Deregulating Social Science Research Ethics – Clipping the Wings of IRBs?

The Federal Register is surely not everybody’s bedtime reading. It is where the US Government formally publishes certain official documents, including advance notice of its intention to make rules that implement Executive policies. For social science researchers, one of the most important of these has long been the so-called Common Rule, introduced in 1981 and revised in 1991. This is the legal basis for the whole system of US ethical regulation through Institutional Review Boards. Although nominally confined to US institutions, it has had a much wider reach, since any country involved in scientific collaboration with US scholars has had to adopt a similar regulatory model. The new revisions to the Common Rule, described in a ‘notice of proposed rulemaking’ published in September 2015, are, then, likely to have global implications.

In practice, the present regime’s international impact has been uneven in the social sciences, because few institutions have relevant collaborative relationships with US researchers. Major European countries like France, Italy and Germany have seen very limited regulatory developments, mainly in their health sectors and in a few elite institutions open to US influence. The UK, Ireland and the Nordic countries have, however, been strongly affected by the US model. The extensive deregulation of social science research anticipated by the notice is, then, as much a challenge to their regulators as to those in the USA.

The implementation of the Common Rule has been increasingly criticized. As the history of ethical regulation has become better understood, its origins are seen to lie less in a desire to protect human subjects than to transfer litigation risk from the Federal government to universities and to defuse a political panic over accounts of historic biomedical research misdemeanors – some of which look less scandalous when properly contextualized. The system has been widely criticized, as much by biomedical scientists as by social scientists, for its costs, complexity, and conservatism. Many critics claim that it has created a self-serving bureaucracy and facilitated institutional practices that have less to do with ethics than with liability and reputation management. Its brakes on biomedical innovation have measurable costs in lives lost. There are less measurable costs in the systematic ignorance about social conditions created by the regulators’ resistance to qualitative methods that are difficult to control. Although attempts have been made at Federal level to press IRBs to be more accepting of flexible and emergent research methods and designs, these have had only a limited impact.

The reforms are intended to deliver radical change. They address new issues – biobanks, biosocial and big data research – that have emerged since 1991. However, they seek to do this within existing, or possibly reduced, resources, by drastically excluding large areas of social science research from review. It is important to note that this is exclusion from jurisdiction not exemption, which has always been an option for IRBs. While informed consent requirements, for example, will be strengthened in many areas of biomedical research, most social science research will be allowed to presume consent from participation, provided that it complies with general legal standards for privacy and data protection. The emphasis will shift from paternalism to respecting the autonomy of research participants, including those in ‘vulnerable populations’, to make their own decisions about participation.

Any research involving standardized testing, surveys, interviews, or observations, including audio and video recording, of public behavior, including behavior online, will be able to proceed without further review. Certain types of experimental work will also be excluded, where participants experience ‘benign interventions [which are] brief in duration, harmless, painless, not physically invasive, not likely to have a significant adverse lasting impact…[or to be] offensive or embarrassing.’ These studies may include an element of deception provided the participant agrees to this in advance. Oral history, journalism, biography and historical scholarship about named individuals can proceed solely with reference to the disciplines’ own codes of ethics. The secondary analysis of administrative data, including many health and criminal justice records, even in identifiable form, will be excluded from review provided it complies with data protection legislation. Some research with children would still require review: standardized tests and non-participant observation would be excluded but not surveys, interviews and participant observation.

Before we celebrate, though, it is important to note, with the reform’s authors, that, while the changes remove statutory requirements for IRB review, they do not prevent universities from ‘extra-regulatory’ actions. The determination that research falls into an excluded category will be made with an online decision tool, but universities may choose to have someone other than the principal investigator complete that form. This creates an opening for university review processes to intervene, although the authors hope to discourage that. However, in the absence of a statutory obligation to undertake ethical review, universities will have to justify the allocation of resources for this purpose, particularly with their increased duties in relation to biomedical and biosocial research. This should flush out the extent to which IRBs are being used mainly for managerial rather than ethical purposes. If university leaders want to block a project that might upset donors or provoke Fox News, it will be harder to use ethical review as a fig leaf.

Many problems with the existing system are due less to the regulations than to their implementation. Although the revisions should restrict the opportunities for abuse, they will not necessarily lead to changes in local practice. The IRB system has a well-organized network of administrators, whose jobs depend on an expansive jurisdiction, and interlocks with the organizational politics of senior university management. However creative the Federal response to social science researchers’ complaints, we may not actually see the deregulation that this notice envisages. The legitimacy of the regulatory system may simply be further undermined without the positive liberalization that many, including the reform’s authors, think to be both necessary and desirable.