Abstract
Much is known about patient attitudes to ethical and legal questions in the context of biobanking, particularly regarding privacy protection and consent. However, little is known about the attitudes of medical researchers who use biobanks for research to these issues. Four focus groups with medical researchers in the UK were conducted in 2010–2011. The study highlights a range of issues associated with the research oversight and consent process (including obtaining ethical approval to use biobank samples and particular concerns for international studies), the benefits and limitations of broad consent and the possibilities of revoking consent. Many of these issues originate in the relatively static consent processes that currently govern the biobanking process. However, it is now possible to develop reliable, dynamic processes using information technology that can resolve many of these ethical and legal concerns. The ‘dynamic consent’ approach therefore offers the opportunity to fundamentally transform the process of medical research in a manner that addresses the concerns of both patients and medical researchers.
Introduction
Biobanks raise important ethical and legal questions, particularly around participant consent, withdrawal, privacy protection, and research governance. While much is known about patient attitudes to these issues in the context of biobanking, far less is known about the attitudes of medical researchers who use biobanks for research. This study draws on insights from 4 focus groups composed of medical researchers held in 2010–2011, as part of the EnCoRe research project to develop technology-enabled, dynamic consent for biobank donors1. In this context, the focus group participants were asked to discuss issues around the governance of biobanks, particularly in the context of research participants granting and also revoking consent. This inevitably involved a discussion of the effects of ethical oversight on their daily research activities.
In this paper, we discuss the key governance issues associated with biobanks around consent and withdrawal, before reviewing prior research on patient attitudes to these governance issues. We describe the empirical research data collection and analysis undertaken with the researchers who were involved with the Oxford Radcliffe Biobank (ORB), UK in various capacities. We use quotations from the focus groups to give ‘face validity’ to the analysis. The paper ends with a discussion of researchers’ concerns about the dynamic consent model.
Consent and Revocation in Biobanking
Caulfield [1] notes that despite the rapid growth of interest in large, population-based biobanking initiatives, a range of ethical, legal and social policy issues still remain unresolved [2]. Moreover, despite international attempts at harmonization, there is widespread recognition that there are diverse approaches to the governance of biobanks [3,4]. This can result in a lack of certainty in biobanking research, thus increasing the (transaction) costs of many studies. Initiatives such as P3G and ISBER, as well as European-funded projects such as PHOBE, ENGAGE, BBMRI, and BIOSHARE, continue to be successful in working toward international harmonization.
Perhaps the most apparent area of divergence is in the role of informed consent and withdrawal/revocation of consent in cases in which personal data and tissue samples are collected for use in biobanks. Whilst withdrawal is the term traditionally used in biomedical research ethics, including biobanking governance, in writing about information governance and the management of personal data in other and broader contexts, EnCoRe partners propose the term ‘revocation’. It includes withdrawal in biomedicine, but it has a wider application. The informed consent of patients and their right to withdrawal from participation plays in a pivotal role in the design of all medical research [5] and is also understood at this time to be the means by which their dignity and autonomy can be respected. Although the precise content of informed consent is contested and there has been little discussion of the nature of withdrawal, they remain the 2 key principles of medical research ethics [6]. The requirement for informed consent and the right to withdrawal are enshrined in law in many jurisdictions and, accordingly, research governance and oversight mechanisms have been designed around them [7]. The particular challenge for consent that biobanks highlight relates to their very nature. As Tupasela [3] notes: ‘The use of large epidemiological sample collections in genome research has raised numerous issues concerning the status of informed consent in reusing samples for purposes other than those they were originally intended for’ (p. 66). To address this, patients are increasingly asked to give ‘broad’ consent that is intended to cover many or all possible future research uses of their samples [8,9].
