IRB: Ethics & Human Research
Public Comments on Proposed Regulatory Reforms That Would Impact Biospecimen Research: The Good, the Bad, and the Puzzling
In July 2011 the U.S. Department of Health and Human Services (DHHS) published in theFederal Registeran Advance Notice of Proposed Rule Making (ANPRM) entitled “Human Subjects Research Protections: Enhancing Protections for Research Subjects and Reducing Burden, Delay, and Ambiguity for Investigators.”1The proposed changes to the regulations governing research with humans (the Common Rule) are designed to enhance protections for research participants and to improve the efficiency of the process that institutional review boards (IRBs) use to review research studies.2
The ANPRM identified biospecimen research as one area in particular need of attention for regulatory change. One of its more controversial proposals is for mandated written informed consent from individuals for any use of their biospecimens, regardless of the identifiability of the biospecimens and the original purpose for which they were obtained (i.e., clinical or research uses). The ANPRM also proposes that consent should be obtained via a standardized consent form that says biospecimens can be used for future unspecified research. Under this consent approach individuals would be giving consent in a broad, open-ended manner that would accommodate all future uses of their biospecimens.
According to guidance from the Office for Human Research Protections (OHRP) about the current provisions in the Common Rule, if a researcher is blinded to any identifiers associated with biospecimens obtained for purposes other than a current research project, their use does not constitute research involving human subjects. For research that does involve identifiable subjects, researchers can request that the IRB waive the requirement for consent, which the current regulations permit an IRB to do when four conditions are met.3Thus, under the current regulations biospecimens with identifiers could be used for research without consent if the IRB determines that a waiver of consent is appropriate. Yet recent large-scale genomic advances and findings about the inherent identifiability of biospecimens have challenged the idea that biospecimens can ever be truly deidentified, raising questions about researchers’ responsibilities to the individuals from whom biospecimens are obtained.4The proposed changes to the Common Rule would require that, in almost all cases, individuals “would have the right to allow or disallow the use of their biospecimens for research.”5
As required by federal rules governing regulatory changes, the DHHS offered the public an opportunity to comment on the proposed changes to the Common Rule and specifically posed 74 questions in the ANPRM to which it invited responses. More than 1100 comments were submitted to the DHHS and are publicly available from OHRP’s website.6In this paper, we present the findings of our content analysis of a random sample of the comments submitted to the DHHS. We focused on questions that were most informative about public views toward proposed changes relating to biospecimen research, including mandated consent, waivers of consent, and the inherent identifiability of biospecimens.
Sample Selection.Using a random number generator tool,7we selected 300 of the 1126 public comments submitted in response to the ANPRM that are available online as unique and complete comments. We downloaded each comment corresponding to the randomly selected comment number and uploaded it to NVivo 9.0, a qualitative data analysis software program. We used a keyword search in NVivo 9.0 to review the 300 comments, in batches of 100, for incidences of the following words (or their stems): “specimen,” “biospecimen,” “sample,” “tissue,” “repository,” “biobank,” “biorepository,” and “collection.” In each batch, we found that approximately half of the comments contained one or more of these keywords, identifying a total of 162 of 300 comments for further review. Following development of our analytic approach, we excluded an additional 53 comments because they did not contain material that was directly relevant to our interests regarding biospecimen research. For example, search terms for “repository” or “collection” occasionally flagged comments restricted to use of data and not biospecimens, “sample” did not always refer to biospecimens, and some comments addressed biospecimens only in relation to the federal HIPAA Privacy Rule. Thus, 109 comments were included in our final analysis.
Analytic Approach.The authors of the comments varied tremendously in their framing of responses to the ANPRM, which presented challenges to developing our analytic approach. Some authors attempted to respond to most, if not all, of the proposed regulatory changes, while others responded only to specific issues. Some formatted their responses as a large table, listing in one column the exact question and its number from the ANPRM and in another column their response to the question. Others wrote their comments in the form of a letter, addressing themes or overarching proposals in the ANPRM rather than referring to specific questions. One set of authors in our sample submitted a published article of theirs as the “comment,” presumably because they believed it addressed the proposals in the ANPRM.
