NHMRC Public Consultations

Skip Navigation and go to Content
Visit NHMRC website

Section 3 (Chapters 3.1 & 3.5), Glossary and Revisions to Section 5 National Statement on Ethical Conduct in Human Research, 2007 submission

ID: 
26
Personal Details
First Name: 
Ainsley
Last Name: 
Newson
Specific Comments
Comments: 
2. Chapter 3.1

Over-arching comment on Chapter 3.1

The changes introduced in this draft take a pragmatic approach to research, and recognise that much research is now multi-and inter-disciplinary. I applaud the new approach, and hope it is also welcomed by the research community as a whole. It is particularly pleasing to see qualitative methods mentioned throughout the text. While debate continues about the exceptionalist nature of genomics research, for now it seems prudent to continue to include content specific to this topic.

In the following, I both suggest edits to certain formulations of words; and make a few more substantive suggestions. I have arranged my comments by Chapter/Element.

 

Introduction to 3.1

  • 3rd paragraph (Page 6 of draft): change 'will have' to 'will involve' as HREC review can't 'have' anything?
  • 3rd paragraph (p6): change 'acquaint themselves' to 'seek independent review about' (or add latter text)?
  • 4th paragraph (p6): some examples may help here. I note that some, but not all, parts of the draft include examples. While adding too many examples risks disproportionately lengthening the NS, they can aid clarity. A range of examples from different fields of research could be used to avoid implying that one research area has more issues than another.

 

Element 1:

  • The use of 'Themes' here may be confusing, as themes are an output of some kinds of research, not something always to be formulated in advance?

 

Element 3:

  • 3.1.25: using "obtaining consent from" is preferable to "consent of" when referencing participants.

 

Element 4:

Data Identifiability (Introduction to Element 4)

  • Noting the query regarding consistency with Privacy Legislation, to me the key question is to ask what purpose(s) the Privacy Act serves in human research and what issues may arise if the National Statement continues to include language that is not consistent with this legislation. Will it lead to difficulties in HREC decisions, or confusion for researchers? Is the terminology in the Privacy Act appropriate for the research context?
  • A compromise may also be possible, whereby the current NS language is maintained, but a footnote or table qualifying how this relates to the Privacy Act is included.

Rest of Element

  • 3.1.41: With respect to changing identifiability details, should a qualifier be added here to the extent that identifiability should not be changed if it will materially affect the interpretation of research results?
  • 3.1.63: should ‘researcher’ be ‘researchers’?

Element 5:

  • Introduction to Element, 2nd paragraph: mention is made here of the risks in returning results to participants. However this does not include scientific risks, for example if the result(s) turns out not to be accurate or its meaning changes with further advances in knowledge
  • 3.1.68(c): should brief mention be made here of the support that will be made available to the participant to enable them to reach a decision? (e.g. study counsellor in the case of genomic research)
  • 3.1.68(e): with respect to family members, it may be worth giving some thought to whether this should take place before or after validation (a point I return to further below). There are papers in the literature that discuss family communication of research results in a genetics context in that the communication occurred but the original finding in the proband was then not replicated under clinical conditions – the research finding was not ‘true’ and family members were needlessly informed of their potential to have the same mutation
  • 3.1.69: The s95AA guidelines mentioned later in the draft should be mentioned here too; as while this section does not pertain directly to genomic research it may be interpreted as such. 
3. Chapter 3.5

Over-arching comment on Chapter 3.5

I welcome the update to the previous genetics chapter. This field has developed rapidly since this chapter was introduced. I have also previously commented that the previous iteration had a rather negative framing of genetic research, labouring risks while ignoring many of the positive outcomes this field has, and will continue to, give rise to.

This draft Chapter is helpful overall. It takes a moderate position, and is appropriately reflective of international debate in this area. I do worry that from a researcher perspective, there will be a need to flick backwards and forwards a lot, including to a previous chapter for the outline of an ethically defensible plan (can it be replicated here?).

One aspect of the chapter that may need refinement is whether strictly somatic genetic research will come under its ambit? (e.g. tumour genomics). Quite a lot of research in that space does not have hereditary implications and if it comes under the ambit of this chapter, that could mean significant departures from current research practice. Clarification as to scope would be welcome.

More could be made of considerations as to uncertainty in results of genomic research. I have uploaded a recent paper for which I am co-author, which considers ethics and uncertainty in genomics. It is written for a clinical context, but may be of interest.

