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: 
7
This submission reflects the views of
Organisation Name: 
General Ethical Issues S-C of Alfred Hospital EC
Personal Details
Specific Comments
Comments: 
1. Introduction to Section 3

No comments.

2. Chapter 3.1

While the intention of combining the three chapters of the current National Statement (3.1, 3.2 & 3.3) into one chapter that covers all types of research is in theory a helpful and more inclusive approach, in practice this has resulted in more abstract and esoteric guidance that can be difficult to understand.  In reading through the revised chapter, it was often necessary to refer back to the current National Statement to try and work out the meaning of a point/clause.

  • Some concepts need to be illustrated with specific examples, rather than removing examples so as not to restrict the types of research the guidance applies to.  For example, 3.1.5 “In their assessment of research quality, reviewers must recognise that thematic and other forms of generalizability may be more relevant than statistical generalizability.”  The corresponding clause in the current National Statement (3.1.6) gave the useful example of data saturation.

  • Some concepts are so broad and general that they offer little in the way of guidance: e.g. 3.1.10 “The potential participant pool for a project should align with the objectives and theoretical basis of the research.”  This either states the obvious or doesn’t make it clear what the issue is.

  • Some points could be more simply worded: e.g. the ‘Key Questions for Element 2: Recruitment’ (p.10)

    • “Will the potential participant pool be screened?”  Why not just “Will potential participants be screened?”

    • “What is the impact of the relationship between researchers and potential participants on recruitment?”  This assumes that there is a relationship.

    • “How will the recruitment strategy facilitate obtaining the consent of participants?”  Is this intended to make researchers and HRECs question or justify the particular strategy chosen (rather than just explaining how participants will be recruited)?

  • The Consultation Document (p.2) suggests that respondents “may wish to consider whether the guidance related to interventional research, particularly clinical trials of unapproved therapeutic substances, devices or biological, is sufficient”.  In response to this, our comment is that for less experienced researchers and reviewers, the current Chapter 3.3 is more helpful because it offers more specific guidance.

Other suggestions:

3.1.24 refers to providing “drafts of recruitment materials” to the reviewers.  As the final versions of participant facing documents need to be approved and cited on approval documentation, the advice to provide drafts is not helpful and suggests that HRECs etc don’t need to see the exact wording.

 

3. Chapter 3.5

Introduction:

  • 3.5.4 (p.27): [“In addition to participants in genomic research identified as index cases (probands), relatives of these individuals who provide information or biospecimens for genomic research become participants in the research in their own right.  Therefore, researchers should be aware of the possibility of de facto recruitment of relatives by virtue of association with a family member who has been recruited.”]  
    It needs to be clearer what the point is. 
    Would “involvement”, which suggests incidental or indirect inclusion, be a better term than “recruitment”, which implies active inclusion?

Element 1:

  • 3.5.1 (p.27): We suggest including the underlined words:
    “Genomic research that uses sequenced information should be designed with attention to what is necessary to achieve the aims of the research and to ensure that ethical issues that arise from activity outside the intended scope of the research are considered and perhaps minimised by, for example, developing a list of genes that are excluded from analysis.“
    The reason for this suggested qualification is that 3.5.1 doesn’t otherwise cover whole genome sequencing, the whole point of which is to find out about ‘unknowns’.

Element 2:

  • 3.5.4 (p.27): [“In addition to participants in genomic research identified as index cases (probands), relatives of these individuals who provide information or biospecimens for genomic research become participants in the research in their own right.  Therefore, researchers should be aware of the possibility of de facto recruitment of relatives by virtue of association with a family member who has been recruited.”]  
    It needs to be clearer what the point is. 
    Would “involvement”, which suggests incidental or indirect inclusion, be a better term than “recruitment”, which implies active inclusion?

Element 3:

  • 3.5.10 (p.28): Should (g) say “identifiable information” – i.e. “whether identifiable information generated by the research will be shared with other research groups.”?

  • The list of points to be considered when developing an appropriate consent process is useful but it is unclear whether they need to be incorporated into a PICF.   If so, a genomic research specific PICF template would be very valuable, and could be included as an appendix.

  • Has consideration been given to whether these points are (or will be) matched by questions in the HREA (new Human Research Ethics Application form)?  They should align.

  • 3.5.15 (p.28): It is not explained why an opt-out approach should not be used in genomic research.  Our view is that this exclusion is too sweeping.  It would be better to have guidelines for the HREC and leave the form of consent/inclusion to case-by-case review.  There is a good argument for an opt-out data collection in some circumstances and in those circumstances an opt-out approach would meet community standards.

Element 4:

  • 3.5.20 (p.29): This clause is inconsistent with clause 3.1.44 in Chapter 3.1 which says that “…biospecimens and, in some cases, the data associated with them should always be regarded as, in principle, re-identifiable.”

  • 3.5.22 (p.29): [“Researchers should include in informational material comment on the potential for re-identifiability of the information generated by or used in the research. “]
    The wording of this clause is cumbersome, making the meaning difficult to grasp.

  • 3.5.25 (p.29): [“Genomic information can sometimes be misused to stigmatise people or to discriminate against them unfairly. Researchers should therefore take special care to protect the privacy of participants. “] This clause is too vague and it is unclear what the point is.  It may also be at odds with making research findings available.

Element 5:

  • 3.5.30 (p.30): Again, this puts the onus back onto research groups and HRECs to decide which findings must, may, or should not be returned.  Some examples of sources for guidance should be referred to here or footnoted; e.g. the American College of Genetics publishes and continually updates a list of reportable findings.

