If no identified data are stored, there can be no engagement between investigators and participants. This is frequently the case in certain types of biorepositories where the precise timing of a participant's transition to adulthood is not known because the birthdate – an identifier under the Health Insurance Portability and Accountability Act (HIPAA) – is not retained.(19) In this type of study, analyses typically continue regardless of whether a participant has reached the age of majority.(20)

More commonly, pediatric data and samples are collected in a larger research study in which identifiers are retained. For this type of research activity, investigators must decide (1) whether they will attempt to contact participants to obtain their consent, and (2) whether it is appropriate to continue analyzing samples from participants whose consent is not obtained. In order to continue using identified samples without the young adult's consent, the IRB must grant a waiver of consent. For this reason, these decisions hinge on the criteria for a waiver of consent set out in the Common Rule. Of the four criteria, practicability is usually decisive, because non-interventional research activities usually confer only minimal risk and a waiver of consent in this context does not typically limit the rights and welfare of subjects.

Practicability can be difficult to operationalize. The Common Rule does not define practicability, and there is no guidance from OHRP that addresses its meaning. However, the Secretary's Advisory Committee on Human Research Protections (SACHRP) has offered several useful observations about practicability in a non-binding letter.(21) SACHRP emphasized that the requirement to obtain consent should only be waived when such a requirement would preclude researchers from attaining a study's research objectives as a result of biasing results, significantly decreasing statistical power, or preventing meaningful conclusions from being drawn.(21) Although SACHRP warns against judging practicability on the basis of convenience, cost, or speed alone, it makes it clear in its letter that these factors do matter if they would prevent scientifically valid research from taking place.

SACHRP's letter addresses waiver of consent at the onset of research but does not directly address the issue of contacting previous pediatric research participants to obtain their consent at the age of majority. In addition, the letter assumes that decisions about a waiver would apply to all participants in a study. In contrast, non-interventional genomic research activities often involve a mix of previously enrolled participants: some might be reached for their consent, while others might not be reachable. We believe, however, that the general principles proposed by SACHRP can be interpreted in the following way: (1) whenever possible, consent should be sought and obtained; (2) it is only appropriate to forego contacting participants for their consent when this effort would preclude investigators from attaining important research aims; and (3) in cases where a participant cannot be reached successfully, or when the IRB deems the requirement to contact participants impracticable, investigators may continue using data and samples from participants under a waiver of consent.

This interpretation of the regulation in light of SACHRP's analysis has several important implications. First, most non-interventional research activities involving identified data will require investigators to reach out to participants as they reach the age of consent. In these cases, investigators should work prospectively with their local IRB to develop a contact plan that is practicable in the context of the study's research aims and resources, and which specifies when an effort to contact a participant could be deemed unsuccessful.

In rare circumstances, however, the local IRB could determine that an effort to contact participants would be impracticable in the context of a particular study, meaning that such effort would prevent the investigators from attaining important research aims. This might apply, for example, to a study with very large numbers of participants or participants who were lost to follow-up many years prior. In this case, the IRB might issue a waiver of consent to allow research to continue without an effort to contact participants.

A second implication of this analysis is that it may be appropriate to retain identifiable samples and data after participants reach the age of majority, even if they cannot be reached for their consent. This conclusion runs counter to practices at some institutions, where IRBs may require samples and data to be de-identified when participants reaching the age of majority cannot be contacted for their consent.(22) Some IRBs even require investigators to destroy samples and data in this circumstance.(22) Notably, even though many sites adopt these practices, there is no strong evidence that participants prefer to have their data and samples de-identified or destroyed if they cannot be contacted. In one study, 54% of adult respondents believed researchers should not have to ask for consent to continue using banked samples once adolescents reach adulthood. Of the respondents who believed researchers should attempt to locate participants to obtain their consent, 41% found it acceptable for researchers to continue using the samples of participants who could not be located.(23) Patients affected with serious and chronic illnesses in particular are often enthusiastic about the use of their data and samples to improve care for future patients.(24-27) And as we will discuss in the next section, many research participants wish to receive important research results.(28-30) Both de-identification and destruction of samples and data preclude returning these results in the future.

Investigators may also have important reasons for wanting to retain a link to identifying data. They may hope that contact can be reestablished in the future so that research results can be returned to participants, so they can obtain follow-up data, or so they can follow-up with participants to answer research questions that were not anticipated when the original study was conceived.(31) Their research aims might also require the analysis of characteristics that are inherently identifying, such as dates or geolocations.