 

Chapter introduction

  • 1st paragraph: add ‘their’ before family members
  • 3rd paragraph: I’m not sure what ‘unchanging’ means here? DNA is not necessarily static, and the interpretation of genetic information is very dynamic. The dynamic nature of genomic information could be emphasised in other parts of this chapter too.
  • 4th paragraph: a comment could be added here that genomic information may change over time, including to be not quite as strongly predictive as initially thought (for example, a gene may be less penetrant in a healthy individual than in an individual with a relevant family history)
  • 5th paragraph: change ‘insurance’ to ‘certain insurance products’ as not all insurance is affected by genomic information
  • 5th paragraph: instead of ‘previously unknown’, state ‘misattributed biological’ as this is more consistent with the relevant literature
  • Final paragraph: there is quite a bit of discussion occurring in the genomics ‘community’ currently regarding slippage between research and clinical practice. The NS contains a definition of research, but some mention of this debate and any advice for researchers who may be uncertain whether they are undertaking research or not would be a valuable addition

 

Element 1:

  • 3.5.1: I am not sure what is meant by “uses sequenced information”?
  • 3.5.2: is this something specific to genomic research, or true of all research? If the latter, then I suggest it belong in the revised Chapter 3.1

 

Element 2:

  • 3.5.6: Some further nuance regarding this statement would be good. What happens, for example, if a family member fails to contact relatives or is reticent to do so? I have attached a paper I published in 2005 on so-called ‘direct contact’ that offers some procedural ethical advice in this regard

 

Element 3:

  • 3.5.11: As it is mentioned elsewhere in this Chapter (eg 3.5.38) that participants cannot expect an ongoing relationship with researchers following the conclusion of the relevant project, this Clause could also stipulate that participants should be provided with advice about where to seek updates to their information
  • 3.5.17: Is there a reason why it is not stipulated that researchers could periodically remind participants (who have made either choice) of the nature of that choice and to ensure they still agree with it?

 

Element 4:

  • 3.5.24 and 3.5.25: would these be better the other way around?
  • Footnote 5, p29: This is an important point that is liable to misinterpretation. However, this Chapter does not currently discuss the point at which a research result may become a clinical result. Is it clinical information at the point of validation? When a referral is made? Is a participant informed when their information becomes clinical? Are they told in advance that their information may be classed as ‘clinical’? This could have s95AA (or state equivalent) implications, and could also have knock-on effects, e.g. if research findings are eventually excluded from underwriting for certain insurance products.

 

Element 5:

  • 3.5.30: an example of ‘must’ would be useful here; and a clarification as to whether something must be returned even if there is no consent
  • 3.5.31: I support this statement and am pleased to see it included. Advocates of returning raw data are adopting a ‘maximising’ approach to autonomy, which overlooks many of the nuances. Research purporting to show support for returning data is usually quantitative and has not included much by way of deliberative reflection and opportunity for critical engagement with the issues. Thus at least at this point in time, this is a prudent approach.
  • 3.5.32 should be joined with 3.5.39.17 or at least cross-refer to it?
  • 3.5.33 and decision tree: The decision tree needs to be better adapted for the research context. For example, the word ‘patient’ is used throughout – but genomic research can involve healthy individuals too
  • 3.5.34: Can a study genetic counsellor be included here too? For healthy individuals, the clause may imply going back to a GP; which will have resource implications. And the GP workforce may not yet be completely genomics ready
  • 3.5.36: Also for consideration here is the framing of the information for relatives so as to allow them a real choice as to whether to engage further
  • 3.5.37: how does this relate to 3.5.17? They seem similar?

 

Ethically defensible plan guidance:

  • 3.5.39.5: Plus details of how the information could change over time and that the relationship with the research team will end when the project concludes?
  • Step 1(a): subject to consent?
  • 3.5.39.11: noting above point about when validated findings may become subject to s95AA guidelines (or state equivalent)?
  • 3.5.39.17: If results are to be validated in advance (if I am interpreting this correctly?), their potential status as clinical information needs to be clarified. It should also be made clear to participants that this could occur without them being asked to give further specific consent. Further, I worry that this is a departure from standard practice in clinical genomics, where testing would never take place without specific counselling and appropriate time for reflection on the specific test being discussed, or the specific rationale for testing. Undertaking clinical validation of findings without specific consent could be said to be at odds with standard and well-entrenched practice in medical practice areas such as clinical genetics. Whilst I recognise there is a risk that a person may experience anxiety through being informed of a research finding that is then not validated, the harm from having a test without specific consent could be greater. A research participant may not remember such a detail about the consent process and if a result is entered into a medical record (which may happen once the validation takes place) other effects could result. The right to refuse to receive a specific clinically validated result must be protected
  • 3.5.39.18: this should be made absolutely clear in a PIS. An alternative would be for researchers to reconfirm choices for all participants, or a random sample in the case of bigger studies – not just those who have refused to receive results
Supporting attachments

Page reviewed: 10 July, 2018