Our codebook was developed in stages. We used the questions posed in the ANPRM as the basis for initial coding of the comments, first reviewing all of the ANRPM questions, then selecting those that related to biospecimen collection and use. We narrowed this list to questions that had the most direct connection to these topics and grouped them into the following broad, preliminary codes: mandated consent for biospecimens and data, waiver of consent for biospecimens and data, and identifiability (Table 1). We included questions about data as well as biospecimens because some of the coded topics were applicable to either source of research material and because some comments on biospecimen use were embedded within responses about data.
We used the first batch of 100 comments to refine the codes. Of these, approximately half had at least one biospecimen keyword. Each of us then independently applied our initial codes to these comments, compared results, resolved inconsistencies, and revised coding rules when necessary. Codes were applied whenever comment authors answered any of the corresponding ANPRM questions, regardless of whether the response was linked to specific questions. For example, we frequently applied the code “identifiability” outside of responses to ANPRM questions 56 and 57 because comment authors often used the issue of identifiability to support their opinions on mandated consent for biospecimens. After refining the codes, we used NVivo 9.0 to generate code report documents that contained all of the text assigned to each of our initial three codes (e.g., identifiability). For each code report, we met as a group to discuss themes that emerged. From these themes, we ultimately distilled our interest in the ANPRM questions to four key topics and coded each comment for the following information:
1. Does the comment support mandated consent for research use of biospecimens?
2. Does the comment support the current regulatory approach allowing for a waiver of informed consent?
3. Does the comment endorse the use of a standard general consent form to obtain consent for research use of biospecimens collected during clinical encounters?
4. Does the comment endorse the view that biospecimens should be considered identifiable in and of themselves?
As described above, our final sample for analysis consisted of 109 of the 300 comments that addressed one or more of these four main questions. For these 109 comments, we compiled and coded the reasons given in support and opposition to each of the four key questions (although not all comments included statements relevant to each topic). Summary codes were applied to the first three questions that ranked responses as either strong yes, qualified yes, or no. Responses related to whether biospecimens should be considered identifiable were coded as either yes or no. While some comment authors provided little explanation for the views they expressed, others provided considerable detail, often in the form of relaying their prior experience with an issue addressed in the ANPRM. Many comments contained extensive critiques of the ANPRM proposals; some of these offered alternative proposals. All of these additional details and critiques were also coded.
In some cases, comments revealed confusion not only about the proposed reforms but also about existing regulations governing research with human subjects. This was especially evident when comments addressed use of a standard general consent form. This seemed to stem from a lack of understanding of the ANPRM’s definition of a standard general consent form and how it might be used. The structure of the ANPRM may have contributed to the confusion. The ANPRM first mentions a standard general consent form under the section entitled “Ensuring Risk-Based Protections,”8which summarizes “five of the most significant changes being considered” in the ANPRM.9Here the ANPRM proposes that the new regulation would generally require
written consent for research use of any biospecimens collected for clinical purposes after the effective date of the new rules (such as research with excess pathological specimens). Such consent could be obtained by use of a brief standard consent form agreeing to generally permit future research. This brief consent could be broad enough to cover all biospecimens to be collected related to a particular set of encounters with an institution (e.g.[,] hospitalization) or even to any biospecimens to be collected at any time by that institution.10
The ANPRM goes on to note that a standard general consent form would allow for “broad, future research.”11However, the first question about a standard general consent form is not posed until several more pages into the document (see Question 49, Table 1). When the first question finally appears it contains no mention of the definition or the specific circumstance for its application (repurposing clinical biospecimens and data for research use); thus, readers of the ANPRM may or may not have connected these references in a coherent manner.
