From Inside Ethics Review Boards

This blog post is part of the symposium “The Rooms We Wanted: What a Year of Decolonisation Seminars Revealed about a Law School”, reflecting on the Decolonisation Seminar Series at the University of Edinburgh.

Dr Morshed Mannan and Dr Rebecca Tapscott spoke at “Decolonising Research Ethics,” the second event of the Decolonial Research and Methodologies Seminar Series, focussing on institutional ethics processes and individual research design. They both write here from inside the system they examine – Dr Mannan as director of research ethics and integrity at Edinburgh Law School, Dr Tapscott as a social scientist who has studied ethics review as a form of regulation. Their contributions here in their own words, focus on the requirement of the signed consent form in empirical research.

Dr Mannan traces where the requirement comes from, the laws and funding rules that make written consent the easiest option. He then shows where it can go wrong – a signature can put a Rohingya refugee at risk rather than protect them. Dr Tapscott looks at the system behind it. Ethics review was built for biomedical research and moved into the social sciences mostly unchanged, and she asks whether it can even recognise these harms.

Dr Morshed Mannan:

Earlier this year, Bhavinee Singh and Matilde Masetti Placci of the Decolonisation Working Group at Edinburgh Law School kindly invited Dr. Rebecca Tapscott and I to speak at an event on decolonising research ethics. I was asked to reflect on my experience as director of research ethics and integrity at Edinburgh Law School as well my experience of navigating ethics processes as a legal researcher. My remarks below reflect both perspectives, but I write this in a personal capacity.

The Preference for Written Informed Consent

Procedurally speaking, research ethics applications at leading law schools in the UK come with a number of set defaults. One of these defaults is obtaining voluntary written, informed consent of research participants. In her response that follows, Dr. Tapscott powerfully explains the limitations of informed consent and research ethics reviews generally, particularly for the reflexive social sciences given its origins in biomedical research. Laws, policies, and administrative regulations have had an important role in constructing the expectation that ethics review committees will assess whether research ethics applications are gathering informed consent from human participants.

For instance, nineteen US federal agencies have adopted a common set of standards known as the ‘Common Rule’ which, among other things, sets out general requirements for obtaining informed consent from human subjects in the US Code of Federal Regulations (see 45 CFR 690.116 for the regulations used by the National Science Foundation, a major funder of social and behavioural sciences). Institutional review boards are required to examine if participants are appropriately informed and that informed consent is documented, regardless of where the research is conducted. While there are categories of research that are exempt from the Common Rule and institutional review boards can waive the requirement of informed consent for research that “involves no more than a minimal risk”, the Common Rule, adherence to which is a requirement for projects receiving federal funding, has contributed to making the gathering of informed consent a default practice across the medical, social, and behavioural sciences in the United States. The Common Rule is one of the principal reasons why we now have the standard requirement for consent in the US. In the UK, beyond biomedical research, (counter-) terrorism research, and research involving individuals with mental incapacity, all of which are subject to legislative requirements, expectations around voluntary and informed consent are shaped by the ethical codes of professional and scholarly bodies, funders, and academic institutions (e.g., British Sociological Association Guidelines on Ethical Research). One of the most recent and important changes in the regulation of research is data privacy regulation.

In recent years, data protection laws, most notably the General Data Protection Regulation (GDPR), have also shaped research ethics application forms in the UK, from requiring researchers to submit data management plans to incorporating questions regarding the respect of data protection rights (e.g., the right of participants as data subjects to withdraw their data). These considerations, in turn, affect the drafting of participation information sheets, consent forms, interview protocols, and other research-related documents. Voluntary informed consent, along with ‘tasks in the public interest’ and ‘legitimate interests’, are three of the common legal bases for processing the personal data of human participants by various types of research organisations (article 6(1)(a), article 6(1)(e), article 6(1)(f), GDPR). It is easy to confuse consent for processing personal data and consent for research, as they both can apply to the processing of information regarding an identifiable living individual, but as a predecessor of mine pointed out, these are distinct topics of consent, and consent may not always be the most appropriate lawful bases for data processing when carrying out scientific research.

Our ethics review form at Edinburgh Law School also has default requirements that are shaped by some of the above rules and ethical codes. We typically ask researchers to prepare written consent forms, participant information sheets and data management plans, enquire if the ‘public task’ ground will be used for lawful data processing, and seek reassurance that data will be kept confidential and securely stored. Departing from these default requirements is possible but requires a persuasive explanation. This places an even greater burden on researchers navigating complex ethics processes, as they may be asked to strengthen their explanations and amend their submitted documents. As a consequence, it may appear dauting to conduct novel research or learn a new research method that would necessitate using an alternative form of consent.

