Opinion Policy

Research ethics: are we minimizing harm or maximizing bureaucracy?

A phenomenon called “ethics creep” is discouraging researchers rather than protecting study participants.

Karen Robson & Reana Maier

October 08, 2018

Posted in

In My Opinion

Read this article in French 9 Comments

Researchers working with human subjects in North America and beyond are very familiar with ethics protocols required by institutions of higher education, protocols rightly put in place to minimize harm to research participants.

In Canada, individual higher education institutions have ethical jurisdiction over the research conducted within their walls and by their employees, although these operations are guided by the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (TCPS). First published in 1998 by the three main federal research funding agencies, the TCPS requires that all university research involving human subjects be approved by a research ethics board (REB) and outlines the principles to be upheld in assessing the ethical merits of an application, though no standardized process of application or evaluation is given. It is understandable that universities, following the TCPS, want to put steps in place to minimize the harm researchers may cause to potential participants.

In recent years, however, the Canadian ethics process seems to have become more of an exercise in bureaucracy than a reasonable examination of the harm posed by research, and we fear this process will prevent actual research from occurring. We have several experiences dealing with research ethics boards at various postsecondary institutions to give us hours of anecdotes to pore over. Our most recent journey was sparked by a successful research grant wherein we proposed to study students who had transferred from the Ontario university system to the Ontario colleges system. We intended to conduct qualitative interviews with 15 to 20 people who had made this transition and we were well aware that our university and the funding agency would require ethics clearance before releasing the funds.

We completed the ethics protocol at our home institution, which was a dense form of 13 pages, covering everything from researcher contact information to data destruction plans. It is worth noting that our project was very straightforward, allowing us to skip certain sections of the application; it did not involve Aboriginal, Inuit or Métis communities, minors, alternative authorized consent, or people in vulnerable circumstances, and did not touch on particularly sensitive topics or illegal activity. Our methodology did not include deception, secondary use of identifiable data (such as health records), or creation of or modification to a database.

Nevertheless, our completed application was 26 pages, including an appendix comprised of all supporting documentation. The research ethics review committee did not have any fundamental objections to our project, but still responded with a dozen suggestions and minor concerns. After addressing each of the points raised and resubmitting the application, the project was finally approved and we could move on with our data collection. Or so we thought.

Our next step was recruitment. We began by advertising the study with friends and colleagues by word of mouth and social media, all perfectly acceptable methods under our ethics clearance. We also contacted various colleges nearby our university and in Toronto, provided our ethics clearance certificate, and asked if they could advertise our study. This is where things got interesting. While one college was happy to advertise our study immediately with recruitment posters, every other college (a total of eight were contacted) indicated that we would have to fill out their institutional ethics protocol. In other words, if we recruited students from their institution by way of social media, it was perfectly within our ethics protocol, but if they assisted us in recruitment by advertising our study on their campuses in any way (i.e. putting up a poster, sending out an email), they required another full ethics application and their own review.

We were discouraged by this setback given how labour-intensive the ethics application was to put together and the time it took to get it approved, but we found the Multi-College Ethics Research Board to which all of the proposed colleges were members. In theory, a single application would give us approval from multiple colleges at once. Except it did not. Once this application was filled in and approved (after making more minor changes and resubmitting it), we were issued with a recommendation letter. We then had to send another full application to each college individually, along with the multi-college recommendation letter, our own institutional ethics certificate, and all the supporting documents.

Each college would decide if we met the criteria for accessing students. In four cases, this was enough to get clearance, but one college required further minor changes and resubmission of the application, while two more would not even consider our ethics application unless both the principal investigator and co-investigator – an associate professor (and Ontario Research Chair holder) and a postdoctoral fellow, respectively – completed an online tutorial on research ethics and sent certificates verifying completion.

In the end, it took over five months and nine application submissions (12 if you count the resubmissions to incorporate REB feedback) to complete the ethics process and have our recruitment materials displayed at all of the participating colleges. Five months of ethics review in order to do interviews with 15 to 20 adults on an uncontroversial subject.

Canadian sociologist Kevin Haggerty called much of what we have just described as “ethics creep,” whereby “the regulatory structure of the ethics bureaucracy is expanding outward, colonizing new groups, practices, and institutions, while at the same time intensifying the regulation of practices deemed to fall within its official ambit.” Dr. Haggerty made this observation nearly 15 years ago, and we argue that in the intervening years ethics creep has become an ethics sprawl.

Adding to this frustration is the knowledge that research outside of academia has not experienced this kind of ethics sprawl. As education researchers, we work closely with school districts that collect data on students often, some even administering surveys to their students. However, as soon as third-party researchers want to collect data on the same students, an ethics protocol is required – certainly by the higher education institution to which they were affiliated and quite possibly by the school district.

We are obviously not arguing against the existence of ethics protocols or REBs, but we believe that ethics sprawl is discouraging researchers rather than protecting participants. The fetishization of rules and bureaucratic process in ethics review and a blanket worst-case scenario approach is a drain on researchers’ time and resources in return for – what? Do we have any evidence that this level of procedural minutiae is providing improved protection of research participants or preventing unethical research? We might want to consider taking a page from our colleagues south of the border: the National Science Foundation has, as of 2017, abolished the need for institutional research board ethical approval on all projects deemed “low risk.”

Karen Robson is an associate professor of sociology and holder of the Ontario Research Chair in Educational Achievement and At-Risk Youth at McMaster University; Reana Maier is a postdoctoral research fellow in sociology, also at McMaster. She received her doctorate in education from the University of Cambridge.