Taking these factors into account, we believe local IRBs should encourage researchers to prospectively develop and implement plans for contacting adolescents to obtain their consent when they reach adulthood. When attempts to contact are not successful, however, IRBs should be willing to utilize the waiver of consent mechanism to allow identified data and samples to be retained. Although this approach would allow research without explicit consent from young adult participants, it can still be done in a way that demonstrates respect for persons. For example, whenever possible an honest broker approach should be used to reduce informational risk. In this approach, investigators performing analyses work with coded data, and the key that links the coded data back to participant identifiers is retained by an honest broker.(32, 33) The honest broker can re-identify records if needed to reestablish contact, such as when a compelling individual research result is uncovered that the investigator wants to return to the subject.

Disclosure of Research Results

In the preceding sections, we argued that current regulations require different approaches to consent at the age of majority for interventional and non-interventional research activities. Interventional research activities nearly always require explicit consent, and non-interventional studies may qualify for a waiver of consent or a non-human subjects approach. Research activities that potentially involve the disclosure of individual research results represent a special case, however, because this practice has the potential to fall into more than one category. Specifically, disclosing research results can be an interventional research activity if this is a planned part of the study. This is especially true if the intention is to gather data about how this information influences outcomes such as healthcare utilization, adoption of preventive measures, or psychological responses. In this case the approach to consent when a participant reaches the age of majority is straightforward. Like other types of interventional genomic research, this research activity typically requires informed consent, because the ongoing interaction increases the feasibility of obtaining consent when participants reach the age of majority.

In contrast, some investigators only intend to disclose research results to participants in whom a clinically significant research result has been discovered. (34, 35) This type of research activity differs from studies that involve disclosure of results to all participants, because there is only a possibility – not a certainty – that any particular participant will receive a result. Also, such results will be selected for disclosure on the basis of their potential to provide benefit to the participant, not to answer research questions.

For these reasons, the requirements for consent differ. Consent should still be sought whenever practicable. However, when a participant cannot be reached for consent, in most cases it is reasonable to continue analysis on his or her identified sample (or to utilize an honest broker to retain identifiers) with the intention of attempting contact again if an important genomic result is discovered. When such an approach is used, a high threshold designed to minimize risks to participants should be used in deciding whether to contact a participant to disclose a result. IRBs should also ensure that the process for returning results protects the rights and interests of participants.

Conclusions

Returning to the three case vignettes introduced earlier, we can see that investigators and the IRB can anticipate these challenges at the start of the study. When such a study is proposed, investigators and the IRB should prospectively agree on a plan that describes when and how, or whether, investigators will attempt to contact participants reaching the age of majority. The plan should also describe how samples and data will be handled when participants cannot be successfully reached. As we have argued here, there is a range of options compliant with current regulations, including several that would permit important research to continue even if participants cannot be reached.

When Anna reached the age of majority, investigators were still analyzing her identified sample and had promised that they would return important research results. The investigators remained in contact with Anna and her parents, and thus had ready access to up-to-date contact information. Because reaching out to Anna for her consent is practicable, the investigators should try to contact her when she reaches the age of majority, update her on the research, and ask for her consent to continue with this work. If she declines, they must respect her choice. If investigators are unable to reach her after making a reasonable effort, they should follow the plan they had developed at the beginning of the study. This plan might allow the investigators to continue the research and retain identified data including Anna's contact information. This approach would allow them to attempt to reach her again in the future, and perhaps even to perform analyses that depend on identifying characteristics like geocodes.

Bryon's situation is similar in many ways. When he reached the age of majority, investigators were still using his sample, but had previously been unsuccessful in contacting him. At this point, investigators should make another attempt to contact him. For example, with the permission of the IRB they could access his electronic health record to see whether his contact information had been updated. Alternatively, if the resources required to contact participants would detract from other important research activities, the IRB might approve a plan that would not require the investigators to attempt to contact participants like Bryon who had been lost to follow up. As with Anna, if Bryon could not be contacted the investigators could retain identifying information with a waiver of consent. It might also be reasonable to de-identify the stored data and samples and continue research under a non-human subjects framework.

When Christopher reached the age of majority, he was still actively involved in the interventional portion of the study. Although our case did not provide sufficient detail to determine whether the physical risk these interventions posed fell above the minimal risk standard, it is at least clear that this research activity requires explicit consent. This conclusion is supported by the fact that the frequent interactions with Christopher through study visits and/or phone calls provide ample opportunities for the investigators to approach him for his explicit consent. In this case, Christopher's consent should be required regardless of the other research activities that were taking place at the time he reached adulthood.

Decisions about how to approach the transition from adolescence to adulthood can prove especially challenging in the context of genomic research studies that involve a range of research activities. We have proposed an approach for consent, consistent with ethical norms and existing regulations, that breaks complex studies into their component research activities. This approach can provide clarity, not only to identify the best approach for genomics studies, but also for any complex research that involves pediatric participants in multiple research activities.