Designing Ethics Around Participants, Not Projects

During our event within the Decolonial Research and Methodologies Seminar Series, we discussed the practical challenges and security risks that can arise from participants recording or signing their names. This concern is also reflected in recent scholarship. Dzhengiz and colleagues, for example, note how in “politically unstable environments, participants fear being recorded or signing documents due to potential repercussions” (p. 1064). This can be particularly significant for vulnerable research participants like refugees, who are in contexts where giving consent to data collection is already a regular feature of their lives.

Hossain and colleagues at the Centre for Peace and Justice, BRAC University, who I am working with on a model refugee law and refugee compact for Bangladesh, have written about being troubled by how informed consent is navigated during biometric data collection in Rohingya refugee camps. The Rohingya are an ethnic minority who have, for decades, been persecuted by the Myanmar authorities. Since 2017, when persecutory violence broke out in the Rohingya-majority Rakhine State, over 750,000 Rohingya have fled to seek refuge in neighbouring Bangladesh.

Given Myanmar’s continued surveillance of the Rohingya population in the country, the Rohingya refugees don’t want their data shared with Myanmar authorities. However, as reported by Human Rights Watch, organisations that collected data for research or humanitarian purposes, whether with and without their consent, have shared this data with the government for repatriation eligibility assessments. The refugees were concerned about the adverse implications of not consenting. While they received formal reassurances that there wouldn’t be any, they were informally told that not consenting to biometric registration would impede their access to rations, opportunity to work in the camps, and potentially returning to Myanmar someday. As they were never fully satisfied that their information would be treated correctly, none of the Rohingya that Hossain and colleagues spoke to during their fieldwork felt that their ejajot (informed consent) had genuinely been taken. At the same time, their attitude appeared pragmatic; that giving their ejajot to data collection was only one of their concerns in the midst of a large-scale humanitarian crisis.

The processing of refugee data, in general, is used in a number of ways that are beyond refugees’ control: for aid disbursement, the construction of official statistics, funding appeals, and political negotiations. This reality means independent researchers encounter real obstacles to doing research, with refugees being wary of what they are consenting to when participating in a research project led by a host country national (e.g., on reproductive health) and camp management being concerned that researchers’ work would upset official narratives.

Changing Procedural Ethics from Within

So, how do we address the procedural and substantive deficiencies in informed consent, especially when it is still a default expectation of academic institutions, funding bodies and other stakeholders? I think one approach is to make it easier for researchers to alter or depart from these defaults. At Edinburgh Law School, we aim to approach this process as a structured dialogue. Researchers initiate this dialogue, reviewers make comments and surface concerns, and researchers have the space to respond and address these points. This can seem daunting and takes time, and institutional review boards have a role in making this process less burdensome. For instance, it isn’t sufficient to write in research ethics application forms that alternative forms of consent can be used, or consent can be dispensed entirely, if an adequate explanation is given. It is necessary for boards to show how explanations can be given in a manner that convinces reviewers that participants will not be harmed or put at risk. At Edinburgh Law School, we have an annotated ethics review form and a primer that is intended to help researchers weigh different considerations in preparing their ethics review applications and have begun putting together resources that show how scholars in a variety of disciplines have navigated ethical challenges.

There is still work to be done: these supporting documents can continue to be updated and reviewers can receive further training on emerging ethical challenges in different research domains (e.g., the use of generative AI). PGRs and ECRs, as well as reviewers, would find attending Decolonial Theory and Methods seminars helpful, as it exposes them to epistemologies, conceptions of justice, research methods, and contexts that they may not be familiar with. This will not only help in better understanding the opportunities and risks of such research but also ensure that research ethics processes are sensitive to contextual ethics systems and local realities, that ethical requirements placed on a researcher don’t inadvertently disrespect or harm participants and distort data collection, or preclude certain types of research entirely.

Dr Rebecca Tapscott:

In the world’s most research productive universities until recently based exclusively in the global metropole of the US, the UK, Canada, and Australia, we too often treat ethics review as self-evident. It is viewed as a “minimum bar”, the only system that, while perhaps imperfect, protects the moral legitimacy of academic research and prevents ethical malpractice. But especially in the social sciences, this assumption should be revisited. It is a partial and limiting understanding, both conceptionally and empirically.

