ACN - 101321555 Australasian Human Research Ethics Consultancy Services Pty Ltd (AHRECS)

Resource Library

Research Ethics MonthlyAbout Us

Australia

Australasian Human Research Ethics Consultancy Services Pty Ltd (AHRECS)

Vigilance versus vigilantism in science: Are ethics no longer important?25

 

Michael James PhD, Senior Researcher, Rheumatology Unit, Royal Adelaide Hospital
Les Cleland AM MBBS MD FRACP, Head of Rheumatology (1982-2015), Royal Adelaide Hospital

In July 2016, the University of Adelaide received an allegation of research misconduct involving the PhD thesis work of a graduate student. We were her supervisors.

We first heard about the allegation not from the University but from a journalist working for ABC TV Australia. It was alleged to have occurred in our laboratory over 15 years ago.

The ABC journalist was in possession of confidential emails between the complainant and the University of Adelaide and the journalist was persistent in attempts to obtain our comments. Although the University process was confidential, it appeared that the complainant was working with the journalist to run a story on unsubstantiated claims before an investigation had commenced.

The claim was made against one of our former students, Dr Maryanne Demasi, who had been awarded her PhD in 2004. Since then, Dr Demasi has worked for the SA Government as a Ministerial advisor and most recently, as a science journalist for ABC TV’s Catalyst program. In 2016 she produced and presented a program covering the scientific debate on the possible health effects of the electromagnetic radiation from Wi-Fi-enabled devices like mobile phones. This attracted criticism from the Telco industry and the ABC imposed a three-month suspension on her duties. Dr Demasi defended her program (Demasi, 2016). However, shortly after her high-profile suspension, the allegation of research misconduct arose.

The complainant’s identity was not known to us but the emails we received from the ABC journalist described the person as a ‘leading Australian scientist’ from the prestigious Walter and Eliza Hall Institute (WEHI). The journalist’s specific mention of WEHI as the employer of the complainant served to give gravitas to the allegations. They appeared to have the imprimatur of WEHI.

The complaint was made in the form of several images of Northern and Western blots with arrows and annotations and alleged that Dr Demasi had duplicated some images in her PhD thesis. The experiments had been done 15-16 years earlier and the thesis was submitted 14 years ago. The topic of her thesis relates to the effect of hypoxia on expression and activity of cyclooxygenase 2 (COX-2), a pro-inflammatory enzyme. There was no context to the complaint such as whether the allegations could have affected the conclusions drawn from the impugned figures or the conclusions of each thesis chapter or the overall conclusions of the thesis. Also, there had been no suggestions of any issue with the research findings in the 15-16 year interim.

The process for managing allegations of research misconduct was specified in the Australian Code for the Responsible Conduct of Research (National Health and Medical Research Council, 2007a). After an initial internal investigation, which took 11 months, the University convened an independent panel, external to the University, to examine the allegations. This was in accord with the Australian Code.

The Chair was the Honourable John Sulan QC, a former Justice of the Supreme Court of SA. There were three other panel members who were all academic experts. The panel took testimony from expert witnesses, from ourselves as the supervisors of Dr Demasi’s PhD studies, and from Dr Demasi. Witnesses, including ourselves, had previously provided written sworn statements and were questioned on these by legal counsel assisting the panel and by Dr Demasi’s legal team.

The hearing was held in Sydney. It commenced in March 2018, approximately 21 months after the University received the allegations, and ran over four days. At the time of our attendance at the hearing, we learned that the complainant had declined a request to attend the hearing to speak to the allegations under the panel’s rules of procedural fairness in place for all witnesses. The complainant chose not to provide a sworn statement presenting the evidence or context for the allegations.

The process placed the entire burden of proof on Dr Demasi. Most of the original documents from the experiments conducted in the years 2001-03 had been discarded years before, especially as our laboratory had closed prior to the building, in which it was housed, being decommissioned for the move to the new Royal Adelaide Hospital. The allegations were based on highly processed images that had been generated on a scanner and computer technology dating back to the 1990s.

In response to three allegations, there had been duplication and this was intended. They represented the ‘baseline’ value for time course experiments and they were intended to indicate there is only one baseline for both the normoxia and hypoxia treatments. The expert witnesses and the panel agreed this was acceptable practice at the time, circa 2002, and did not constitute a breach in any code. The panel found no evidence of duplication for the remainder of the specific allegations where the respondent had denied duplication. The panel ruled that it could not substantiate any of the allegations made by the complainant (University of Adelaide, 2018).

Within 24 hours of the panel findings appearing on the University website, the online Post Publication Peer Review (PPPR) site ‘Retraction Watch’ posted the headline ‘Controversial Australian science journalist admits to duplication in her PhD thesis’. This headline thus began by branding Dr Demasi as controversial and then misrepresented the findings of the panel (Anonymous – Retraction Watch, 2018).

