You find a partially submerged car bobbing in the local harbour.
A big problem
So, you rush out to hire a crane and pay an operator to lift it out of there.
It’s working perfectly. Soon the pesky vehicle will be out of everyone’s way and normal life can resume.
Except…
Except you hadn’t allowed for the weight of the car and water, and you hadn’t factored in the leverage effect of the degree to which the car was below the crane.
WOOSH, ARGH! SPLASH
So now you have both a crane and car blocking the harbour.
And by the way you might now also have a crane operator who is badly injured.
This latest development really isn’t a problem because you rush out to get an even bigger crane to lift out the first crane with the attached pesky car.
All is going well.
Until it isn’t.
Most of these images record a costly sequence of blunders in Galway, Ireland (which apparently was initiated by a parking accident).
At the risk of undermining my metaphor, the last image is a photoshop fake.
For human research ethics, the big problem is that researchers might not be seeing human research ethics as a vital component of doing research well, and researchers might not perceive ethics as their responsibility. A related problem is ensuring the ethical design and conduct of research might not be perceived as also institutional responsibility.
The usual response has been to
grab research ethics review as a governance weapon to be wielded with ardent fervour.
slam submitted applications we perceive as being incomplete or ill-advised,
respond with scores of directive conditions. And
use enforcement and sanctions procedures to punish the researchers who dare to resist.
But in attempting to solve this problem, have we created a worse one?
Reflections on the limitations of our current approach
The international approach was created as a response to some egregious biomedical and psychology scandals. Safeguarding against such incidents has driven our implementation and the continue evolution of the arrangements.
The approach is wedded to a compliance and enforcement way of seeing any problem.
The approach tends to treat the 99+% harshly because of the misbehaviour of <1%.
It also assumes it is possible and helpful to try to direct the complex behaviour of professionals across a vast array of (sub)disciplines, methods, topics, populations and contexts.
Reflecting back on the time, effort and the resources we have expended over the last 60 years, we find ourselves confronted with an uncomfortable question: Does this approach catch the dangerous minority? I would suggest it does not.
Symptoms of the problem
Institutions that almost exclusively focus their human research ethics efforts on ethics review.
The institution’s executive is anxious about institutional risk exposure.
There is a crippling workload for reviewers and administrators.
It is not uncommon to hear researchers ask each other, “Have you done ethics yet?”
Too many researchers outsource their ethical responsibility to the research ethics committee.
There is evidence of unthinking or grudging compliance, avoidance and misrepresentation.
Ethics training at the institution has been reduced to better form filling.
The National Statement (2007 updated 2018)
The National Statement is not the problem, though it has its limitations.
Are we trapped by past practice? “This is the way it has always been done.”
We recognise the challenges and flexibility that are open to us by the framing of the national arrangements (by way of example, the Australian National Statement on Ethical Conduct in Human Research (2007 updated 2018) allows for great diversity in the selection of a consent strategy, even though there are committees that treat an information sheet and consent form as the preferred consent strategy).
Having excellent policies, processes and forms aren’t enough for us to claim we are promoting a culture of ethical and appropriate conduct.
Could we actually be encouraging researchers not to be reflective?Could we actually be making things worse?
But there is a better way: resourcing reflective practice
Stressing personal, rather than outsourced, ethical responsibility.
Encouraging ethical thinking before commencing an ethics application.
Seeing an approval certificate as a step, not the end of thinking about human research ethics.
Continued reflection on: merit and integrity matters (such as non-financial conflicts of interest); beneficence (such as public goods and community harms); respect for persons (such as a communal approach to consent); and justice (such as ensuring an otherwise disenfranchised voice is not lost).
Developing skills to monitor ethical conduct throughout the lifecycle of a project.
Responsibilities continue through analysis, write up, reporting/publication, and management of data/material after a project is completed.
Regarding research ethics as a design, conduct, quality and professional responsibility – not ‘just’ a matter of compliance and bureaucratic concern.
Recognising a reciprocal obligation for those of us who participate in and/or administer research ethics reviews, to approach the review as facilitating research, not policing it.