While it offers a pragmatic solution to the current problems faced by biobanks, broad consent arguably fails to meet the tenets of informed consent as provided by the 1964 Helsinki Declaration of the World Medical Association. As defined there, informed consent means ‘each potential subject must be adequately informed of the aims, methods, sources of funding, any possible conflicts of interest, institutional affiliations of the researcher, the anticipated benefits and potential risks of the study and the discomfort it may entail and any other relevant aspects of the study’ [10]. Unless a loose definition of ‘informed’ is taken (perhaps one where the information provided covers all issues that are relevant for a person’s choice [9]), the potential subject cannot be adequately informed at the time of recruitment about all future research that may take place on the sample. The use of ‘broad consent’ that describes the broad purpose of the research, as opposed to all of the research uses, has been seen as practical solution to these challenges. However, difficulties arise when new research techniques and research questions raise ethical concerns that could not have been foreseen at the time the initial consent was given.
A common response to these problems with broad consent has been to require approval from a research ethics committee (REC) or an institutional (ethics) review board (IRB) [8] that acts ‘on behalf of’ the patient who provided the sample and gave the initial broad consent [11]. Thus, the REC will review research proposals that seek to use biobank samples and will decide whether the existing ‘broad consent’ is, in fact, broad enough to cover this future use [12] or if new consent must be obtained. In forming their decisions, RECs may consider the scientific merit of the proposed study as well as existing legal and regulatory requirements (such as the Human Tissue Act 2004 in the UK [13]). Anonymization and coding has been used as a way of removing the need for new consent from individuals for additional research on the same samples or information (secondary research). This is because data and samples are no longer identifiable, but these mechanisms have also been used as a means to protect privacy. However, with greater sharing of data and more data being publicly available on the internet, disclosure risks increase and the effectiveness of coding techniques are brought into question [14].
Another key aspect of the Helsinki Declaration [10] concerns the revocation of consent: ‘The potential subject must be informed of the right to refuse to participate in the study or to withdraw consent to participate at any time without reprisal’ (p. 3). In the context of biobanking, this raises particular practical and ethical concerns regarding when and how the withdrawal of consent can apply and how it is effected [15].
Prior Research
A number of empirical studies have sought to gauge the attitudes of various stakeholders, particularly variously defined ‘publics’ [16] in terms of these contested concepts. Frequently, these studies have explicitly sought to enhance public support for and hence participation in biobanks.
Many such studies involve large-scale, representative telephone and face-to-face surveys of ordinary citizens [e.g. [17,18]]. Such quantitative studies have been usefully complemented by detailed, qualitative studies of citizen attitudes [19,20]. In recent years, a number of studies of deliberative democracy workshops have been undertaken on biobanking topics [21,22,23]. These workshops have been particularly effective at articulating the views of ‘ordinary people’ about technologically sophisticated topics [24]. These studies have demonstrated that citizens are able to understand and even to challenge perspectives presented by experts and stakeholders [4,25], even if their understanding of the consent given in normal circumstances is less clear [26]. Other qualitative studies since the early 2000s have involved focus groups of patients [e.g. [27,28,29]] and healthy volunteers [30].
These studies have sought to understand patient attitudes toward a variety of issues, including trade-offs between individual privacy concerns and societal benefits of research [4] and patient perspectives on whether seeking explicit reconsent for new research uses would increase participation [17].
They have highlighted the important role played by research governance mechanisms (such as restraints on the commercialization of results, transparency of research goals and assurances of confidentiality) in supporting participation in biomedical research more generally [e.g. [18,22,31]]. They also indicate a growing shift towards better research accountability and the consideration of research participants’ expectations in building more intuitive and adaptive governance models [32,33,34].
While there is increasing knowledge of the attitudes of patients and donors, far less is known about the perspective of medical researchers [35,36,37,38]: How do they view the current processes of obtaining REC approval for their research, particularly in an increasingly global scientific arena? What constraints result from existing processes around obtaining consent and informing patients? How might moves to provide patients with more control over their data, for example by providing effective revocation of their consent, affect the work of medical researchers?
The aim of this study is to add to our knowledge of the attitudes of medical researchers, as they are an important stakeholder in the development of the biobanking governance. A series of focus groups with medical researchers from the Oxford Radcliffe Biobank (ORB) were undertaken in 2010–2011 as part of the EnCoRe cross-disciplinary research project. The intention of the focus groups was to elicit the views of medical researchers on the issues of biobank research governance and, in particular, the new patient interface using the ‘dynamic consent’ approach developed in the EnCoRe project.
