22.48 A related concern, raised by the PMCI and Dr Graeme Suthers, involves the use of family genetic information provided by one individual in the diagnosis or treatment of another genetically related family member who is attending the same clinical genetics service. Dr Suthers, the Head of the Familial Cancer Unit at the South Australian Clinical Genetics Service, provided the following example of such a situation:
Both Brendon and Bradley have been tested by me. As far as they know, they have no living relatives who would be at risk of having inherited this mutation. However, a long lost cousin, Claude, is seen by a doctor in another department of my hospital. Claude says that he had a relative called Bradley whose brother had early-onset bowel cancer, but he cannot remember Brendon’s name. The further medical care of Claude would be greatly enhanced if the doctor seeing him could find out details of Brendon, Bradley, and the genetic testing that is now available. Should he ask me? Should the doctor have access to my database?
22.49 Dr Suthers noted that this issue often arises within the South Australian Clinical Genetics Service, which has a number of discrete, yet related, units sharing a common database for reasons of cost and to assist in managing familial disorders.
It is very common for one unit to have data about one end of a family, and another unit to have relevant information about another branch of the family. Should this information be shared?
22.50 Dr Suthers also noted constraints on the disclosure of such information to the treating doctors of other family members. Using the facts described above, Dr Suthers noted that:
If Claude’s doctor was not part of our genetics service, he would not have access to our database, and I would not be able to release any information to him about Bradley and Brendon. He would not necessarily know that I had this information. In practice, we are the main repository of familial cancer information in the State, and Claude’s doctor would probably call me. I could tell him non-identified information about the family and recommend that Claude be referred to our Service for genetic counselling and testing. But—as noted above—when a family has a limited number of affected people, it is impossible to avoid identifying the person with a familial mutation.
22.51 Under the Privacy Act, use or disclosure of health information for a secondary purpose is permitted if the secondary purpose is directly related to the primary purpose of collection and the individual would reasonably expect the organisation to use or disclose the information for the secondary purpose or where the individual has consented to the use or disclosure.
22.52 The requirement that a secondary purpose be directly related and within the reasonable expectations of the individual appears to restrict the use or disclosure of genetic information collected from one patient in the diagnosis and treatment of another.
22.53 However, there may be no reason why this cannot be addressed by routinely seeking consent from patients to the use or disclosure of their family genetic information for the diagnosis and treatment of other patients of the same clinic or hospital. Even broader consent could be sought to the use or disclosure of family genetic information in the treatment or diagnosis of other family members, including use by or disclosure to their treating doctors. Alternatively, in the situation described by Dr Suthers directly above, the clinic has the option of later seeking consent to a particular disclosure.
22.54 Dr Finlay Macrae, Head of Colorectal Medicine and Genetics at the Royal Melbourne Hospital, raised related issues concerning the application of privacy legislation to the disclosure of family genetic information in the course of genetic counselling. He stated that genetic counselling is impeded by the need to ensure that information on an accumulating pedigree is not shared with the family member being interviewed, in case the already collected information includes information ‘provided confidentially by others in the family’. He noted that while, in theory, consent might be obtained for the disclosure of family genetic information to other family members
there may still be information about other [live family] members provided indirectly for the pedigree, and the status of consent for this disclosure is never clear, and potentially could lead to privacy difficulties.
22.55 Dr Macrae observed that:
Working pedigrees rarely include, on the pedigree, the source of the information (except if death certified or cancer register retrieved), so the usual practice is not to disclose anything other than the immediate family to the person being interviewed. The full impact of the pedigree pattern of inheritance is not then pictorially available to the family member being interviewed, and this deprives the counsellor of one of his/her most cogent educational tools to promote surveillance in the family. Displaying an anonymous pedigree is possible, but leads to confusion for both the counsellor and the family member (potentially) as names act as anchors to understanding the pedigree to the naive witness.
22.56 He suggested that this constraint on the disclosure of genetic information hampers effective counselling and causes more ‘overall harm than good’.
22.57 For a variety of reasons, pedigrees contain information which may include personal information about relatives who have not consented to having their information recorded. These relatives may have died, not been approached for consent, or have denied consent to have the information collected or disclosed. The Cancer Council Victoria Cancer Genetics Advisory Committee highlighted this concern and, in particular, the fact that current state legislation requires consent from a deceased’s legal representative, who is not always easy to identify or locate.
22.58 Some genetics services have developed detailed guidelines on how such information should be dealt with. The South Australian Clinical Genetics Service Familial Cancer Registry has developed Confidentiality and Consent Guidelines which, among other things, provide guidelines on the recording and release of pedigree information. These distinguish between situations where the relative (to whom the information relates): has died; is alive and consent has yet to be sought; is alive and consent has been sought but no response has yet been received; and where consent has been denied. For example, the guidelines state that:
Pending consent of a relative, identifying data may be recorded but not released to another party outside the Familial Cancer Registry and the Familial Cancer Unit; identifying data is not released to other services (eg laboratory services) of the South Australian Familial Cancer Service. Pending consent, confirmation of diagnoses may not be sought from other sources.
Where consent has been denied, the following non-identifying data may be recorded on the database: gender, cancer diagnosis, and age at diagnosis. These data are not released to another party outside the Familial Cancer Registry and the Familial Cancer Unit.
 Peter MacCallum Cancer Institute, Submission G104, 20 February 2002; G Suthers, Submission G026, 30 November 2001.
 G Suthers, Submission G026, 30 November 2001.
 In the Guidelines on Privacy in the Private Health Sector, the Office of the Federal Privacy Commissioner has indicated that the Commissioner would consider a reasonable interpretation of ‘primary purpose’ in the health context to be the ‘main and dominant reason’ a health service provider collects information: Office of the Federal Privacy Commissioner, Guidelines on Privacy in the Private Health Sector (2001), OFPC, Sydney, 12.
Privacy Act 1988 (Cth) NPP 2.1(a).
 Ibid NPP 2.1(b).
 F Macrae, Submission G069, 14 January 2002.
 Ibid. Dr Macrae also noted that while the pedigree can be discussed with the patient (and verified against the patient’s own understanding) without the patient directly viewing the document, this ‘leads to an awkward interview where the counsellor is seen to be a controller of family information which “does not belong to [the counsellor]”’.
 Cancer Council Victoria Cancer Genetics Advisory Committee, Submission G195, 27 November 2002.
Health Records Act 2001 (Vic) ss 3, 95(2).
 G Suthers, Submission G026, 30 November 2001.
 Ibid. The SA guidelines note that where consent is denied identifying details (ie name, date of birth, and address) must be deleted from the person’s record in that family. However, ‘a record of these people must be kept so that we know not to bother them again’.