Affiliations of Comment Authors.All but eight of the 109 comments in our sample included information that identified the affiliation of the author(s). Many of the authors indicated who they were representing within the body of their comments (e.g., “On behalf of XYZ University . . .”). We categorized comment authors into five groups: 1) IRB or other oversight group, submitted on behalf of an institution (N = 37); 2) IRB or other oversight group member, submitted on behalf of an individual (N = 6); 3) scientific organizations including professional societies, associations, journals, or research groups, submitted on behalf of the organization (N = 39); 4) scientists as individuals (N = 12); and 5) those writing from a patient or advocacy perspective (N = 7). For ease of presentation, we combined these groups into three major categories representing the “Oversight Perspective” (groups 1 and 2), “Scientist Perspective” (groups 3 and 4), and “Patient/Advocacy Perspective” (see Figure 1, on theIRB: Ethics & Human Researchwebsite, for distribution of these perspectives in the final sample).
Content Analysis of Comments
Views on Mandated Consent for Specimens.Of the 87 comments that addressed mandated consent for biospecimen use, 9 offered full support, 9 offered qualified support, and 69 were opposed. Table 2 depicts responses by comment author affiliation. The majority of both the oversight and scientist perspective groups opposed mandated consent, whereas the majority of those in the much smaller patient/advocacy perspective group supported this consent approach.
Of the comments that supported mandated consent (N = 9), the most common argument for this approach had to do with the inherent identifiability of biospecimens. Several comments stated simply that if biospecimens are identifiable, consent is needed. As one author noted, “We will be as ethical as possible by explaining upfront that with today’s technology biospecimens are inherently at increased risk for breach of privacy due to today’s and future technology.” Several authors wrote similarly about the importance of consent as “more consistent with basic ethical principles,” with one noting that this requirement should be applied “regardless of where [in what country] biospecimens are collected. There is no good ethical justification for doing otherwise.” Others noted that “people should be permitted a measure of control.”
In contrast, comments that offered qualified support for mandated consent (N = 9) questioned the extent to which biospecimens are identifiable and, if they are, how much risk would actually be involved. These comments also raised concerns about logistical problems and the potential impact on science, but balanced these concerns against the importance of consent. According to one author, “This appears to be a very complex issue where support for autonomy may be in conflict with advancing biomedical knowledge that could be obtained from samples that are not readily identifiable, especially those from individuals of diverse genetic backgrounds.” Another argued that we need more public input before proceeding: “It is vital that . . . public attitudes be thoroughly and rigorously assessed so that the appropriate policies can be developed. Until the public is better engaged in this issue, the proposed approach is premature.”
The majority of the comments expressed opposition to mandated consent (N = 69), with logistics and the impact on science emphasized over all other considerations as reasons for opposition. Comments predicted that mandated consent would be “a logistical quagmire” and “wholly unworkable,” that it would create “an unfunded mandate on research enterprises” and “would profoundly impede clinical research for biomarkers and basic science studies.” One comment expressed the view that “Loss of ability to use certain types of archived tissues without obtaining consent may be the death knell of life-saving translational research.” Another described the potential delay for research on rare diseases, concluding, “[W]e would have to consent 50k individuals and wait 10 years to do the project.” Several comments suggested that mandating consent would introduce biases in subject recruitment, as well as decrease the numbers of available specimens and data for research, thus further limiting the conduct of science. One comment pointed to the perceived failure of a comparable regulation in the UK and warned of catastrophe in the U.S. if the proposals in the ANPRM are promulgated by the DHHS in a final rule:
A similar requirement was introduced in the United Kingdom in the early 2000s and essentially disabled research involving patient material. Moreover, this approach would adopt the very problems researchers have identified with the HIPAA Privacy Rule. … The ANPRM proposal to require consent for all future uses of biospecimens would have a chilling effect on the research landscape for decades to come and would be profoundly harmful to scientific progress.
Finally, some comments questioned whether consent in the clinic when biospecimens were collected met the standards of ethical consent and might actually increase research risks. One stated that “strengthening the current regulations and penalties for violation or misuse of archival specimens will go farther to ensure patient (subject) protection and privacy than mandated informed consent for all specimens.” Another alleged, “permission to use de-identified tissue [actually] creates a greater risk for security breaches.”