Strategies to resource reflective practice
Shift institutional ethos – from relying solely on HREC review as a demonstration that the institution is serious about ethical conduct – to an ethical research culture.
Ethics review to advise and inform, rather than to clear/approve/police.
Ethics administrators empowered to work with both researchers and reviewers.
The institution having resource materials rather than rules.
Use of mentors and collegiate advisors, such as a network of Research Ethics Advisors.
An ongoing commitment to upskilling ethics reviewers.
Training, training, training – professional development focused upon principles, strategies to apply the principles to challenges, not just form filling.
Top down (institutional and discipline leaders) and bottom up (HDR and supervisor training, new staff and experienced staff).
Inviting two-way communication – welcoming innovation, praising thoughtful and elegant strategies, inviting feedback, and supporting creative problem solving.
Continuous improvement and striving towards a learning institutional approach, where complaints and negative feedback are opportunities to improve.
No cranes were harmed in the drafting of this post
Health services are often operated by people who strive to improve the way they deliver care. In the public imagination improvements arise from ‘breakthroughs’ such as the discovery of new disease mechanisms and drugs or devices to address these. However, it is not just novel treatments that lead to better outcomes. Sadly, it is not widely recognised that eliminating sub-optimal practices or variations in healthcare practices play a major role in improving clinical outcomes. Indeed, I don’t recall a headline announcing an increase in operational efficiency in any health service as this is hardly exciting news regardless of its value. Funders of healthcare are interested though, and in a report entitled Exploring HealthCare Variation in Australia: Analyses resulting from an OECD Study, published by the Australian Commission on Safety and Quality in Health Care in 2014, the authors stated that:
Unwarranted variation may also mean that scarce health resources are not being put to best use. As countries face increasing pressure on health budgets, there is growing interest in reducing unwarranted variation in order to improve equity of access to appropriate services, the health outcomes of populations, and the value derived from investment in health care.
All consumers of health care should therefore be interested in this and support those working toward improving health services. Unfortunately doing this work is difficult and often unrewarding. The ethical imperative to do this work is also often thwarted by the ‘ethics’ and governance framework that too often encumbers those doing it {Clay-Williams, 2018 #516}. It is also largely left to the NHMRC to fund Health Services Research (HSR) and the Comparative Effectiveness Research (CER) studies that generate evidence to reduce wasteful practice. In contrast, very little funding from health services themselves go to these activities despite them being the direct beneficiaries of the research.[1] Importantly, those engaged in HSR and CER are becoming an increasingly larger proportion of the total medical research endeavour in Australia, and by classification constituted almost one third of NHMRC competitive funding in 2019[2]. This is despite the fact that the studies undertaken often take several years to complete and therefore the number of publications is smaller than for life sciences. For HSR, publication is rarely in the ‘higher impact’ journals, whereas for some CER Trials the outcomes are so profound that they are of international significance and will be published in widely read international journals. Pleasingly this suggests that the criteria for assessment do not necessarily disadvantage such research in terms of competitiveness for funding, but also reflects the fact that clinical trial funding from the NHMRC supports a great deal of CER studies.
Those doing HSR and CER are also often involved in working in health services as clinicians which reduces the amount of time they can devote to academic research and to build a competitive personal research portfolio. The NHMRC has implemented a “Relative to opportunity” process in an attempt to address the almost impossible task of taking into account personal circumstances, but I doubt anyone is truly comfortable in applying it across the breadth of candidates and disciplines. Indeed, it could be argued that it is a surprisingly unscientific and subjective approach to use in schemes that are rewarding the quality of scientific approaches to address major societal issues. In 2019, only 7% of investigator grants went to applicants identified as HSR researchers.
It is difficult to think of what could replace this system across all areas of scientific endeavour but there is a possibility of rethinking how we fund and manage HSR and CER clinical trials. In both types of activity the end points are focussed on providing evidence to inform changes to clinical practice and health service delivery. As such the end users are health care providers and their funders. It would therefore seem much more appropriate that the end users play a much greater role in judging what kind of research should be done as well as the value of the outcomes of existing projects. The problem with this is that Health Services do not have the internal infrastructure and capability to manage research and have no incentive or means to do so.