Karen Robson & Reana Maier

9 Comments

It depends on each university’s policies. In Canada, some universities have very strict policies for conducting research on human and animals. However, some universities don’t have. Rather, some departments in some universities in Canada even don’t know Tri-Council Policy. I experienced that the department chair even let his friends to come to my classroom to conduct research without any ethical approval, and without noticing me beforehand. I was shocked. When I talked to the chair about this inappropriate behaviors , he even said “I don’t think it needs any ethical approval. The researchers just ask students to fill up 20 pages questionnaires and do some interviews.” I was speechless. I doubt whether those chairs in some universities have any knowledge about research.

In Australia approval by an ethics board accredited by its analogue of the tri councils is accepted by all universities, colleges and hospitals.

I wonder if the researchers would have felt the same way had they been applying to university REBs? Having worked for a university/hospital REB I can say that a hospital REB would not only have required REB review but the researcher would also have to find a Locally Responsible Investigator.

Colleges receive numerous requests to access Faculty and students for research purposes. We’re seen as a big pool of participants just waiting to be studied. REB review is necessary in order to prevent certain members of our community from being over targeted as research participants, e.g. Indigenous peoples, people with mental health issues, people with disabilities, immigrants and refugees. REB review is also necessary to avoid overlap between mandatory college surveys and research studies.

Canada has an accreditation program for human research that is rooted in the TCPS 2 and other normative documents. It is Orion Human Research Accreditation (Orion HRA). We are currently accrediting human research protection programs of Canadian organizations, including REBs. Among the important elements of our accreditation program is the work we do with organizations to help them optimize their resources and increase efficiencies. There is no reason for a research advertisement to take more than 72 hours to review. Contact us for more information http://www.orion-hra.org, jp******@or*******.org. We’re here to promote Canadian excellence in research through accreditation.

A brave and important opinion piece. I sense that the vast majority of researchers agree wholeheartedly with this lamentation of ethics sprawl, yet none will write comments, as the subject is too toxic. Ethics review is much like institutional risk assessment, just with harm as the main criteria. Thus, rather than trained and expert actuarial assessment of what the likely harm threshold is in a proposed research, the assessment looks much more a “where’s Waldo” of harm. If you squint in just the right way, maybe you can spot the potential harm that can come through the research. If someone told me that their proposed use of pencils for student surveys was flagged for the possibility of giving splinters to the participants, and thus it was recommended that only ergonomically-proven non-toxic pens are to be used, I would hardly register as strange.
I would love to see a section on the harm caused by not doing the research. Then weigh the expected harm thresholds. Now that would be something!

As the writers acknowledge, this is an old charge with which REBs are familiar. The review system can be improved to be sure, and efforts of REBs to reciprocally acknowledge the expertise of other boards are afoot. This can streamline submissions when it works well. I think that most who have laboured in the field of ethics review are sympathetic to the frustrations of our researchers (and many of us are researchers too). The problem, however, is more complex than the article suggests. Boards are responsible for studies that fall within their jurisdiction and they need to use existing mechanisms to ensure that they are exercising that responsibility appropriately. I have been around long enough to remember a time when little to no review was required for much non-biomedical research in some Canadian institutions and that could be deeply problematic too. What counts as low risk is not always an uncontroversial matter, and even low risk research can squander time and resources – and erode participant good will – if it is not well done. I think we can find common ground when our research review boards include members who bring social scientific expertise to the table. Low risk boards with that expertise may streamline processes too. It is not too much to ask, however, that all researchers complete the TCPS2 tutorial or that we have some arm’s length review of even low risk research. A common vocabulary and an awareness that we all have partial perspectives (and conflicts of interest and commitment) is a good place to start if we want to do this work better.

Few scholars seem to realize that research-ethics review is derived from the medical framework of ethics in research. As such, we can speak of the colonization of research ethics by the field of medicine.Unaware of this process, researchers in the social sciences are now using terms that have principally belonged to medical research, such as Protocol, Investigator,Research-Study, etc It is interesting to note that the only social-science term adopted by medical researchers is “research participant,” not “human subjects.” Should we not decolonize research ethics?

Given the amount of research in the social sciences and humanities that continues to be conducted I see no empirical evidence of researchers in Canada being discouraged from doing research. All evidence points the other way.

It is a shame that their experience of a harmonized review was so long – but I would note over the 20+ years at UBC the speed, timing, and response of our BREB has improved markedly.

The biggest problems I have noted (from both serving on our local ethics review board and through submitting ethics applications) are when researchers themselves over imagine the various risks involved (or claim they have no risks).

While researchers have for years mastered the form (but at times), not the content, of ethical research – that is nothing new. I was among many who have written on this problem (see article linked in webpage URL below)

It is common in my department for grad students to wait 18 months to two years before receiving even just a response from the REB, let alone receive approval. Keeping in mind that Masters programs are typically 2 years in duration and PhD’s are 4 years in duration, there seems to be something unethical about such a prolonged ethical review process, which can in turn delay one’s progression through the program and entry to the workforce. Given that the majority of grad students are earning at or below the poverty, there are ethical considerations around prolonging their duration of study by way of inefficient ethical review processes.

Once approval is achieved, one faces additional challenges, such as inability to recruit through other institutions without first completing their ethical review process (which can only be achieved once one receives ethical approval through one’s institution of study) and costly ethical review processes, such as those at hospitals, which frequently require submission of 15 to 20 copies of their multi-page applications. Such requirements create barriers for less well-funded researchers, thereby influencing what gets studied and by whom. Less cumbersome, more efficient processes would reduce the costs associated with the ethical application process, making research more accessible to those with less funding.

Given that Canada has a standardized ethical research framework, could we not streamline the process, as Gavin Moodie suggests with Australia as an example. For instance, establishing reciprocal agreements between institutions which have REB’s that adhere to the TCPS standards so that researchers can simply submit their certificate of approval by their institution’s REB, along with the application package submitted to their REB to another TCPS abiding institution.

Reply to