In this model, consent is not a mere communication exercise, but a bidirectional, ongoing, interactive process between patients and researchers [39]. Individuals can make and express preferences about the choices they are given about the use of their data and samples for research and change or revoke these over time. This system includes run-time compliance monitoring, cryptographically based linking of consent preferences to data items and the ability to cascade consent preferences across institutional boundaries. The benefit of this interface is that it enables individuals to exercise autonomy by giving informed consent for new types of research in real time rather than being asked to give a broad consent at the beginning of the research process, when they are recruited into a biobank. The potential benefits for the research process of a dynamic consent approach are that recruitment is easier, less costly and more efficient, the legal and ethical requirements of consent can be met with ease, there is greater transparency and accountability in the research process, and research findings can be returned to research participants as part of a personalised-medicine approach [40].
Data Collection
This study draws on empirical data from 4 focus groups with a range of medical researchers associated with the ORB. In particular, they included a range of expertise and specialisms from basic researchers to researchers with dual clinical and research roles, recruitment support coordinators, research coordinators, and data managers. The application of focus groups varies considerably between disciplines, but has not been often used to research ethical issues associated with biobanks. This section therefore briefly introduces the focus group approach to data collection, the characteristics of the focus groups and how they were run and the methods that were used to analyze the data generated.
Focus Groups as a Methodological Approach to Data Gathering
As a research method, focus groups provide the opportunity for the collection of detailed, qualitative data about a particular product, concept, or innovation and offer distinct advantages over individual interviews [41]. They are intended to be interactive activities whereby differing viewpoints and perspectives emerge from the interaction between focus group participants. As such, focus groups are a powerful and flexible means to generate hypotheses, explore opinions, ideas, or attitudes and test new products or concepts (p. 1) [42].
Attitudes and perceptions are developed, in part, by a structured process of interaction with others in the group that would not be expected to emerge in a sequential series of one-on-one interviews with the same individuals [43]. Indeed, in some cases, group interaction might cause individual perceptions and opinions held by focus group members to shift over the course of the discussion or to be more clearly articulated in response to countervailing perspectives, developments that the focus group process can help document quite effectively.
Because managing the dynamics of focus groups requires specialist skills, it is important that the facilitator of the focus group creates a permissive environment that nurtures different perceptions and points of view without pushing for consensus [41,42,43]. The facilitator needs to ensure that he or she does not give any implicit support or recognition of particular viewpoints or speakers, as this might hamper the contributions of other participants and other perspectives [43]. This caution can involve subtle body language and using neutral responses such as ‘uh-huh’ rather than ‘yes’ or even ‘thank you,’ which can convey approbation.
It is common to have a second person involved as a ‘note taker’ who also supports the facilitator. This individual can focus on the content of the discussion rather than how it is being run and is a useful backup in case the recording devices fail. Depending on the requirements of the focus group, it is common for the focus group discussion to be recorded (audio and/or video) and for the recording later to be transcribed. As with all such recordings, the informed consent of participants is required. A fuller discussion of the background to focus group research can be found in Whitley and Kanellopoulou [44].
The EnCoRe Biobank Focus Groups
Ethical Approval
Research ethics approval was sought and obtained from a National Research Ethics Service for this study, to conduct a series of focus groups with non-NHS-based as well as NHS-based researchers. For reasons of expediency, ethics approvals were sought in 2 phases, to avoid unnecessary delays with recruitment. The sessions were advertised with the help of the ORB biobank manager through targeted e-mail lists, with attachments of the project information leaflet and a detailed information sheet, which were also available on the ORB and HeLEX Centre websites. This method allowed us to recruit a wide range of specialisms from basic clinical researchers to research and recruitment managers, data programmers, clinical research fellows, group leaders, and senior consultants across various cancer, pathology and oncology groups, involved or interested in data access management practices in the biobank. The focus groups were held in 2 sites, at the University of Oxford and at the John Radcliffe Hospital, in Headington, Oxford. The 4 focus groups were held between December 2010 and October 2011 (table 1).