Motive for vigilantism has raised considerable discussion in the PPPR environment. The Editor-in-Chief of the journal Plant Physiology is critical of the anonymous aspect of online comments on the PPPR site, PubPeer, saying, ‘The overwhelming majority of posts on PubPeer are negative and occasionally malicious.’ He further states, ‘What of the bulk of comments posted on PubPeer? These relate to small errors and oversights, not the stuff of misconduct nor likely to arouse any but the most obsessive of temperaments’ (Blatt, 2015).

By contrast, PPPR sites are supported in the article ‘Science Needs Vigilantes’ which argues that the motive for making an accusation is not relevant – ‘are the vigilantes really doing something that requires explanation?’ (Neuroskeptic, 2013).

In the present case, the complainant had made allegations about thesis work conducted 15-17 years earlier for which there had been no previous suspicion of misconduct. If the motive of any science vigilante must be to correct the scientific record, there was no apparent reasonable motive for the complainant to trawl through this old thesis. Further, because the complainant worked with a journalist to run a story on unsubstantiated claims, it is presumed the motive was other than to correct the scientific record. The collaboration with the journalist overrides the requirement of procedural fairness, also described as natural justice, as required by Section 12 of the Australian Code.

In our view, the argument that the motives of vigilantes are irrelevant fails to recognise that their activities can preclude conformity with the standards for investigation of allegations and, consequently, the protections that those standards accord to participants.

The investigations were conducted according to the Australian Code (2007a). This stipulates that research misconduct has several mandatory elements, one of which is that it must have ‘serious consequences’. Therefore, an allegation of research misconduct requires content and context. However, this was absent in the allegations presented to the University by the complainant and to us by the ABC journalist. There was no suggestion that the alleged image manipulation could allow the student to make a claim or conclude anything different from that which was already in the thesis. Nevertheless, the process lasted for 21 months at great financial cost to the accused and at great cost to the health and wellbeing of the accused. It was a process that required no input from the complainant after the initial allegation.

The Code states ‘A person who makes an allegation must also be treated fairly and according to any legislative provisions for whistleblowers during and following investigation of the allegations.’ The standards for taking a complaint seriously are low, presumably to protect the vulnerable whistle blower. However, the standards for refuting even a mischievous complaint are high and onerous. Once the accused is shown to have no case to answer, the mischievous complainant should have a case to answer. The 2007 Australian Code is unbalanced and failed to protect the falsely accused.

The 2018 revision of the Australian Code devotes more space to dealing with a complaint that is made in ‘bad faith or is vexatious’ and states that ‘action to address this with the complainant should be taken under appropriate institutional processes’ (NHMRC 2018). However, such action will necessarily be limited or will be non-existent when the complainant is not an employee of the institution as in the present case.

This revision of the Australian Code and the Guide to Managing and Investigating Potential Breaches of the Code (the Investigation Guide) published in 2018 together provide a manual for dealing with allegations of research misconduct while allowing scope for subjective judgements. The Guide does contain a list of “principles of procedural fairness” that institutions are “expected” to incorporate in their investigation processes. However, for these to clearly address circumstances such as those in the current case, it would benefit greatly from specifying ethics principles to be used in making judgements. Respect for persons, beneficence/non-maleficence, and justice are ethics principles that are adopted across medicine and human research (National Health and Medical Research Council, 2007b). The current case demonstrates unethical behaviour when measured against these principles. It can be argued that a complainant is not obliged to observe these ethics principles when making an allegation under the Australian Code because they are not explicit in the Australian Code. However, such an argument does not pass a reasonable person test nor would it prevent an institution requiring a complainant to observe the fairness principles that are now specified in the Guide.

Research Ethics Monthly invites debate on issues raised by items we publish. However, we will only publish debate about the issues that the items raise and expect that all contributors model ethical and respectful practice.

References

Anonymous (2018) Retraction Watch. May 9. [cited 2018 June 5] Available from: https://retractionwatch.com/2018/05/09/controversial-australian-science-journalist-admits-to-duplication-in-her-phd-thesis/

Blatt M.R. (2015) Vigilante science. Plant Physiology 169: 907-9.

Demasi M. (2016) Sometimes asking questions provides you with answers that may be uncomfortable. Huffington Post. Feb 19. [cited 2018 June 5] Available from: https://www.huffingtonpost.com.au/maryanne-demasi/sometimes-asking-questions-provides-you-with-answers-that-may-be-uncomfortable_b_9267642.html

National Health and Medical Research Council (2007a) The Australian Code for the Responsible Conduct of Research. Available from: https://www.nhmrc.gov.au/guidelines-publications/r39

National Health and Medical Research Council (2007b) The National Statement on Ethical Conduct in Human Research. Available from: https://www.nhmrc.gov.au/guidelines-publications/e72