What is particularly important to reflect on here is that publication metrics and university-based career milestones are largely irrelevant to the health services and arguably should not be the drivers of why the work is done. It would be more appropriate to have a regular employment relationship between the health services researchers and the health services in a manner that does not differ to clinical safety and quality activities. Sadly, health services have not seen the need to invest in this out of their operating budget and one can see why they would not if universities will do so. The problem though is that health service managers inevitably regard them as academic exercises with no direct relevance to routine health practice and when budgets are tight any support rapidly evaporates.
Like other industries that are reliant on R&D, it could be argued that a defined proportion of all health funding should go to HSR and CER that is conducted and run within the health services themselves. In the UK the National Institute for Health Research (NIHR) was established in 2016 with £1 billion to do just this, representing just under 1% of the total National Health Service Budget at the time (£126 billion[3]). The current public expenditure on health in Australia is $81.8 billion AUD, however, the total health spend, including private and personal (out of pocket) expenditure was $185 billion in 2017/2018[4]. The combined NHMRC and MRFF expenditure on all HSR and clinical research is estimated to be around $800million in 2019, less than 0.5% of the total health expenditure. The difference with the UK is that the Australian funding is largely administered through universities and not by the health services as in the UK.
I would propose that researchers engaged in HSR and CER should be employed by the health services themselves and be regarded as intrinsic to the operations of the health service. In addition, I propose that these staff would not publish papers under their own names but instead publish under the health service banners, either singly or as collaborations of organisations. In this way individual career progression would be based upon demonstration of outcomes in the same way that other activities in the health service are evaluated. Staff who do their job well would continue in their employment and be eligible for ongoing employment. Career progression would be through building a demonstrated portfolio of achievement that is attested to by their employers in a similar way that professional references are provided. Success for individuals would therefore be entirely based on being able to show how they had contributed to productive activity within the organisation.
If this major change to operating this type of research was made then it might also change one of the other major barriers that currently exists, the disproportionately burdensome ethical and governance requirements for such ‘research’ which is mostly treated as having the same risk profile as novel interventional studies. The aims of research are stated to differ from those of providing clinical care to patients, and this is at the heart of the Declaration of Helsinki’s ethical principles as well as those known as The Belmont Report published in 1979. In the Australian National Statement on Ethical Conduct in Human Research, such a distinction between clinical care and research is no longer identified and not commented on so explicitly.
For HSR and CER the distinction between what is routine practice and what is research is extremely blurred, particularly within the context of a self-improving/self-learning healthcare system in which a constant cycle of analysis of the current status of clinical activity informs the delivery of healthcare into the future. The robust methodologies employed to do the analysis and to test potential alternative practices aimed at improving care are indistinguishable to those used for researching novel and potentially more risky interventions. However, the risk profile is completely different, particularly where the research involves one or more practices already in widespread use. A more embedded framework that ensures that ethical issues are addressed at a systemic level rather than through the existing ethics and governance system that treats such work as ‘other’ is needed. The 2019 draft Clinical Trial Governance Framework developed by the ACSQHC goes some way toward creating a culture in which this can be delivered although it will likely require significant cultural change at most health services engaged in research.
We need a system that values those doing this work as core employees and is directly vested in the outcomes of the work and their implementation into improved practice as the prime demonstration of productivity. Such a cultural change will provide the drive to streamline the overly burdensome regulatory framework that currently exists. Such a framework would deliver its own efficiency dividends in a positive cycle enabling more of this work to be done and an acceleration in avoiding wasteful practice and generation of data that brings real improvements to people’s lives. It would seem that this is the very definition of an ethical outcome.
Over the last decade, AHRECS has worked with institutions of various types, size and maturity. The project brief often refers to fixing or implementing proportional review. When you drill down, the work item will generally owe its origins to calls from researchers, the research ethics review body, the research office and the institution’s executive. They might all use the words proportional review or perhaps expedited review, but are they really talking about the same thing?