Process
The literature suggests that controlling the dynamics of focus groups can require specialist skills and so for this study a professional, external facilitator was hired to facilitate all the focus groups [41,42,43]. Audio recordings were made of the 4 focus groups. In addition to the facilitator and participants, the first 2 authors attended each focus group to act as note takers. One of the authors undertook the role of facilitator for the final focus group because it was important to run the focus groups within a certain period of time, and we would have had to seek additional REC approval to use the external facilitator for this final group.
The focus group began with the facilitator giving a brief outline of what the focus group was trying to achieve. Participants were told that there were no ‘right answers’ but that, instead, the project was interested in learning their expert opinions about the topic. Participants were asked to sign a consent form and were informed that the session was being recorded. They were also told that ‘The discussions will be recorded, transcribed and analyzed for key themes that emerge from the discussion. The data from the session will be available to all researchers working on the project, but the transcripts will be kept anonymous. The data may also be used in reports and publications and direct anonymised quotations from the transcript may be used in published output.’
As required by the ethics approval for this part of the study, an interview agenda had been prepared, and this was shared with the facilitator beforehand. The facilitator then used this interview agenda to structure the focus-group discussions. A copy of the interview agenda is attached as an appendix.
Analysis
A commercial transcription company produced transcripts of the 4 focus groups from the recordings. The transcripts were then formally coded by the first 2 authors using the Atlas.ti qualitative analysis software (version 6.2.27). The transcripts were loaded into Atlas.ti, and the relevant text was coded into approximately 30 themes and subthemes derived from the literature. The codes were normally allocated to a ‘paragraph’ of the transcript, although occasionally the code would run over a number of related paragraphs. Some paragraphs, naturally, were allocated a number of different codes [45].
After the initial codes were produced, the first 2 authors then reviewed the codes iteratively to produce 11 high-level ‘code families.’ Each of these code families was then analyzed to provide further detailed subcodes within each family. Around 70 subcodes were produced in this way and these drive the analysis presented below.
A selection of coded quotations is presented in the remainder of this paper. Each quotation is accompanied by a 2-part code. The first part of the code identifies the focus group it was taken from; see the codes in table 1. Because the analysis explicitly did not identify which participant made the utterance, the number in the second part of the code refers not to a particular participant, but only to the paragraph number the quotation was taken from. Also, because the raw data is a transcription of the natural conversation, some work has been undertaken by the first author to make the quotations more legible. In the focus-group quotations below, clarifications and additional words appear within parentheses and transcription annotations are in curly brackets.
For example, ‘It doesn’t really matter if my sample was really contributing, it’s ... my participation that contributed’ (FG4 327) was taken from focus group 4, paragraph 327. The original transcription was ‘It doesn’t really matter if my sample was really contributing, it’s that my participation that contributed,’ which as indicated was adjusted slightly to improve the readability of the quotation.
Results
The Challenge of Obtaining Ethical Approval
Perhaps the most striking feature of the focus groups was the frustration felt by most participating medical researchers at the ways in which seeking ethical approval for research was affecting their daily (research) work. The metaphors used by the researchers to describe this frustration varied: some just complained that ‘it’s difficult, yeah’ (FG2 39); others felt that the requirements for ethical approval were ‘just bonkers’ and ‘completely over the top’ (FG2 589), ‘getting out of hand’ (FG2 635), ‘horrible for everybody’ (FG2 665), causing them to have a ‘lot of sleepless nights’ over it (FG2 63) and ‘all gone a bit mad, really’ (FG3 353).
Some participants talked of having to go through a ‘number of different hoops before you can get your access to that tissue’ (FG1 619) and given the speculative nature of much research, others complained that ‘people are much less willing to do the paperwork needed to sort out the ethics on the off chance you might get something’ (FG3 117).
As already noted, a particular concern related to the time and effort expended on obtaining ethical approval. As a result, one coping strategy was to ‘make our ethics as broad as possible because we are trying to give ourselves a bit more room to manoeuvre without actually having to go and fill in another form and say yes, but now we’ve decided that we want to do slightly more or want to look at something slightly differently’ (FG4 267). This same respondent noted, however, that there was a conflict ‘between how tightly things are drafted and how much you can do within your possibilities’ (FG4 267).