National Health and Medical Research Council (2018) Guide to Managing and Investigating Potential Breaches of the Australian Code for the Responsible Conduct of Research. Available from: https://www.nhmrc.gov.au/_files_nhmrc/file/publications/17653_nhmrc_-_guide_to_managing_and_investigating_potential_breaches_-_v1-3-accessiblefinal_0.pdf

Neuroskeptic (2013) Science needs vigilantes. Discover magazine. Dec 31. [cited 2018 June 5] Available from: http://blogs.discovermagazine.com/neuroskeptic/2013/12/31/publish-damned/#.WwEm-iDhXct

University of Adelaide (2018) May 9 [cited 2018 June 13] Available from: https://www.adelaide.edu.au/research-services/oreci/integrity/inquiries/docs/summary-panel-report-for-publication-9-may-2018.pdf

Contributors
Michael James, Senior Researcher, RAH – Profile (see below#) | contactmjjames@gmail.com

Les ClelandHead of Rheumatology Unit, RAH – Profile (see below*)

# “Research activity in the Rheumatology Unit at the Royal Adelaide Hospital (RAH) from 1988 – 2016 involved laboratory based studies on lipid inflammatory mediators and their metabolic enzymes and also included clinical trials with omega-3 fatty acids in cardiac and rheumatoid arthritis patients. I was an Adjunct Professor in the Dept of Medicine at the Univ of Adelaide until 2018. I was Chair of the Royal Adelaide Hospital HREC for 14 years from 1995 to 2009 and have been Chair of a Bellberry HREC since 2009.”

* “Research interests have included production and regulation of lipid mediators of inflammation, clinical effects of dietary omega-3 fatty acids, early intensive treatment of rheumatoid arthritis and the role of T cells in  experimentally-induced arthritis. Member of Institute of Medical & Veterinary Science Ethics of Animal Research Committee 1984-2001.”

This post may be cited as:
James M. and Cleland L. (22 June 2018) Vigilance versus vigilantism in science: Are ethics no longer important?. Research Ethics Monthly. Retrieved from: https://ahrecs.com/research-integrity/vigilance-versus-vigilantism-in-science-are-ethics-no-longer-important

Australian Code 2018: What institutions should do next1

 

Gary Allen, Mark Israel and Colin Thomson

At first glance, there is much to be pleased about the new version of the Australian Code that was released on 14th June. A short, clear document that is based upon principles and an overt focus on research culture is a positive move away from the tight rules that threatened researchers and research offices alike for deviation from standards that might not be appropriate or even workable in all contexts.

The 2007 Code was rightly criticized on several grounds. First, weighing a system down with detailed rules burdened the vast majority with unneeded compliance for the recklessness and shady intentions of a very small minority. Second, there was reason to suspect the detailed rules did not stop the ‘bad apples’. Third, those detailed rules probably did not inspire early career researchers to engage with research integrity and embrace and embed better practice into their research activity. Finally, the Code did little to create an overall system able to undertake continuous improvement.

But, before we start to celebrate any improvements, we need to work through what has changed and what institutions and researchers need to do about it. And, then, maybe a quiet celebration might be in order.

Researchers have some fairly basic needs when it comes to research integrity. They need to know what they should do: first, as researchers and research supervisors in order to engage in good practice; second, if they encounter poor practice by another researcher; and, third, if other people complain about their practices.

The 2007 Australian Code offered some help with each of these. In some cases, this ‘help’ was structured as a requirement and over time was found wanting. The 2018 version appreciated that these questions might be basic but that the answers were often complex. The second and third questions are partly answered by the accompanying Guide to Managing and Investigating Potential Breaches of the Code (the Investigation Guide) and we’ll return to this. The answer to the first question is brief.

The Code begins to address responsibilities around research integrity through a set of eight principles that apply to researchers as well as their institutions: honesty; rigour; transparency; fairness; respect; recognition of the rights of Indigenous peoples to be engaged in research; accountability, and promotion of responsible research practices. Explicit recognition of the need to respect the rights of Aboriginal and Torres Strait Islander peoples did not appear in the 2007 version. There are 13 responsibilities specific to institutions. There are 16 responsibilities, specific to researchers, that relate to compliance with legal and ethical responsibilities, require researchers to ensure that they support a responsible culture of research, undertake appropriate training, provide mentoring, use appropriate methodology and reach conclusions that are justified by the results, retain records, disseminate findings, disclose and manage of conflicts of interest, acknowledge research contributions appropriately, participate in peer review and report breaches of research integrity.

In only a few cases might a researcher read these parts of the Code and conclude that the requirements are inappropriate. It would be a little like disagreeing with the Singapore Statement (the one on research integrity, not the recent Trump-Kim output). Mostly, the use of words like ‘appropriate’ within the Code (it appears three times in the Principles, twice in the responsibilities of institutions and five times in responsibilities of researchers) limit the potential for particular responsibilities to be over-generalised from one discipline and inappropriately transferred to others.