Views on Waiver of Consent for Specimens.Seventy of the 81 comments that addressed the current system for a waiver of consent for research with deidentified biospecimens offered strong support for this system, 8 offered qualified support, and 3 argued that consent should be required (Table 2). Support for mandated consent for biospecimens was strongly and inversely related to support for the current waiver system for research use of deidentified biospecimens (Figure 2, available on theIRB: Ethics & Human Researchwebsite). That is, if a comment supported the current waiver system, it was likely to oppose mandated consent for biospecimens, and vice versa. As with the views of authors who commented on mandated consent, the views of authors commenting on the waiver of consent issue were not related to their institutional affiliation or personal perspective.
Comments in support of the current waiver system (N = 70) argued that it worked well and facilitated research in a manner that did not unduly violate the rights or welfare of research participants. Using the example of tissue banks, several comment authors pointed out that access to resources that operate under the current waiver provisions is essential to research. As one individual noted, “[I]t is not possible to know how a researcher will want to use specimens in the future, and it may be impracticable to go back and get consent.” And one set of authors drew from their professional experience in supporting the current system for a waiver of consent:
The current system recognizes the value of archival material and the complexities and impracticability of obtaining consent, especially if time has elapsed or a subject is deceased since collection of their material. In our extensive professional experience working with biospecimens on a daily basis, the current system has worked well and has greatly enriched the opportunity for discoveries that were unknown at the time of collection.
Two comments in the category of qualified support for mandated consent (N = 8) noted uncertainty about how to apply the waiver criteria, especially in determining how to apply the criterion for impracticability. One comment noted that identifiability might be an issue for some populations, and in those cases, the waiver might not be ethical. Another argued that while the IRB should be permitted to issue a waiver when potential risks to the individual are low and societal benefit is great, “The criteria related to impracticability will differ among institutions and assessment of such should be left to the IRB reviewing the research. At the same time, some examples of what might contribute to ‘impracticability’ would be helpful to IRBs as they try to make consistent determinations.” In the view of one set of comment authors,
We believe that requirements for consent for future use…should be based on the identifying information or likelihood that a research subject could be identified. If data were de-identified and non-genotypic/phenotypic data were collected, then consent should not be required. In addition, we believe that the final regulations should allow the flexibility to waive consent if the circumstance warrants.
Comments that expressed opposition to the current system for a waiver of consent (N = 3) acknowledged that obtaining consent is difficult or “impracticable” under certain circumstances, but argued that impracticability should not be used to justify failure to adequately protect human subjects. According to one comment author,
The Common Rule should accommodate the reality that not all possible research uses of biospecimens and data can or will be contemplated at the time of collection. This reality, however, must be balanced with assurances that future uses of human tissue and personal data will be ethically tenable . . . . Obtaining consent for secondary analysis of existing biospecimens or identifiable data should not be presumed impracticable simply because the sample size is “too large.”
Views on Use of a Standard General Consent Form.As discussed above, analysis of the comments revealed confusion regarding a standard general consent form for research with biospecimens originally collected for clinical purposes. Confusion seemed related to what the ANPRM actually meant by a standard general consent form and how it would be used.
Sixty-seven comments addressed use of a standard general consent form (Table 2). Of the comments that endorsed its use (N = 8), most gave no explicit reason, and those that did tended only to support simply getting consent for biospecimen use, rather than focusing on the use of a standard general consent form specifically. For example, one set of authors stated, “We believe that research participants should always have the ability to consent to what happens to the clinical data and biospecimens collected from them, and we support the suggestion that there should be a standard and general form so that individuals can consent to future open-ended use of their biospecimens for research purposes.” Some authors who opposed mandated consent reluctantly noted that if obtaining consent for specimens becomes mandatory, then using a standard general consent form would at least make it easier to do so.