As the discussions continued, some participants expressed frustration with the whole process, with one remarking that they would be ‘going back to botany quite frankly if we get to that [overly constrained] stage’ (FG3 229). Some noted that the Human Tissue Act was a problem because it ‘was drafted very quickly after the Alder Hey thing’ (FG4 247).2 In particular, it was felt that ‘everybody is so nervous now, they cover themselves absolutely for everything and it doesn’t necessarily help the patient to give informed consent’ (FG4 247).
In general, focus-group participants saw the process of applying for REC approval as being time-consuming, which they detailed in terms of both filling out forms and appearing before the ethics committee. These problems were not alleviated by the fact that what might require REC approval could vary from one health authority to another. Indeed, in one case, a participant noted that ‘we ended up having to send somebody to go and do the experiments in their lab because it was easier than the paperwork to try and transfer the samples’ (FG3 113).
Nevertheless, despite these grumbles, there was widespread recognition of the role that REC approval plays in ensuring public trust and support for the scientific work that they were doing. For example, it was recognised that, by working under the auspices of an ethics committee, the researchers were asking their participants ‘do you trust the ethics committee? because that’s where we’re putting our trust in’ (FG1 617).
There was some support for pro forma ethics forms, standard phrases and the ‘magic sentence [about future genetic testing]’ (FG2 61) that would prevent you from having to ‘reinvent the wheel’ (FG2 591) each time you sought ethics approval. Such an approach would mean that ‘if you have these 6 points on your consent form, you’ve probably covered most of the important points’ (FG2 595). However, this was countered by a recognition that each study was going to be different and that going through the forms ‘allowed you think in more detail about these special things [such as whether feedback to participants would be provided]’ (FG2 623). The advantages and limitations of broad consent are discussed in more detail below.
There was also an appreciation for specialist staff who may either be based in the respondents’ own department or division or in other units and be available to provide expert advice upon request. These people are able to help with the preparation of REC applications and could point out common concerns as well as forms of words that generally didn’t raise unnecessary problems with the ethics committee.
The Challenge of Ethical Approval for International Studies
Biobanks are intended to facilitate cross-institutional and international research studies by enabling the sharing of samples across organizational and national boundaries. However, the focus group participants repeatedly emphasised the practical constraints that the oversight of biobanks was imposing on this aspect of their work. As noted above, standardised consent conditions present a significant challenge even within the context of a single biobank operating within a limited geographical scope. When the samples in the biobank are potentially available for sharing across national boundaries, the issues associated with the lack of common standards become even more important.
This can cause problems both for the export of tissue samples (where groups not affiliated with the biobank would need to provide more information to the REC about what they intended to do with the samples they received) and the use of samples from overseas biobanks. The logic of ‘delegated trust,’ implicit in the governance of biobanks means that if one group of researchers receives ethical approval to receive samples from one biobank, then, if this ethical approval translates to the other biobank, the research group should be able to obtain samples from that other biobank as well.
However, this model of delegated trust is not always complete and one participant noted that they were ‘a bit worried about my colleagues getting samples from countries where this is not so much enforced this ethical thing’ (FG2 549). That is, they wondered how the trials and consent processes were set up elsewhere. As one noted, ‘So I’m a patient, I’m donating my samples, it’s an ORB consent form; it’s kind of a contract between me and the ORB. If the ORB then pass my samples to x, y, z labs, as a patient can I influence what x, y z do, or can I just influence what happens at the biobank level?’ (FG3 451).
Concerns about the efficacy may be raised by the local research ethics committee or, in some cases, by the journals publishing the outcomes of the research study: ‘They would have to have informed consent because if you don’t use ... if you don’t work with people who have informed consent you can’t now publish. But that would be local informed consent, for example in Canada or [in] Spain we would have no knowledge of what that consent was, we would obviously discuss with the people involved this is the project we’d like to do, it’s a collaboration with them, they would say whether their ethics and their opinion covered it. We were asked with one collaboration with ... to prove consent, one of my colleagues was and in this case he had all the consent forms and they could just show the journals and say ‘‘yes we do; it’s not exploiting people somewhere else’’ ’ (FG3 99).