There are some exceptions, and some researchers may find it difficult to ‘disseminate research findings responsibly, accurately and broadly’, particularly if they are subject to commercial-in-confidence restrictions or public sector limitations, and we know that there are significant pressures on researchers to shape the list of authors in ways that may have little to do with ‘substantial contribution’.

For researchers, the Code becomes problematic if they go to it seeking advice on how they ought to behave in particular contexts. The answers, whether they were good or bad in the 2007 Code, are no longer there. So, a researcher seeking to discover how to identify and manage a conflict of interest or what criteria ought to determine authorship will need to look elsewhere. And, institutions will need to broker access to this information either by developing it themselves or by pointing to good sectoral advice from professional associations, international bodies such as the Committee for Publication Ethics, or the Guides that the NHMRC has indicated that it will publish.

We are told that the Australian Code Better Practice Guides Working Group will produce guides on authorship and data management towards the end of 2018 (so hopefully at least six months before the deadline of 1 July 2019 for institutions to implement the updated Australian Code). However, we do not know which other guides will be produced, who will contribute to their development nor, in the end, how useful they will be in informing researcher practice. We would hope that the Working Group is well progressed with the further suite if it is to be able to collect feedback and respond to that before that deadline.

There are at least nine areas where attention will be required. We need:

1. A national standard data retention period for research data and materials.

2. Specified requirements about data storage, security, confidentiality and privacy.

3. Specified requirements about the supervision and mentoring of research trainees.

4. A national standard on publication ethics, including such matters as republication of a research output.

5. National criteria to inform whether a contributor to a research project could or should not be listed as an author of a research output.

6. Other national standards on authorship matters.

7. Specified requirements about a conflicts of interest policy.

8. Prompts for research collaborations between institutions.

For each of those policy areas the following matters should be considered:

1. Do our researchers need more than the principle that appears in the 2018 Australian Code?

2. If yes, is there existing material upon which an institution’s guidance material can be based?

3. Who will write, consider and endorse the guidance material at a national or institutional level?

Many institutions will conclude it is prudent to wait until late 2018 to see whether the next two good practice guides are released and discover how much they cover. Even if they do so, institutions will also need to transform these materials into resources that can be used in teaching and learning at the levels of the discipline and do so in a way that builds the commitment to responsible conduct and the ethical imaginations of researchers rather than testing them on their knowledge of compliance matters.

Managing and Investigating Potential Breaches

The Code is accompanied by a Guide to Managing and Investigating Potential Breaches of the Code (the Investigation Guide). The main function of this Guide is to provide a model process for managing and investigating complaints or concerns about research conduct. However, before examining how to adopt that model, institutions need to make several important preliminary decisions.

First, to be consistent with the Code, the Guide states that institutions should promote a culture that fosters and values responsible conduct of research generally and develop, disseminate, implement and review institutional practices that promote adherence to the Code. Both of these will necessitate the identification of existing structures and processes and a thorough assessment to determine any changes that are needed to ensure that they fulfil these responsibilities.

This means that institutions must assess how their processes conform to the principles of procedural fairness and the listed characteristics of such processes. The procedural fairness principles are described as:

  • the hearing rule – the opportunity to be heard
  • the rule against bias – decisionmakers have no personal bias in the outcome
  • ‘the evidence rule – that decisions are based on evidence.

The characteristics require that an institution’s processes are: proportional; fair; impartial; timely; transparent, and confidential. A thorough review, and, where, necessary, revision of current practices will be necessary to show conformity to the Guide.

Second, when planning how to adopt the model, institutions need to consider the legal context as the Guide notes that enterprise bargaining agreements and student disciplinary processes may prevail over the Guide.

Third, the model depends on the identification of six key personnel with distinct functions. Some care needs to be taken to match the designated roles with the appropriate personnel, even if their titles differ from those in the model, in an institution’s research management structure. The six personnel are:

  • a responsible executive officer, who has final responsibility for receiving report and deciding on actions;
  • a designated officer, appointed to receive complaints and oversee their management;
  • an assessment officer or officers, who conduct preliminary assessments of complaints;
  • research integrity advisers, who have knowledge of, and promote adherence to, the Code and offer advice to those with concerns or complaints;
  • research integrity office, staff who are responsible for managing research integrity;
  • review officer, who has responsibility to receive requests for procedural review of an investigation.]

Last, institutions must decide whether to use the term ‘research misconduct’ at all and, if so, what meaning to give to it. Some guidance is offered in a recommended definition of the term but, as noted above, this will need to be considered in the legal contexts of EBAs and student disciplinary arrangements.

Conclusion

The update to the Code provides a welcome opportunity to reflect on a range of key matters to promote responsible research. The use of principles and responsibilities and the style of the document offers a great deal of flexibility that permits institutions to develop their own thoughtful arrangements. However, this freedom and flexibility comes with a reciprocal obligation on institutions to establish arrangements that are in the public interest rather than ‘just’ complying with a detailed rule. We have traded inflexibility for uncertainty; what comes next is up to all of us.