Comments that reflected qualified support for a standard general consent form (N = 17) expressed willingness to endorse use of such a form but provided stipulations about where more robust forms of consent should be used. For example, one comment stated that a standard general consent form should not be used when biospecimens will be used for commercial purposes, and another indicated support for a standard general consent form that included tiered options for participants to determine how biospecimens could be used. One set of authors thought that a standard general consent form would create logistical burdens for clinical staff and researchers and thus undermine its effectiveness:
We believe that a standard, global consent form would be very desirable, but the use of such a form may present substantial challenges. [It] could add considerable time and effort to those clinical situations in which it is used and we worry that neither the subjects or those obtaining consent will fully understand the implications and thus consent may not be truly informed.
Comments that expressed opposition to the use of a standard general consent form (N = 42) fell into two broad and divergent categories: those that opposed a standard general consent form because of opposition to mandated consent and those that opposed a standard general consent form because it could not accomplish obtaining truly informed consent. The latter category of opposition is reflected in this comment:
It is difficult to envision how a “standard, brief general consent” could adequately provide the information needed for an individual to make a meaningful decision about research use of their existing tissue and data. By definition, the contemplated form would seem to be too vague to provide useful information about anticipated uses, risks and benefits of the materials. Of particular concern, the concept of “all future uses” changes with time. . . . Additionally the concept of risk can change with time. An individual’s medical information incrementally changes over time. Incremental change that includes sensitive personal data may not have been considered at the time that “consent” was obtained.
Almost no comments in our sample were directly responsive to the second part of the ANPRM’s question 49, which asks whether there are options other than a standard general consent form that should be considered, such as a public education program coupled with notification and an opt-out option.
Views on Whether Human Specimens Should Be Considered Identifiable.Sixty of the 109 comments we reviewed addressed whether a human biospecimen ought to “be considered identifiable in and of itself.” While many of these views were expressed in response to the ANPRM’s specific questions regarding identifiability, they were also found elsewhere, such as in arguments to support or oppose the proposal to mandate consent for all biospecimens. Of the 60 comments, 15 stated that biospecimens should be considered identifiable (Table 2). The majority (44) stated that they should not. One set of authors said they were divided on this issue, with some believing that biospecimens should be considered identifiable and some believing that they should not (comment not included in Table 2). In contrast to the lack of association between comment author affiliations and views on other key questions, authors of comments on the question of identifiability appeared to be associated with affiliation: those affiliated with the scientist perspective were somewhat more likely to state that biospecimens should be considered identifiable (29%) than those in the oversight group (19%).
The most interesting finding regarding responses related to identifiability is that virtually all comments that included rationales acknowledged that specimens could be identifiable with proper conditions in place. These comments then differed as to whether, given the potential for identifiability, specimens ought to be considered identifiable.
We identified two specific arguments in the comments reflecting the view that specimens should be considered identifiable (N = 15). One argument was a practical one about right conditions for reidentifying biospecimens: “It is impossible for biospecimens to be completely anonymous—DNA matching might be cumbersome and expensive but it is possible in practice and that is enough to render anonymity impossible.” The other argument, expressed by a small minority, focused instead on a perceived beneficial outcome for biospecimen contributors: “The advantage of considering all future research with biospecimens to be research with identifiable information is that the privacy and confidentiality of the individuals from whom those biospecimens were obtained will be adequately protected.”
Of comments that stated that biospecimens should not be considered identifiable (N = 44), the vast majority argued that because external tools are necessary to identify biospecimens, they should not be treated as identifiable in and of themselves. As one author noted, “Use of DNA sequencing to associate a biological specimen with a specific individual requires an identified reference sample. Neither the biological technology nor the computing technology to otherwise establish identity is available at this time.”
Another comment raised the question, “If the tissue specimen is not being used for DNA analysis, why must the tissue specimen itself be considered identifiable?” The comment author went on to make several points about the issue of identifiability:
The majority of current and prospective research using tissue does not involve analyses that could be used to generate a person-specific genetic fingerprint . . .
The same oversight would be provided for non-genetic research using de-identified specimens as for genome sequencing of identifiable specimens. The result could be both over-protection that increases the difficulty of doing certain lower-risk tissue research and under-protection that allows more sensitive research to proceed with no independent oversight.