An additional concern with international studies was the possibility of the unauthorised, reidentification of donors. As with national studies, our respondents fell back on the benefits of anonymization: ‘It’s almost becoming easier, does that mean a lot of these genetic things do actually do the experiments in your own country and then just pull this anonymised data because it’s just such a nightmare’ (FG2 565). That is, ‘you do your experiments, we’ll do ours and pull it all together at the end rather than trying to send samples abroad’ (FG2 565).
The problems associated with introducing new analytical techniques and research questions become magnified in this context as well where it is necessary to ‘go back and get more ethics’ (FG4 275). As one researcher explained: ‘If you’re doing a centre, a multi-centre study then not only do you have to amend your ethics, but you have to get everybody else to amend theirs as well, so it can take you 6 months to get your ethics once everybody’s done it and there’s always somebody dragging their feet. So before you can do it and if it’s changing samples ... once you’ve got it set up it’s fine, but when you start off it’s a lot of work to get back’ (FG4 275).
Broad Consent and Future Uses
One of the more controversial features of biobanks is that donated samples might be used for purposes other than those envisaged when the original sample was donated. As such, the consent form signed at the time of donation might not cover all possible future uses. The researchers recognised this aspect of unknown scientific futures, noting that ‘of course we don’t know what would be useful at the moment we collect it, so we just get permission to use it for any possible research’ (FG1 125). No one knows what tests can be done in 20 years’ time, ‘so that’s really the whole point of a generic tissue bank’ (FG1 607). This broad consent is taken to allow researchers ‘to do things I’m not even dreaming [of] today’ (FG1 615). Indeed, this raised an ethical dilemma of its own, in terms of ‘how much information you give to the patient because ... you have to be general, you have to be vague because you don’t know ... what you’re going to be using the tissue for and in what ... way’ (FG1 127). Moreover, additional challenges arise if the patient is being treated for a life-threatening illness. In such circumstances, discussing longitudinal data collection of tissue samples might require explicitly discussing the fact that the patient might not be alive in 5 years’ time.
In practice, however, current ethical oversight does not allow consent forms that give permission for ‘any possible research,’ and in particular, there is normally a need to distinguish DNA-related tests from other forms of research [47]: ‘our ethics are quite broad to a certain extent but if you try ... if you start using a new technique then you do have to ask for an amendment to use that technique on your sample’ (FG4 169). Moreover, when a new study seeks to use samples that were collected using a different form of consent, the REC will take an ethical perspective and they might say: ‘Well, actually we think you need to go back to individuals because it’s outside of the scope of the original consent’, or they might say ‘well, if you can anonymise this information and there’s no harm to individuals, then you can use it for different research purposes’ (FG1 15).
Such decisions, it was felt by focus-group participants, were taken by their REC on the basis of whether the research was both scientifically and ethically good. For example, an ethics panel might reasonably question whether a broad consent for tissue donation really covered a study that sought to explore if there was a genetic link between criminality and certain ethnic groups. In that case, the panels, which include patient representatives, are there ‘on behalf of the person who gave the generic consent’ (FG1 645).
Revocation of Consent
Given the purpose of the overall research project, the interview agenda for the focus groups explicitly included questions that sought the participants’ views on the revocation of consent and, while they acknowledged that officially research subjects can withdraw their consent at any time, it soon became clear that there were subtle practices at play that affected both the likelihood that a potential donor would give a sample and consent for its use by the biobank and whether there was the possibility of revoking that consent.
For example, while all participants of a scientific study are entitled to revoke their consent at any time (including, of course, participants in the focus group), in practice there were very few cases of revocation known to any of the focus-group participants, other than proactively opting out of DNA-related studies. Indeed, 1 participant noted that decisions on whether to give consent or not were frequently made before discussing the consent form: ‘My experience [is that] if someone wants to consent for research, they go and consent. You can sit with them and go over the consent form and give them information, but if the person has already made up their mind way back and ... doesn’t want to consent, I don’t think you are going to convince them at all’ (FG4 175).