Click here to read about the AHRECS Australian Code 2018 services

The Contributors
Gary Allen, Mark Israel and Colin Thomson – senior consultants AHRECS

This post may be cited as:
Allen G., Israel M. and Thomson C. (21 June 2018) Australian Code 2018: What institutions should do next. Research Ethics Monthly. Retrieved from: https://ahrecs.com/research-integrity/australian-code-2018-what-institutions-should-do-next

We invite debate on issues raised by items we publish. However, we will only publish debate about the issues that the items raise and expect that all contributors model ethical and respectful practice.

Stop centring Western academic ethics: deidentification in social science research – Anna Denejkina0

 

This blog will provide a discussion of issues present in deidentifying marginalised research participants, or research participants who request to be identified, in the publication of qualitative research. As my research is mixed-method (quantitative and multi-method qualitative) it included several data collection techniques and analyses. For this discussion, I will specifically focus on the face-to-face and Skype interviews I conducted with participants in Russia and the United States.

My PhD study investigates intergenerational transmission of combat-related trauma from parent to child, focusing on the Soviet–Afghan war, 1979–89. This research includes interviews with Soviet veterans and family members of veterans; it was these interviews that raised questions of participant erasure and agency. From 12 face-to-face and Skype interview participants, one participant requested complete deidentification; one requested that their real name not be used but their location and other identifying details remain; two participants requested that only their first names be used and their location and other identifying details remain; the eight remaining participants requested that they be fully identified, with some participants sending me photographs of them and their families for inclusion in research publications. Given the social and political sensitivity that persists in Eastern Europe around the discussion of the Soviet invasion into Afghanistan, I had to consider and discuss with participants that requested they be identified the issue of their safety.

My research participants are marginalized participants by virtue of the topic of my research, the Soviet–Afghan war, and the ongoing silencing treatment they’ve received during and following the war by the state:

To take just two examples: in the hope of obscuring the true impact of the war, some local authorities refused to allow special areas in cemeteries to be set apart for the graves of soldiers killed in Afghanistan; while others forbade the cause and place of death to be stated on gravestones or memorial shields. (Aleksievich, Whitby & Whitby 1992, p.5–6)

Given academic broad-stroke standards of deidentifying research participants, we must review the ethics of this practice as it can promote and perpetuate erasure of marginalised participants and the silencing of their voices. Some textbooks on the topic of ethics in the social sciences approach anonymity and deidentification of participants from the angle that anonymity is part of the basic expectations of a research participant, without elaborating that anonymity is not always desirable nor ethical (see for example Ransome 2013), essentially replicating the medical model of human research ethics developed for the regulation of biomedical research in the United States (Dingwall 2016, p.25). Such an approach does not address the issues of presenting anonymity as a status-quo in social research, and makes a sweeping – and a Western academic – generalisation that anonymity is one of the vital assurances researchers must give to their participants to keep within their duty of care (that is, that researchers have at least some obligation to care for their research participants).

This approach to research ethics negates participant agency, particularly those participants that request they be identified in research. Furthermore, forced anonymity can be an act of disrespecting participants (Mattingly 2005, p.455–456) who may have already experienced invisibility and who are then further erased through anonymity by researchers (Scarth & Schafer 2016, p.86); for example, “in some Australian and, in particular, some Indigenous cultures, failing to name sources is both a mark of disrespect and a sign of poor research practice” (Israel, Allen & Thomson 2016, p.296).

As researchers, we must also question if presenting this approach as a vital tenet of social research can become a damaging rule-of-thumb for new researchers who might, therefore, not question the potential undermining of participant agency, and use deidentification unethically as a sweeping regulation within their research without consideration for the individual situations of their research participants. This is part of the issue created by applying a medical model of ethics assessment processes to the social sciences, in which the prevailing interpretation is that deidentification is also required within social research, whereas the reality is that specific agreements between the researcher and the research participant must be honoured.

The ethical dilemma, therefore is: can researchers ethically deidentify participants at the expense of the participants’ agency, potentially perpetuating the historical and symbolic erasure of their voices and experiences? I argue that, based on research design and data collection methods, this decision-making process is an ‘ethics in practice’ and must be approached in context, individually for each study, and for each individual participant.
As scholars, we want to minimise or eradicate harm that might come to our participants through our research. While we think “in advance about how to protect those who are brought into the study” (Tolich 2016, p.30) this must be a continual process throughout our project, in which we “work out the meaning of what constitutes ethical research and human rights in a particular context” (Breckenridge, James & Jops 2016, p.169; also see Ntseane 2009). This is important to note, because protection does not only refer to participants but also to others connected to them. For example, the use of a real name at the request of a participant may expose their family member(s) who were not part of the research.

