NHMRC Public Consultations

Skip Navigation and go to Content
Visit NHMRC website

Draft Policies for NHMRC Committees and Working Groups Developing Guidelines Submission

ID: 
15
Personal Details
First Name: 
Jane
Last Name: 
Johnson
E. Submission
Submission options: 
Section 1 – Policies Document
Section 2 – Key Elements and Flow Chart
Section 1 – Policies Document
1. Are there any gaps in the issues addressed in this document?: 

The document effectively limits itself to a concern with financial and relationship based conflicts of interest, which are the most straightforward to identify. As a consequence, it appeals to remedies which are also quite limited (transparency and recusal). The document pays little attention to within-in role conflicts, where experts may have legitimate interests relating to their multiple roles, with the potential for conflict between these. For example, a specialist clinician with expertise in heart disease may also be a researcher with significant funding from NHMRC. Being part of a guideline-development group builds this person’s reputation, thereby assisting them to secure further funding and giving them an interest in retaining membership of the committee; and such a person has an interest in having their findings incorporated into the guideline, as this will be a very attractive outcome to report to the research funder. This example is not meant to indicate that clinician/researchers should be excluded, but rather to highlight some of the complexities of conflicts of interest, and the problem of limiting consideration to largely external conflicts such as financial ones.

2. Does the document strike the right balance between ensuring adequate expertise is available to be deployed on the committee and creating a framework to ensure conflicts are managed appropriately? : 

Given the webs of interests of many of the relevant professionals, and the limited population base in Australia from which to select ‘experts’, there is a real issue about finding experts without conflicts. One possible response to this problem is to widen the criteria for expertise, and to select younger clinicians and professionals rather than always seeking the most well-known people in a field. This may have some advantages: younger professionals may have fewer established ties with industry or fewer conflicts; they may be more critical of the evidence used to develop guidelines, and more adept at reviewing such evidence; they may be more available for such appointments; and they may be less intimidating for lay and other committee members. Well known experts could be harnessed for review of specific recommendations or to make a presentation to the committee if this was appropriate, in which case their influence would be limited to the specific interaction, rather than being permeated throughout the committee processes. In addition, it may be possible to enlist international experts with their participation facilitated by electronic means such as video-conferencing, if there is a lack of relevant and non-conflicted expertise within Australia

Given the structural nature of many of the interests that may conflict, both within role and external, the remedies of disclosure and recusal are inadequate. Disclosure may paradoxically increase the influence of the disclosing member, as they will be perceived to be honest and trying to be impartial. Experience from the US indicates that, in relation to decisions made by the Food and Drug Agency, for every one committee member with a conflict, there was a 10% greater likelihood that the meeting would favour the drug under review (Lurie et al 2006, cited in Rogers and Ballantyne 2009); this is despite disclosure of the interests. It is beyond the remit of this consultation to instigate systemic changes, but long term changes to the culture of research and medicine is required to loosen at least some of the ties between industry and clinical experts. In the absence of such changes, it may be worth providing committee members with some education about conflicts of interest and bias. In particular, it may be helpful for members to know about different sorts of conflicts, and the tendency for bias (often unconscious rather than intentional) to operate in favour of the person with the interests.

3. Do you have any suggestions for further guidance on matters to be declared?: 

Religious convictions may constitute a conflict of interest under 1.4.1 b where these involve a ‘strong position or prejudice’. And clearly religious views do influence the thinking of individuals on ethical matters. Our concern would be that some people would not recognize these affiliations as conflicts or might be reticent to reveal their religious orientation. If the NHMRC regard such convictions as conflicts (as we do), we think it would be best to explicitly state this as an example in 1.4.2 to avoid any confusion or lack of clarity on the matter.

4. Is there anything that should be declared which has not been included in this section?: 

See 1 above re within role conflicts.

5. Is the amount to trigger a declaration of interests reasonable?: 

With respect to 1.5.2 e, the guidelines state ‘Disclosure is required if these disbursements amount, in total, to $1000 or more per year, or to $3000 or more over the past three years.’ However i) evidence suggests even small gifts can create conflicts of interest (Grant and Iserson 2005), rendering this threshold amount irrelevant; and ii) it is unclear why it is that only amounts received within the last 3 years should be considered germane. We are unaware of any evidence or research which supports this proposition. Any gifts or hospitality have the potential to bias the recipient in favour of the provider of the gift (Dana and Loewenstein 2003).

References

Dana J and Loewenstein G. 2003. A social science perspective on gifts to physicians from industry. JAMA 290 (2): 252-255.

Grant DC and Iserson KV. 2005. Who’s buying lunch: Are gifts to surgeons from industry bad for patients? Thoracic Surgery Clinics 15 (4): 533-542.

Lurie, P., et al. 2006. Financial conflict of interest disclosure and voting patterns at Food and Drug Administration Drug Advisory Committee meetings. JAMA 295(16):1921–88.

Rogers W and Ballantyne A. 2009. Justice in health research: what is the role of evidence-based medicine? Perspectives in Biology and Medicine 52 (2): 188-202.

Section 2 – Key Elements and Flow Chart
1 - Are there any gaps in this section in relation to the guidance for the responsible Senior Officials at the Office of NHMRC to apply their judgement in how to manage any conflicts?: 

We strongly support the appointment of an independent and authoritative chair with no conflicts. In our view, the independence of the chair is more important than their expertise, and lack of expertise may in fact be an advantage in identifying patterns of bias amongst committee members.

We are unaware of any evidence to support the option of “participation with constraint” option. If a conflicted member is present at some or all of the meetings, and excluded solely for limited periods such as while drafting recommendations, it is not clear how this will limit their overall influence upon on the development of the guidelines. Other committee members will have heard their views and formed their own opinions about the credibility of the conflicted person, and it seems likely that physical exclusion will not mitigate against these influences.

2. Is the flow chart an adequate representation of the proposed process?: 

The flowchart shows that members should declare any changes to their potential conflicts of interest at the start of every meeting, but we question whether this allows sufficient time to adequately address new conflicts which may have arisen. Hasty decisions to manage new conflicts may be unduly affected by pragmatic considerations such as the fact that a member is already at the meeting.

Page reviewed: 5 November, 2012