Social Science and Research Ethics

Conceptually, a vast and growing scholarship highlights the limitations of ethics review for the reflexive social sciences. Sociologists, historians, political scientists, lawyers, scholars of business and management, education, and others have all documented in varied ways how ethics review can miss the key ethical challenges faced in their research, while imposing requirements that undercut the very values they purport to promote. Signed informed consent is an oft-referred to example for a variety of reasons: respondents can experience it as a protection for the researcher rather than for themselves; as evidence of research participation, it can compromise confidentiality and put respondents at risk; and it treats consent as one-and-done, rather than evolving and relational. Scholars also critique understandings of vulnerability, risk, and of the realities of conducting research in politically charged and changing contexts. These issues are each linked to a disjunct, in which processes of ethics review see the researcher as autonomous from the subject of study, while reflexive social sciences see the researcher as embedded in the field of study, such that research shapes and constitutes its very object.

This means that the myriad challenges documented by social scientists are not ad hoc bad experiences of a technical nature. Rather they are underlying and foundational to ethics review and its understanding of research, including social sciences, as “scientistic”. In this view, the researcher is autonomous from their object of study, making it conceptually possible to assess risks and benefits. But as noted, this does not hold for reflexive social sciences, fields which understand the researcher as embedded in the world they study. Understanding research as reflexive in this way also means that the very principles that underpin ethics review are socially constructed; in a non-trivial sense, one person’s benefit may be another person’s harm.

Ethics review is not tooled to grapple with this; instead, it draws on language, analogies, and assumptions of bioethics, understanding the purpose of research as preserving embodied human health and longevity. When ethics review is extended to incorporate research beyond biomedical and clinical research, it frequently does so by adopting more generic language rather than grappling with these more fundamental questions.

These insights suggest that ethics review is better conceptualised not as a minimum bar that prevents unethical research, but a regulatory system that understands research, and its potential risks and benefits through a positivist, incremental, and value-neutral lens. When ethics review encounters research, these are the elements that it best locates and engages with. In turn, this can cause researchers to reformulate their research in these terms to make it palatable to the regulatory system.

Empirically, in social science fields, ethics review is also not self-evident in its suitability or even availability. While almost all countries worldwide require ethics review for clinical and biomedical research, comparatively few have extended these requirements to the social sciences (in 2020, the number was 21%, with a third of those applying to health-related research). Ethics review also requires resources; it is notoriously time consuming and slow. Ethics review has also been found to reproduce the priorities and ethos of the institution within which it functions, with institutional incentives coming to shape interpretations of open-ended principles. This means that from a very practical standpoint, many social science researchers outside globally leading research universities lack access to suitable systems of ethics review. Imposing it as a prerequisite to publish or to secure funding risks reinforcing the already unequal structures that govern who can speak authoritatively about the world.

These conceptual and empirical observations about ethics review are baked into its institutional design, which uses a committee form as a mechanism to give interpretive meaning to open-ended ethical principles (typically, beneficence and non-malfeasance, justice, and respect for persons). This structural flexibility allows ethics review to be adapted to wide ranging socio-cultural contexts, and to accommodate diverse methods and disciplines at a surface level. But a key part of their functioning rests on agreement over underlying assumptions about what research is and what (and who) it is for. As a result, our models of ethics review are limited for the reflexive social sciences due to institutional design.

An Uneven Research Landscape

The resultant costs are uneven, falling more heavily on those with limited resources, time, and professional standing. For example, in some countries, researchers must pay for ethics review. On one hand, this is defensible, under-resourced institutions in global majority countries may be asked to provide “local” review for foreign sponsored research; there is a compelling argument that local institutions should not foot the bill. However, when applied to all research including unfunded and low-cost social science studies as well as those conducted by domestic academics lacking sponsorship, the cost of ethics review can become a barrier. Where this prevents domestic researchers from doing (or publishing) primary research, or where it stifles entire areas of low-risk research, this should be cause for concern. Other standards of research integrity and ethics, such as norms surrounding data storage, demand access to sometimes costly or impractical technologies.

To be sure, harms of research are real and widespread. Many are mundane, built into the very fabric of a research ecosystem that prizes outputs and impact; others can be severe, with the potential to people’s lives and livelihoods at risk. As a tool to guard against these ills, ethics review requires significant and ongoing interpretive work to identify the “right” ethical questions in social science research, and to avoid imposing requirements that create new ethical issues. It is therefore important to approach ethics review with a critical lens, analyzing its functioning, role, and implications in the context of the institutions within which it is embedded. This allows us to understand ethics review as one of many institutions that regulates knowledge production, shaping ideas about what research is legitimate, and constituting an important gatekeeper to secure funding and to publish findings for a global audience.