The researchers were only able to provide sketchy details about the ‘likely’ process that would be followed should a request to revoke consent be received. This typically would involve the legal office of the hospital trust responsible for the relevant research project(s).
A common perspective on the issue of revocation was encompassed within broader notions of altruism (‘it’s a big gift that they’re giving us to give us their samples’) (FG2 507), whereby ‘once it’s given, it’s given’ (FG1 21). Others recognised that ‘there’s very much a push within healthcare in general to allow people to have a greater say in their clinical care and what happens to their information’ (FG1 23).
One risk of allowing patients to revoke their consent was that ‘you might not have as much planning security as you used to have ... if people keep dropping out half way through, it might make it difficult to execute the trials and stuff’ (FG1 49). The costs of both obtaining consent and managing any potential revocation processes were also raised, with some researchers suggesting that ‘if we asked patients, the majority of them would say “we’d rather get the research done” ’ (FG1 275). Others noted that ‘we are now using so much money just to keep all the legal aspects going that 10 years ago would have been used to help somebody in the lab to work’ (FG1 146).
Other concerns were raised with the possibilities of patients revoking their consent to continue participation in longitudinal studies or of a revoked consent in a small cohort study where ‘the meaning of that could be completely different and that almost then invalidates the other people who participated and potentially destroys years of research and investment’ (FG3 293).
In an interesting parallel to the amount of understanding associated with giving informed consent to use a sample, some participants suggested that while donors have rights, it was unclear if ‘they fully understand the consequences of the decisions they might subsequently make, and I think to give people that amount of power they have to have the understanding to go with it and that I think is difficult’ (FG3 359).
Conclusion
This paper contributes to our understanding of attitudes to the ethical oversight of biobanking by focussing on an underrepresented stakeholder group, namely the medical researchers who use the resources of the biobank in their daily work lives. We found that these researchers understood the requirements for ethical oversight of their work, in the form of RECs and consent forms. Nevertheless, they expressed much frustration with the practical challenges of obtaining ethical approval for the use of samples, with some suggesting that ethical oversight in the UK had gone ‘too far’ and was now actually hampering their work. Therefore they recognised that the system could be improved and come more in line with the way that research was being carried out.
A particular challenge related to the oversight of research across institutions and national boundaries, which is an increasingly important characteristic of biobank longitudinal studies. Here, despite various harmonization initiatives, the lack of interoperability between ethics processes in various countries and institutions was a clear constraint on innovative medical research. The research approval process was made quicker with the help of staff who had been through the process before and if researchers could provide clear evidence that informed consent had been obtained. Therefore, effective ways to demonstrate that informed consent had been obtained could have the effect of making the process of obtaining research approval in cross-jurisdictional collaborations quicker.
In response to the constraints of doing international research, researchers adopted a series of defensive strategies, including seeking even broader consent for possible future uses than perhaps is envisaged in the general discussion of broad consent in the literature. This trend was not something that they felt happy with, but many felt it was the only solution for secondary research because going back for a new consent for a new research use meant going back through the cumbersome research ethics process again. For them, the need to obtain consent was never in question, as they recognised the value of research participation and the altruism involved in participating in research, which was seen as part of a wider trend in healthcare to allow people to have a greater say about what happens to their information [48]. The issue for most researchers was not to do away with consent, but rather to find an expedient way to obtain consent that would be approved by a research ethics committee.
Researchers also recognised the value of oversight bodies and the personal reflection on their work that occurred when applying for REC approval. For example, if researchers are choosing not to provide feedback to participants in their study, this had to be a conscious choice, rather than an implicit consequence of the design of the generic REC application. This mirrors findings in other studies, where although researchers found the ethics approval system frustrating at times, they still considered that it was necessary and beneficial to their research as a means of providing legitimacy and instilling public trust [49].