Abbreviations and Acronyms

CSER

Clinical Sequencing Exploratory Research Consortium

dbGaP

Database of Genotypes and Phenotypes

eMERGE

Electronic Medical Records and Genomics Network

HIPAA

Health Insurance Portability and Accountability Act

IRB

Institutional Review Board

OHRP

Office of Human Research Protections

SACHRP

Secretary's Advisory Committee on Human Research Protections

Appendix 1

In addition to the authors, the following are members of the CSER Consortium:

Benjamin E. Berkman, JD, MPH, Leslie G. Biesecker, MD, Sara C. Hull, PhD (NHGRI); Sawona Biswas, MSc, MS, CGC (Children's Hospital of Philadelphia); Wendy K. Chung, MD, PhD (Columbia University); Barbara Koenig, PhD (University of California – San Francisco); Lisa S. Lehmann, MD, PhD (Dana-Farber Cancer Institute); Michelle Lewis, MD, JD (Mayo Clinic); Amy L. McGuire, JD, PhD, and Melody J. Slashinski, PhD, MPH (Baylor College of Medicine); Lainie F. Ross, MD, PhD (University of Chicago); Joseph S. Salama, BS (University of Washington); Debra Skinner, PhD (University of North Carolina); Holly K. Tabor, PhD (Seattle Children's Hospital); Susan M. Wolf, JD (University of Minnesota).

In addition to the authors, the following are members of the eMERGE Network: Cassandra Perry, MS, and Chandler, Ariel (Boston Children's Hospital); Beth Cobb, MBA, John Harley, MD, PhD, and Myers, Melanie, PhD, MS, LGC (Cincinnati Children's Hospital Medical Center); Rosetta Chiavacci, BSN, CCRC, and John Connolly, PhD (Children's Hospital of Philadelphia); Andy Faucett, MS, Samantha Fetterolf, David Ledbetter, PhD, and Janet Williams, MS (Geisinger Health System); Kelly Ehrlich, MS, Malia Fullerton, DPhil, Kelly Hansen, Andrea Hartzler, PhD, and Aaron Scrol, MA (Group Health Cooperative/University of Washington); Lucy Carruth, PhD (Johns Hopkins University); Elizabeth Chau, Chris Chute, MD, DrPH, FACMI, Iftikhar Kullo, MD, and Richard Sharp, PhD (Mayo Clinic); Murray Brilliant, PhD, Norm Frost, MD, MPH, Terrie Kitchner, and Cathy McCarty, PhD, MPH, RD (Marshfield Clinic/Essentia Institute of Rural Health); Kadija Ferryman, MS, Carol Horowitz, MD, MPH, Yolanda Keppel, Rosamond Rhodes, PhD, Saskia Sanderson, PhD, and Randi Zinberg, MS, CGC (Icahn School of Medicine at Mount Sinai); Sharon Aufox, MS, CGC, Michael Heathcote, Vivian Pan, CGC, and Maureen Smith, MS, CGC (Northwestern University); Nanibaa' Garrison, PhD (Vanderbilt University); Melissa Basford, MBA, Jacqueline Kirby, MS, Mollie Bodin Claar, MA, Lauren Melancon, and Sarah Stallings, PhD (eMERGE Coordinating Center at Vanderbilt University).

Appendix 2

The CSER Consortium is funded by NHGRI and NCI (U01 HG006485 [Baylor College of Medicine], U01 HG006500 [Brigham & Women's Hospital], U01 HG006546 [Children's Hospital of Philadelphia], U01 HG006492 [Dana-Farber Cancer Institute], UM1 HG007301 [HudsonAlpha Institute], UM1 HG007292 [Kaiser Permanente], UM1 HG006508 [University of Michigan], U01 HG006487 [University of North Carolina], U01 HG006507 [University of Washington], R01 HG006615 [Boston Children's Hospital], R21 HG006596 [Columbia University], R01 HG006600 [Columbia University], R21 HG006613 [Children's Mercy Hospital], R21 HG006594 [Johns Hopkins University], R01 HG004500 [Mayo Clinic], R01 HG006618 [Seattle Children's Hospital], R01 CA154517 [UC - San Francisco, Mayo College of Medicine, and University of Minnesota], R21 HG006612 [Vanderbilt University and McGill University], U01 HG007307 [University of Washington serving as the Coordinating Center]). The eMERGE Network was initiated and funded by NHGRI (U01HG006828 [Cincinnati Children's Hospital Medical Center/Boston Children's Hospital], U01HG006830 [Children's Hospital of Philadelphia], U01HG006389 [Essentia Institute of Rural Health], U01HG006382 [Geisinger Clinic], U01HG006375 [Group Health Cooperative/University of Washington], U01HG006379 [Mayo Clinic], U01HG006380 [Icahn School of Medicine at Mount Sinai], U01HG006388 [Northwestern University], U01HG006378 [Vanderbilt University Medical Center], and U01HG006385 [Vanderbilt University Medical Center serving as the Coordinating Center]).