Consequentialist approaches to ethics suggest that “an action can be considered morally right or obligatory if it will produce the greater possible balance of good over evil” (Israel, 2015: 10; also see Reynolds, 1979). This is an approach we could take to issues around deidentification; however, this also means that researchers must know what is good or bad. In studies like mine, this would mean knowing (or making an attempt, or an assumption to know) what is good or bad for my research participants. This action is infantilising, and places the researcher above the research participant by making the final call ourselves, which is to remove participant agency – if we can assume participants are autonomous during the research consent process, we must also assume that they are autonomous in making decisions with respect to their identification (Said 2016, p.212). Additionally, this action may be culturally insensitive given that Western human research ethics committees follow Western cultural guidelines, centring the dominance of Western academia.

The ethical issues I faced during my PhD research highlight why researchers cannot take a sweeping approach to deidentification in qualitative research – not even for a single study. ‘Ethics in practice’ means that each participant’s situation is analysed individually, and issues around erasure, safety, and their agency weighed against each other to reach a conclusion. I propose that if this conclusion is at odds with the preference of the participant, that it must then be taken back to the participant for further discussion. Not implementing this aspect of ‘ethics in practice’ goes against social science ethics, that we must avoid doing long-term and systemic harm, both of which come through erasure and silencing. We must also remember that “any research project has the potential to further disenfranchise vulnerable groups” (Breckenridge, James & Jops 2016, p.169), and ignoring the wishes of participants regarding their identification due to a Western model of ethics can cause further damage to these groups.

References:
Aleksievich, S., Whitby, J. & Whitby, R. 1992, Zinky Boys: Soviet voices from a forgotten war, Chatto & Windus, London.

Breckenridge, J., James, K. & Jops, P. 2016, ‘Rights, relationship and reciprocity: Ethical research practice with refugee women from Burma and New Delhi, India’, in K. Nakray, M. Alston & K. Whittenbury (eds), Social Sciences Research Ethics for a Globalizing World: Interdisciplinary and Cross-Cultural Perspectives, Routledge, New York, pp. 167–186.

Dingwall, R. 2016, ‘The social costs of ethics regulation’, in W.C. van den Hoonaard & A. Hamilton (eds),The Ethics Rupture, University of Toronto Press, Toronto, pp. 25–42.

Israel, M., Allen, G. & Thomson, C. 2016, ‘Australian research ethics governance: Plotting the demise of the adversarial culture’, in W.C. van der Hoonaard & A. Hamilton (eds),The Ethics Rupture, University of Toronto Press, Toronto, pp. 285–216.

Mattingly, C. 2005, ‘Toward a vulnerable ethics of research practice’, Health: An Inderdisciplinary Journal for the Social Study of Health, Illness and Medicine, vol. 9, no. 4, pp. 453–471.

Ntseane, P.G. 2009, ‘The ethics of the researcher-subject relationship: Experiences from the field’, in D.M. Mertens & P.E. Ginsberg (eds), The Handbook of Social Research Ethics, 1st edn, Sage, Thousand Oaks, pp. 295–307.
Ransome, P. 2013, ‘Social research and professional codes of ethics’, Ethics and Values in Social Research, Palgrave Macmillan, Basingstoke, pp. 24–53.

Said, D.G. 2016, ‘Transforming the lens of vulnerability: Human agency as an ethical consideration in research with refugees’, in K. Nakray, M. Alston & K. Whittenbury (eds),Social Sciences Research Ethics for a Globalizing World: Interdisciplinary and Cross-Cultural Perspectives, Routledge, New York, pp. 208–222.

Scarth, B. & Schafer, C. 2016, ‘Resilient Vulnerabilities: Bereaved Persons Discuss Their Experience of Participating in Thanatology Research’, in M. Tolich (ed.), Qualitative Ethics in Practice, Left Coast Press, Walnut Creek, CA, pp. 85–98.

‘Tolich, M. 2016, ‘Contemporary Ethical Dilemmas in Qualitative Research’, in M. Tolich (ed.), Qualitative Ethics in Practice, Left Coast Press, Walnut Creek, CA, pp. 25–32.

Statement of interest
No interests to declare.

Contributor
Anna Denejkina | Casual Academic and PhD  candidate in the Faculty of Arts and Social Sciences, researching intergenerational trauma transmission UTS | Staff profileAnna.Denejkina@uts.edu.au

This post may be cited as:
Denejkina A. (24 May 2018) Stop centring Western academic ethics: deidentification in social science research. Research Ethics Monthly. Retrieved from: https://ahrecs.com/human-research-ethics/stop-centring-western-academic-ethics-deidentification-in-social-science-research-anna-denejkina

“More what you’d call guidelines”0

 

In a notorious scene from Pirates of the Caribbean: The Curse of the Black Pearl, Captain Barbarossa refers to the Pirate’s Code cynically as ‘what you’d call guidelines’ suggesting that conformity is merely a matter of choice:

Elizabeth: Wait! You have to take me to shore. According to the Code of the Order of the Brethren…

Captain Barbarossa: First, your return to shore was not part of our negotiations nor our agreement so I must do nothing. And secondly, you must be a pirate for the Pirate’s Code to apply and you’re not. And thirdly, the Code is more what you’d call ‘guidelines’ than actual rules. Welcome aboard the Black Pearl, Miss Turner.