The area of revocation and withdrawal of consent is widely recognised to be problematic from a legal and ethical perspective and our focus groups provided detailed evidence of researcher concerns in this area. Researchers could provide few examples of participants who had withdrawn from research, though they acknowledged that this was an important right for participants to have. However, focus-group participants also were concerned that if a dynamic consent approach was used, this might hamper their research, which would affect the societal benefits of their studies. They identified a number of detrimental effects that withdrawal might have on research, such as invalidating findings based on cohort studies and potentially destroying years of research and investment. Therefore, researchers were concerned that any right of withdrawal would have to be accompanied with comprehensive information on the effects of doing so.
The dynamic consent model developed through the EnCoRe project has the potential to address the concerns of researchers, who recognised the benefits of being able to go back to participants on a regular basis for new consent. Being able to demonstrate that consent had been obtained when involved in international collaborations would also help to make things easier. However, there was also resistance and concern about the benefits and new challenges that such a model might raise. While researchers found the current research governance structures frustrating they also had concerns about the dynamic consent model. The idea of revocation was particularly challenging for some, as there was concern that this might lead to actions by participants that was detrimental to research and would not be in the public interest. This suggests that the right of withdrawal in the dynamic consent model would have to be accompanied by information about the implications of withdrawing not just for the individual, but the study as a whole. Providing people with information is an important dimension of the dynamic consent model and would be able to address researcher concerns that they felt that participants did not have a sufficient understanding of the nature of research. These focus groups have provided valuable insights into some of the concerns that researchers face and the ethical requirements they must adhere to when navigating the research governance system. On the whole, researchers could see the benefits of the dynamic consent model in improving the current research governance framework, which they find does not deal adequately with the need to obtain a new consent for secondary research purposes. It could also make the oversight process more efficient and effective which would greatly assist collaborative research. However, most of the concerns were around withdrawal and how this could be managed in such as way to enable participants to exercise this right but to do so in a way that did not undermine the research process.
A particular challenge was the tension between anonymization of sample data (more accurately described as pseudonymization of sample data), which was frequently used as a means of addressing concerns with broad consent and the benefits of longitudinal studies that could link samples more closely to the donors. An important element of this problem relates to the way in which the initial consent is typically seen as a paper-based, one-time process where, in practice, consent is given and is only revoked by determined individuals who are able to make their way through the (rarely used) mechanisms for withdrawal. Moreover, typical consent forms don’t accommodate the option for researchers to recontact the patient to discuss expanding their consent or to convey macro-level study results.
The benefits of more intuitive and easy-to-use systems that can improve the communication of information to and from participants over time and can facilitate recontact opportunities throughout the research process were discussed by our researcher participants in this study, to enable more rewarding and sustainable biobanking research. Many of the researchers who took part in our pilot expressed a genuine interest in more nuanced and effective approaches that can be achieved through technological interfaces such as the solutions being developed by EnCoRe.
Acknowledgements
The authors gratefully acknowledge Prof. Peter Dabrock, Prof. Herbert Gottweis and Dr. Andréa Vermeer for their support and organization of the PRIVATE Gen Workshop ‘Privacy and Post-Genomics Medical Research: Challenges, Strategies, Solutions’ supported by the German Federal Ministry of Education and Research (Bundesministerium für Bildung und Forschung, BMBF).
Appendix: Focus group topic guide
References
EnCoRe (http://www.encore-project.info) is an interdisciplinary research project, which has developed an IT system to make giving consent and revoking consent as easy for individuals as turning on and off a tap. The project partners are Hewlett-Packard Laboratories, HW Communications, QinetiQ, the London School of Economics and the University of Oxford. The project runs from June 2008 to May 2012. It receives funding from the UK Government’s Technology Strategy Board, Economic and Social Research Council and Engineering and Physical Sciences Research Council (grant EP/G002541/1).
The Alder Hey organs scandal involved the unauthorised removal, retention, and disposal of human tissue, including children’s organs, during the period 1988 to 1995. During this period organs were retained in more than 2,000 pots containing body parts from around 850 infants. These were later uncovered at Alder Hey Children’s Hospital, Liverpool, during a public inquiry into the organ retention scandal which followed the outrage and grief of many parents upon finding that their children’s organs and tissues had been retained [46].