While many comments that opposed considering biospecimens as identifiable in and of themselves argued that the existing system of “standard de-identification” was sufficient to protect human subjects, several suggested alternatives that sanctioned improper uses of identifiable data. One argued for “the use of thoughtful but strong restrictions with genuine negative consequences for violations (use of such data for any purpose that was not approved by an IRB). Restrictions could be similar to those in use for restricted access to secondary datasets of behavioral, educational, and sociological data.” Another referred to legal mechanisms to protect privacy:
Informed consent in and of itself cannot guarantee privacy protection. Enforcing and strengthening penalties against violations of the Genetic Information Nondiscrimination Act of 2008 (GINA law), specifically the OHRP Guidance on the Genetic Information Nondiscrimination Act: Implications for Investigators and Institutional Review Boards (dated March 24, 2009), will go farther toward providing actual protection.
Some alternative proposals called for further input. One was framed formally, in terms of a task force that would regularly review technological advances that may “increase our capability of identifying individuals from smaller and smaller sequences.” Another suggested frank public discussions of “actual” uses and dangers, rather than “knee jerk” regulatory reactions:
Rather than this knee-jerk reaction to growing fears around data security and privacy, the agency should engage the public in a frank discussion about the ways their personal and medical information is actually used and what the actual dangers are. Some researchers have experimented with an approach wholly opposite to the proposed rule by acknowledging the inherently identifiable nature of genetic material, accepting it, and making the information public (with direct identifiers included). Such an approach suggests that perhaps the dystopic scenarios of science fiction are best avoided through transparency rather than placating ‘consent forms’ that do little to inform subjects.
In sum, while many supporters and opponents of the proposition that biospecimens should be treated as inherently identifiable recognized that biospecimens can be identified under certain conditions, opponents argued that these conditions are currently so cumbersome, expensive, and unlikely that biospecimens should not be considered identifiable. Additionally, many of those who argued that biospecimens should not be considered identifiable, and even some who argued they should, proposed that researchers should be sanctioned for failing to protect individuals’ privacy or misusing technologies in order to attempt to identify the individuals from whom the biospecimens were obtained.
We examined 109 comments on the ANPRM’s proposals about research with biospecimens that mentioned at least one of the four proposals we focused on: mandated written consent, the current system permitting a waiver of consent, the use of a standardized general consent form, and the inherent identifiability of biospecimens. The comments in our sample gave somewhat more attention to the first two proposals (87 and 81 comments) than toward a standardized general consent form and identifiability of biospecimens (67 and 60 comments). IRB members, university officials, and scientists writing on behalf of their organizations were the authors of most of the comments.
Reflecting the tension presented in the title of the ANPRM (“Enhancing Protections for Research Subjects and Reducing Burden, Delay, and Ambiguity for Investigators”), the proposals for revising the Common Rule related to biospecimens seem to offer a series of trade-offs. First, in an effort to enhance protections, the ANPRM proposes that biospecimens used for research must be collected with consent, regardless of whether the biospecimens were originally collected for research or clinical purposes. However, appearing to recognize that this might be burdensome, it suggests obtaining consent through use of a short standard form that requests broad consent for all future research uses of biospecimens. Public comments also seemed to reflect a trade-off. Most of the comments we reviewed reflected opposition to the proposal for mandated consent, and many of these similarly argued that the current system for a waiver of consent was adequate protection for research participants. However, of the comments reflecting opposition to mandated consent, many of the authors said that if consent was mandated, they would support the use of a standard general consent form in order to make the process easier (although warning that even the use of a standard general consent form would still make the process logistically difficult). Most comments in our sample seemed to presume that research with biospecimens would advance science and used this presumption to weigh the risks and benefits of research with biospecimens. Others have argued, however, that the potential benefits of research should not be used to trade off against reasonable regulatory requirements.12Moreover, some scholars maintain that the ANPRM proposals do little to resolve the ethical and legal concerns related to research with human biospecimens and may in fact exacerbate some of the problems.13
Botkin and colleagues14also analyzed responses to the ANPRM, looking specifically at comments addressing biospecimen research from authors affiliated with 26 institutions in the National Institutes of Health’s Clinical and Translational Science Awards consortium. Perhaps predictably, they found that the majority of comments from individuals at these research-oriented institutions opposed the ANPRM’s proposals related to biospecimens. Our study expands on their findings since the comments in our sample were chosen at random and reflect the broader range of perspectives contained in the entire collection of comments submitted to the DHHS. Overwhelmingly, the comments in our sample were from people actively engaged in research with biospecimens or involved in the oversight of that research. Our sample included very few private citizens, research subjects or their advocates and thus reflects the small number of people with these perspectives who chose to submit comments. However, the data we gathered from advocate comments suggest that these perspectives may differ in important ways from those in the scientist or oversight world. While the small sample size of advocate comments is an obvious limitation, the majority of these comments supported mandated consent and argued that biospecimens ought to be considered inherently identifiable; the opposite was true of both the oversight and scientist groups.