Recently, some evidence has emerged that the same observation could be made about another set of guidelines, namely, those relating to the ethics review and conduct of human research in Australia: the National Statement on Ethical Conduct in Human Research issued by the National Health and Medical Research Council, the Australian Research Council and Universities Australia in 2007 and modified to the current version of May 2015. These guidelines set out the principles and processes for ethics review by human research ethics committees (HRECs) and conduct of research in which people are participants. The guidelines also set out requirements for the establishment, membership and operation of HRECs and assign obligations to institutions to see that these are followed. Since 2001, the NHMRC has established and maintained a register on which institutions list their HRECs and agree to operate them according to the National Statement.

Annually, these institutions provide to the NHMRC, on request, reports on the conduct of the HRECs they have established. It is these reports, covering 2014, 2015 and 2016 that provide revealing evidence about the extent to which HRECs and institutions in fact conform to the National Statement.

.

ACTIVITY

‘When an institution has established an HREC, the institution is responsible for ensuring that…

  • review of research proposal is thorough;
  • review processes and procedures are expeditious;
  • the workload of an HREC does not compromise the quality or timeliness of the ethical review;’ (National Statement 5.1.28 (c), (d) & (i))

Some reasons for these guidelines are that ensuring an adequate workload maintains review skills and that avoiding an excessive workload weakens and prolongs reviews.

Evidence:

In 2014, of the 216 HRECs that reported, 11 did not review any proposals and 59 met between 1 to 5 times and 41 HRECs considered not more than 10 proposals.

In 2015, of the 212 HRECs that reported, 10 did not review any proposals and 54 met between 1 to 5 times, and 42 HRECs considered not more than 10 new proposals.

In 2016, of the 210 HRECs that reported, 15 did not review any proposals and 50 met not more than 1 to 5 times and 42 HRECs considered not more than 10 new proposals.

The published data support the conclusion that in each of the last three years about 5% of the HRECs did not review any proposals and about 20% did not review more than 10 new proposals. What would be the effect of such light workloads on the review expertise of HREC members?

.

TRAINING

One other means by which committees can maintain their review skills is to undertake regular continuing education or training. The National Statement contains two relevant guidelines that recognise this:

‘where an institution has established an HREC, the institution is responsible for ensuring that… (b) members undertake: (ii) continuing education;’ (5.1.28 (b)(ii)), and

‘…each member of a review body should: …(c) attend continuing education or training programs in research ethics at least every three years.’ (5.2.3 (c)).

Evidence:

In 2016, 161 of 210 HRECs reported that ‘1 or more’ members had attended training during the year.

In 2015, 185 of the 212 HRECs reported that their institutions made opportunities for training available to members, but only 160 reported that 1 or more members had attended relevant training during the year.

In 2014, although 179 of the 217 HRECs indicted that their responsible institution provided opportunities for members to attend training, only 149 of the committees reported that 1 or more members had attended relevant training during the year.

The data published by the NHMRC does not allow us to identify whether the same HRECs failed to take advantage of training possibilities in different years nor to work out what kinds of institutions were more or less likely to ensure professional development of committee members. Nevertheless, in each of these years, between 30 and 40 HRECs undertook no training at all and the data for those that did could mean that not more than one member attended one training opportunity in that year.

While some HRECs may be investing in the professional development of their members, it is difficult to conclude that the requirements are being taken seriously across the sector.

Indeed, it is doubtful that the sector as a whole is engaging with capacity building of HREC members. Since 2008, the NHMRC has not devoted any resources to training of HRECs, in contrast with its counterparts in the United Kingdom. Canada and the United States. The provision of such professional development as is available has fallen to voluntary national gatherings, such as the Australasian Ethics Network or to commercial providers.

.

MEMBERSHIP

There is a minimum membership for HRECs of eight constituted by specified categories, namely,

  • a chair,
  • at least two community members,
  • a member with experience in counselling or treatment of people,
  • a person who performs a community pastoral role
  • a lawyer
  • at least two researchers. (National Statement 5.1.30)

The specification relates to HREC decision-making, as the guidelines provide that HREC decisions ‘must be informed by an exchange of opinions from each of those who constitute the minimum membership…’

Evidence:

In the 2016 year, 16 of the 210 HRECs reported that they did not have the minimum membership during the year.

In the 2015 year, 42 of the 212 HRECs reported during the year that they made decisions on proposals when there was a vacancy in their membership, conduct that the report noted as being ‘contrary to the National Statement’.

In the 2014 year, 34 of the 216 HRECs reported that they had continued meeting when there was a vacancy in their membership, again noted in the report as being ‘contrary to the National Statement’.