A casual glance at the NHMRC’s annual activity report shows many institutions are conducting reviews outside their HREC. So why are researchers still calling for change?
While cases like the Racist bus driver, Laughing at the Disabled and the Sexual health survey cases do raise questions about the (mis)use of triggers for review outside an HREC, there remains an apparent tension between the meanings that different stake-holders have of proportional review.
For institutions, proportional review is often understood to include:
From the date of application, how long does it take to get a review outcome to the applicants.
that the length of the application form is determined by the specifics of an individual project.
the review process allows the review body to move quickly through ‘low risk’ and ‘low ethical sensitivity’ projects leaving more time for more contentious work.
Within that frame proportional review can be achieved by committees meeting relatively often, everyone using a smart form and a good structuring of the research ethics committee’s business.
But for researchers the term ‘proportional review’ often means:
Not having a form with questions they struggle to relate to their research design.
The review body including members who understand their (sub)discipline and methodology.
The reviewers being aware of relevant good practice
The objective of the review is to provide useful and constructive feedback.
What we understand researchers look for in proportional review arrangements are the following:
A short and tight form where answers can be as expansive as necessary, guided by clear and useful help text.
The reviewers are different from the main committee and include persons with a background in qualitative, participatory and iterative designs that are relevant to their research.
The triggers between different review pathways, e.g. previously reviewed projects, not more than negligible risk, not more than low risk should be clear, predictable, transparent and consistently applied.
Within this frame, a researcher is unlikely to be satisfied with the review of their research by the ordinary committee or the use of the same form as for full HREC reviews. In such circumstances, they are likely to perceive (even if that perception is inaccurate) that they are being forced to use a process intended for higher risk biomedical research.
One of the advantages of a discernibly separate process is that it is easier to stress to reviewers to adopt a proportional approach to their reflections and that an important part of their role is the facilitation of ethically sound research.
“Regulations don’t solve things. Supervision solves things” Wilbur Ross 2015
Dr Gary Allen, Prof. Colin Thomson AM and Prof Mark Israel AHRECS Senior Consultants
HDR supervisors should, and often do, play an important role in the formulation of a candidate’s research ethics review application. If you talk to an experienced busy research ethics committee member, they will tell you they’ve seen too many applications where there wasn’t any indication the supervisor even saw the application prior to its submission.
Many institutions consider the supervisor to be the lead investigator for candidate research. Even those that don’t usually will expect the supervisor to be a key adviser and mentor for the candidate’s passage through the research ethics review process.
A supervisor not taking an active role in a candidate’s review application can reflect a worrying attitude: “I don’t have time to know about research ethics in detail. The candidate should submit what they have, the experts on the committee can tell them what they need to fix and how they want it changed.”
Such thinking is irresponsible and concerning at numerous levels, not least because research ethics is a fundamental component of the quality design of research.
Being able to think and write about ethical challenges is an essential component of the research training of new researchers.
A supervisor or a research school that isn’t systematically engaged in a candidate’s ethical capacity-building and professional development is failing them.
But a not unreasonable question is “What should I be looking for, when I read an HDR candidate’s application?” A reader of the Research Ethics Monthly asked for a tool she could use.
Attached here is the start of such a tool. The Research Ethics Monthly community includes some very experienced research ethics reviewers. It would be deeply appreciated if you could comment on what we have drafted and suggest other elements for the tool.
The senior consultants started AHRECS in 2007. We were looking for a way of responding to requests for advice on research ethics and integrity from the government, health and education sectors read more...
Comments’ rules
We decided to include comment functionality in the Blog because we want to encourage the Research Integrity and Human Research Ethics communities to contribute to public discourse about resourcing and improving practice. read more...
Random image from the AHRECS library. These were all purchased from iStockPhoto. These are images we use in our workshops and Dr Allen used in the GUREM. Less
Before you send your plan please ensure you have read the frequently asked questions. Below are terms and conditions that you must accept before you can submit your research plan.