Numerous survey and focus group studies indicate that while the public is generally supportive of broad uses of their biospecimens and associated data, most individuals want to know how they will be used in research and be asked permission for this use.15The heart of this issue appears to be that the transparency involved in asking for consent bolsters trust in the research enterprise.16Perhaps though, as one of the comments in our sample suggested, transparency and trust might better be attained through large-scale public education about the various ways biospecimens and data are acquired and used in research, as well as about the various risks and benefits of their use, rather than relying on a broad consent template.
While one might have predicted that the oversight perspectives in our sample would favor additional requirements aimed at enhancing human subjects protections, these comments were no more supportive of the proposed mechanisms than the scientists in our sample. In fact, more comments from those in the oversight group opposed the use of a standard general consent form than comments from those in the scientist group, although potential confusion over the standard general consent form proposal dictates caution in interpreting these comments. However, we believe that the relative lack of support for proposed reforms from those in oversight positions may reflect the complexity of issues at stake when considering research use of biospecimens and the workplace realities in applying current protections. In fact, Edwards and colleagues17drew a similar conclusion from the results of their survey of genetic researchers and IRB officials, noting that there was a range of viewswithineach group on a variety of ethical issues related to genetic research with human biospecimens.
While the ANPRM represents a much-needed first step in revising the regulations related to human subjects protections, perhaps the proposals were too ambitious. At the very least, the ANPRM presented readers with internal contradictions among the various proposals, and this confusion was reflected in commentary on the questions we investigated. Thus, it was not always clear if comment authors really understood the current regulatory landscape or what was being proposed in the ANPRM to change it. Additionally, it is not clear when and how the rule-making process will proceed, with some suggesting that the process is stalled.18At the time of this writing, the Notice of Proposed Rule Making that constitutes the next step has yet to be issued by the DHHS, although there have been public assurances that regulatory reform is a priority and that the process will be moving ahead.19When and if it does, there will be opportunity for another round of public comments on the next iteration of proposals. We offer this systematic analysis of a random sample of comments on the proposals in the ANPRM not only to assist in the next stages of the regulatory-reform process but also to provide insight on the current views of some members of the research community and the general public about human subjects protections for research with biospecimens.
R. Jean Cadigan, PhD,is assistant professor in the Department of Social Medicine at the University of North Carolina School of Medicine;Daniel K. Nelson, MS, CIP,is professor of social medicine and pediatrics at the University of North Carolina School of Medicine;Gail E. Henderson, PhD,is professor in the Department of Social Medicine at the University of North Carolina School of Medicine;Anders G. Nelson, BSBA,was a research assistant in the Department of Social Medicine at the University of North Carolina School of Medicine;Arlene M. Davis, JD,is associate professor of social medicine in the Department of Social Medicine at the University of North Carolina School of Medicine.