  • 3.5.36 (p.30): We endorse the requirement that the communication of relevant results to relatives should be done by the appropriate clinical service, or (in the absence of such a service) the participant’s clinician in consultation with the research team.  However, we believe that the researchers should not be obliged to search for relatives if the participant has died.  We recommend that it be made clear in this section that the obligations of the researchers are (should be) limited to the participant.

  • Decision Tree for the management of findings in genomic research and health care (p.31)

    1. The diagram is helpful  and gives HRECs a crucial role in deciding how (the protocol) such decisions are to be made, by whom and in what context.  (However, guidance for HRECs in making these decisions, to ensure a consistent approach nationally, is lacking in the National Statement’s revised chapter.)

    2. In the ‘research’ pathway of the submission tree, it should refer to the ‘participant’ not the ‘patient’ (e.g. “Did the patient consent…”.)

    3. There is a step missing.  If there is a ‘no’ answer to the step “Does the Protocol include criteria and a process for the return of findings, including secondary and/or incidental findings” the next step is to “Consult the HREC re: establishment of a process for review of findings, consultation with clinicians and criteria for contact with research participants”.  This assumes that the participant must be, and/or has already consented to being, informed of pertinent findings.  If the Protocol does not contain the necessary process and criteria to guide people, when it does permit the return of findings, this should be rectified and any process should include asking for participants’ consent to the return of pertinent findings.

  • Under 3.5.39.5 – Step 1: Determination of Whether Findings Will Be Returned (p.32)
    The “relevant factors to be considered to determine whether findings must, may or should not be returned” should also include consideration of participant’s wishes about whether or not to be advised of pertinent findings.

  • 3.5.39.19: It is recommended that the statement about respecting participants’ wishes not to receive information, or routinely confirm this later, also include a caveat that there may be exceptional circumstances, such as a new treatment becoming available, which could warrant re-visiting with the HREC decisions not to contact.  This possibility could be mentioned in participant information and a range of options/contact scenarios included in the consent section.  As suggested above, providing a genomic participant information and consent form template with recommended wording would be a very useful appendix to this chapter. 

General comments:

  • Our main concerns are:

    1. The guidelines leave too much latitude to research groups and HRECs to make decisions rather than setting the standard.  This is likely to lead to inconsistent approaches nationally, especially given the potential for onerous requirements on returning results.

    2. The chapter does not sufficiently cover a key area of concern and risk, namely the data security of genomic databanks.  HRECs need to review security arrangements.  Could ISO/IEC 27001 Information security management standards be referred to?

    3. The guidelines do not address the potential for genetic discrimination under current Australian insurance regulations.  This is an issue that cannot be ignored as it affects both research and clinical practice.

Given the complex nature of genomic research and related ethical issues, should there be requirement that genomic research with the potential to generate findings of significance (intended or incidental) be reviewed by specialised HRECs/HRECs that are accredited to review such research?

4. Section 5
  • 5.1.39 and 5.1.40: Guidance as what the insurance and indemnity requirements are when the Sponsor is not a commercial entity in the case of investigator-initiated studies would be beneficial.

  • Chapter 5.2: Does this section also apply to the review undertaken by an RGO of an accepting application reviewed and approved  by a certified HREC?

  • 5.2.7: Is it reasonable to expect that research proposals are written in lay language?

  • 5.2.18: “Capitation payments” is a term that many researchers are unfamiliar with.  It would be helpful to include an example.

  • Chapter 5.5, under “Discontinuation or suspension of research” (p.52), in some situations stopping a study might not be in the best interests of, or might adversely impact on, existing participants.  Other sanctions should be considered, to elicit full compliance.

  • 5.5.3: For consistency, we recommend the inclusion and definition of device adverse events, e.g. Unanticipated Serious Adverse Device Effect (USADE).

  • 5.5.9 (d): Amend to, “or may indicate the need for amendments to the trial protocol and/or Participant Information & Consent Form;”.


We would also like to express specific support for the following additions to Section 5:

  • 5.2.11: The new points (a) and (b), about what should be disclosed to the review body, are useful additions to the clause about disclosure of COI.

  • 5.5.2: We support the addition of wording to clarify that monitoring arrangements should be commensurate with the risk, size and complexity of the research.

5. Glossary

The term “innovative intervention” [3.1.2 (b)] should be defined in the Glossary.

General Comments
Comments: 

Terminology relating to data identifiability

The Consultation document (p.2) asks about use of terminology relating to data identifiability and whether it might be advisable for the National Statement to use language that is consistent with privacy legislation.

We agree that the term “de-identified”, in particular, creates difficulties as it is used in legislation and also in other ‘regulatory’ contexts such as journals’ publication and data deposition policies, but is not used (and its use is implicitly discouraged – see p.16) in the National Statement.

Assuming that the terms used in the National Statement and the privacy legislation are interchangeable, it seems unnecessary and possibly counter-productive that different terms are used to define the same things. 

The view of lawyer members of the Ethics Committee is that if one considers the National Statement to be a quasi-legislative instrument it should use terms that are consistent with the Privacy Act and not use terms that are similar to, but different from, those contained in the Act. Where the term is defined in the Privacy Act the National Statement should use exactly the same term and to do otherwise will only lead to confusion and endless debates about the meaning of the term. The argument will be that the National Statement has used a different form of words from the Privacy Act therefore those drafting the Statement must have had a different meaning in mind when in fact that is not necessarily the case.  It could also be argued that the definition in the Privacy Act being legislation approved by Parliament overrides the National Statement which is only quasi-legislative. It is best to avoid arguments of statutory interpretation and use consistent language.


Data deposition/data sharing

A number of journals, organisations and regulatory bodies (e.g. NIH) require the data underlying research publications to be deposited in a publicly accessible repository, or otherwise make data available for future research use.  This needs to be addressed in the revised National Statement.

Page reviewed: 10 July, 2018