TERMS AND CONDITIONS – RESEARCHERS
Questions about these terms and conditions and/or the warranties should be directed to gary.allen@ahrecs.com. Allegations of any breaches or disputes should also be directed to gary.allen@ahrecs.com.
This platform is a service to connect researchers with research ethics advisers.
Advisers must not provide advice with regard to a project where they might be involved in the research ethics review of that project or otherwise involved in the governance or conduct of that project.
Advisers must not provide advice with regards to a project where they have a conflict of interest (regardless of whether that conflict might be perceived or actual).
Advisers must act with integrity and honesty when using this platform, including the provision of advice to clients of the platform (researchers).
Advisers must not misappropriate the original ideas, or other intellectual property, shared with them by researchers.
If an adviser is employed by, or has a role with, a research institution then before providing advice an adviser must confirm that they are permitted to do under the policies of that institution.
This platform does not indemnify advisers for their provision of advice to researcher clients. Advisers should confirm the degree to which they are covered by existing policies band seek appropriate advice with regard to the desirability of seeking additional cover.
It is the responsibility of the adviser to pay all the relevant taxes, duties and fees that are required for the monies they receive from the platform for the provision of advice. Advisers are encouraged to seek independent financial advice with regard to their obligations.
Researchers pay AUD50 for basic advice and AUD500 for full advice.
Basic advice would generally be in the form of around a paragraph of text reflecting on what kind of shape the project described by the submitted research design is in and some dot points of the most significant ethical challenges and difficulties.
Full advice would reflect upon all of the significant ethical challenges and difficulties related to the submitted research design. This is likely to include suggested refinements or other changes to the design, the adviser should direct the researcher to relevant academic literature or other resources. The adviser might also usefully comment upon the matters likely to come up during research ethics review.
Payments are initially paid by EFT to Australasian Human Research Ethics Consultancy Services – the current operator of this platform.
The operators will deduct a flat 20% operator overhead from all payments. The balance will be paid by EFT to the relevant adviser.
An adviser must not solicit or otherwise seek payments outside of this platform or seek engagement from a researcher client for work that is provided by this platform.
In communication with researchers, advisers must not use racist, sexist or otherwise derogatory language. Advisers are urged to approach communications in a collegiate, courteous and constructive manner.
Advisers must not bring the Research Ethics Adviser platform to disrepute.
Failure to observe these conditions may result in the suspension or termination of the adviser’s user account.
Warranties
The operators of this platform offer no warranties to users with regard to the uninterrupted operation of this service.
The operators of the Research Ethics Adviser platform make every effort to keep the platform free of computer viruses and malicious software but the operators provide no warranties that the platform is free of viruses and malicious software. Users are urged to use an effective internet security software and keep it up to date.
Transfers of payments to advisers will be commenced as promptly as possible and an email confirming the commencement of a transfer will be sent to the relevant adviser. The actual commencement of a transfer and the time a transfer takes can be dependent on a number of factors, such as interruptions to the online services of the relevant financial institutions. Consequently no warranty is provided as to the actual time between the receipt of the payment by the operators and the adviser’s receipt of the payment.
The terms and conditions that researchers must accept make it clear that the advice they receive is not a research ethics review and, though it is independent and experience advice, no guarantee is offered that there will be no difficulties when the work is submitted for research ethics review. The operators of the Research Ethics Adviser platform offer no professional indemnity or other cover to Advisers. Advisers are encouraged to arrange their own professional indemnity and other insurance cover, Advisers with institutional cover are urged to check that cover does include advice provided via the Research Ethics Adviser platform.
The operators of the platform will not share the name/email address of users with third parties.
The operators of the platform will maintain the confidentiality of adviser and researcher users of the platform. Aggregate or otherwise de-identified information about the operation of this platform (including the provision of advice to researchers) might be used to-
improve the operation and functionality of this platform
prepare a business case for seeking funding for the development of the full platform
Symptoms of the problem