The authors thank Kriste Kuczynski for assistance with graphics. Funding for this research was provided by several grants: R01HG005227-01A1 (G. Henderson, principal investigator, “From Specimen to Biobank: Using An Organizational Perspective to Study ELSI Issues”), from the National Human Genome Research Institute; 5UL1RR025747-04S1, a supplement to the UNC CTSA U54RR024382-01A1 (M. Runge, principal investigator, “Enhancing Biobank Capacities Across CTSAs”), from the National Institutes of Health and the University of North Carolina Center for Genomics and Society, P50HG004488 (G. Henderson, principal investigator), from the National Human Genome Research Institute. The content of this article does not necessarily reflect the views of the funding agencies.
The figures for this article are available on theIRB: Ethics & Human Researchwebsite.
1. The Department of Health and Human Services and the Food and Drug Administration. Human subjects research protections: Enhancing protections for research subjects and reducing burden, delay, and ambiguity for investigators. Advance notice of proposed rule making.Federal Register2011;76(143):44512-44531. https://www.hhs.gov/ohrp/humansubjects/anprm2011page.html.
2. U.S. Department of Health and Human Services. ANPRM frequently asked questions (FAQs) July 2011. https://www.hhs.gov/ohrp/humansubjects/anprmqanda.html.
3. 45 CFR 46.116
4. Gymrek M, McGuire AL, Golan D, et al. Identifying personal genomes by surname inference.Science2013;339(6117):321-324.
5. Emanuel EJ and Menikoff J. Reforming the regulations governing research with human subjects.NEJM2011;365(12):1145-1150.
6. U.S. Department of Health & Human Services. ANPRM for revision to Common Rule. https://www.hhs.gov/ohrp/humansubjects/anprm2011page.html.
7. Research Randomizer. www.randomizer.org.
8. See ref. 1, pp. 44514-44515.
9. See ref. 1, p. 44515.
10. See ref. 1, p. 44515.
11. See ref. 1, p. 44519.
12. Rothstein MA and Shoben AB. Does consent bias research?American Journal of Bioethics2013;13(4):27-37, p. 28.
13. Javitt GH. Take another little piece of my heart: Regulating the research use of human biospecimens.Journal of Law, Medicine & Ethics2013;41(2):424-439; Williams BA and Wolf LE. Biobanking, consent, and certificates of confidentiality: Does the ANPRM muddy the water?Journal of Law, Medicine & Ethics2013;41(2):440-453.
14. Botkin JR, Anderson R, Murray T, et al. Proposed regulations for research with biospecimens: Responses from stakeholders at CTSA Consortium institutions.American Journal of Medical Genetics PartA 2014;164(4):892-897.
15. Ludman EJ, Fullerton SM, Spangler L, et al. Glad you asked: Participants’ opinions of re-consent for dbGap data submission.Journal of Empirical Research on Human Research Ethics2010;5(3):9-16; Murphy J, Scott J, Kaufman D, et al. Public perspectives on informed consent for biobanking.American Journal of Public Health2009;99(12):2128-2134; Tomlinson T, Kaplowitz SA, Faulkner M. Do people care what’s done with their biobanked samples?IRB: Ethics & Human Research2014;36(4):8-15; Trinidad SB, Fullerton SM, Bares JM, et al. Informed consent in genome-scale research: What do prospective participants think?AJOB Primary Research, 2012;3(3):3-11.
16. See ref. 15, Tomlinson et al. 2014; Trinidad SB, Fullerton SM, Bares JM, et al. Genomic research and wide data sharing: Views of prospective participants.Genetics in Medicine2010;12(8):486-495.
17. Edwards KL, Lemke AA, Trinidad SB, et al. Genetics researchers’ and IRB professionals’ attitudes toward genetic research review: A comparative analysis.Genetics in Medicine2012;14(2):236-242.
18. Basken P. Federal overhaul of rules for human research hits impasse.Chronicle of Higher EducationMarch 7, 2013. https://chronicle.com/article/Overhaul-of-Rules-for-Human/137811/; Puglisi T. Reform within the Common Rule?Hastings Center Report2013;43(1):S40-S42.
19. U.S. Department of Health and Human Services. Minutes from March 12-13, 2014 Secretary’s Advisory Committee on Human Research Protections Meeting. https://www.hhs.gov/ohrp/sachrp/mtgings/index.html.