Accordingly, there were numerous decisions – it is impossible to calculate how many – made by HRECs during each of the last three years that lacked the range of input that the guidelines require. Further, as the reports in 2014 and 2015 note, such decision-making is contrary to the National Statement and, in turn, a failure by the responsible institution to fulfil its responsibility to ‘see that any human research for which they are responsible is… ethically reviewed and monitor in accordance with this National Statement.’ (National Statement 5.1.1(b)).

.

GENDER BALANCE

One other guideline about membership provides that, ‘as far as possible, there should be equal numbers of men and women’ (National Statement 5.1.29(a)).

Evidence:

In the 2016 report, it was stated that because ‘It is recognised that this may be difficult to attain’, the ‘NHMRC considered instances in which there was at least an 80:20 gender imbalance as significant and requiring attention.’ Only 5 of the 210 HRECs reported such an imbalance. No data were reported of the number of HRECs in which there was a lesser gender imbalance.

In the 2015 report, the same statements appear and only 3 of the 212 HRECs reported an imbalance of 80:20 or more.

In the 2014 report, that statement does not appear and 24 of the 216 HRECs are reported to have ‘less than a 70:30 gender balance in either direction.’

Accordingly, gender imbalances of anything less than 1 in 5 are not regarded as in need of attention, an interpretation of the guideline that provides very little incentive to correct imbalances that, in many other contexts in 2018, would be regarded as unacceptable. Were the sector taking gender seriously in HREC membership, it would be far better to create more meaningful targets and, if necessary, to phase these in over time. It would also be sensible to track where gender imbalance lies within committee membership and compare that to broader patterns within the host institutions. For example, if the small number of female academics on a committee reflected a broader problem in an institution, the latter may need to be addressed first rather than placing an increased burden on a small number of more senior female academics.

Year Reviewed applications Complaints about research Complaints about review
2007 10777 138 49
2008 21087 96 19
2009 22306 100 11
2010 23696 121 21
2011 25022 n/a n/a
2012 26257 161 19
2013 24882 145 20
2014 20892 226 58
2015 18768 229 34
2016 18039 237 37

.

IMPLICATIONS

Does any of this matter? If so, why: do deficiencies in the processes by which HRECs reach their conclusions and decisions contribute to more of those decisions being inappropriate or unacceptable? Trends in complaints data may be some indication, and these data, available from 2007 to the current year (except 2011), tabulated below, do show some increase in both kinds of complaints. However, it is notoriously difficult to generate reliable conclusions from complaints data.

Implications for NHMRC Administering Institutions

Institutions take responsibility for annual HREC reports and many of those institutions will be administering institutions with NHMRC grants, the conditions for which are contained in a standard Funding Agreement. Clause 24 of that agreement provides that ‘in carrying out this Agreement, the Administering Institution must comply… with… the NHMRC Approved Standards and Guidelines.’ (which are defined to include the National Statement). Clause 30.4 of the same agreement requires Administering Institutions to ‘immediately notify the NHMRC in writing if it ceases fully to comply with… the NHMRC Approved Standard and Guidelines. Such failures are among the grounds on which the NHMRC can suspend or terminate research funding’ (Funding Agreement clause 15).

Should reporting of a non-compliant HREC be treated as such a breach? Given that an institution is required to notify the NHMRC when it ceases ‘fully to comply’, any of the deficiencies recorded above from the last three annual reports of HRECs would appear to be sufficient.

.

Some wider implications and questions

These data raise the question: when is an HREC decision sufficiently defective as not to merit respect or recognition? When there is no input from any one of the minimum members? When the gender balance is lower than 1 in 5 in either direction? When the decision is the only one or only one of 5 that the committee has made in a year? When none of the committee members have attended training in the last three years? When the decision is one of 30 made at the same meeting? When there is no reference in the decision to any one of the four key review criteria: research merit, justice, beneficence or respect?

Will tolerance of nonconforming practice lead to declining support for and recognition of HRECs and, in turn, of ethics review itself? If that recognition declines, will the need for ethics review be questioned? Would recognition that Australian ethics review lacks accountability and conformity to national guidelines threaten the reputation of Australian research, researchers and research institutions?

Doing nothing in the face of this evidence condones the nonconforming practices and risks breeding an indifference to ethics review and, in turn, to seeing it as irrelevant and unnecessary.

However, if a response to these data of deficiency is needed, at what level should that response be made: that of the HRECs, institutions or all the sector stakeholders? In short, who should take responsibility for the reliability of ethics review and how should that responsibility be implemented?

Contributors
Colin Thomson – Senior Consultant, AHRECS | AHRECS biocolin.thomson@ahrecs.com

This post may be cited as:
Thomson C. (2017, 22 March 2018) “More what you’d call guidelines”. Research Ethics Monthly. Retrieved from: https://ahrecs.com/human-research-ethics/more-what-youd-call-guidelines

Page 1 of 41234