Issues Facing a National Genomic Data Governance Framework in Australia: An Interview Study

The use of genomic testing in both clinical and research applications has become more prevalent with the decreasing costs and increasing cost-effectiveness of genomic testing and analysis techniques. However, ensuring that such data are effectively collecting, storing, sharing and used across clinical, research and population health settings is not guaranteed by lowered testing costs. Encouraging efficient, impactful, and ethically-justifiable uses of human genomic data require appropriate forms of governance. Good data governance approaches enable data access, encourage compliance with national laws and international agreements, mitigate potential harms of data use, promote equity in the use of data, and direct the use of genomics to the public benefit. A national genomic data governance framework is needed to make better use of Australia's genomic datasets in a way that is ethically defensible, is socially valuable, and promotes public trust. This report contributes to the development of a national genomic data governance framework by some of the key Ethical, Legal and Social Issues (ELSI) and technical challenges surrounding genomic data. We present findings of a qualitative interview study with 40 of Australia’s foremost genomic data experts from across clinical, research and consumer advocacy settings. Participants were directly asked about the issues facing the more effective use of genomic data in their work, as well as the opportunities and pitfalls they foresaw in developing a national approach to genomic data governance. Interview data were openly coded to allow for participant’s own understanding and uses of genomic data to shape the research results. Our research found that issues that a national approach to genomic data governance would need to address include: 1. Data Collection and Consent, such as: (i) the collection, governance and consent of data from Aboriginal and Torres Strait Islander peoples; (ii) the question of how informed consent is applied across different genomic data contexts; and (iii) the trialling of dynamic consent in genomic research. 2. Data Storage, such as: (i) data storage costs that include cloud computing costs, research funding limitations, and also impact data retention/deletion practices; (ii) data security issues concerning cloud-based storage and data breaches; (iii) the need to consider how other personal and identifying information are linked to genomic data; (iv) how to appropriately enact data custodianship for communally-managed data and how custodianship is procedurally delegated; (v) data management plans, their cost, and insufficient data management practices and succession planning; (vi) limitations to data linkage posed by the presence of clinical healthcare information; and (vii) the sheer volume of some research data sets. 3. Data Sharing, such as: (i) jurisdictional barriers, including state/territory-based and international barriers to data sharing; (ii) data formatting issues, including national and international standards, interoperability and the impact of formatting decisions on data preservation; (iii) challenges of enabling third party access, including through bespoke data sharing agreements, on-sharing to consumer services, and through using technological solutions to facilitate access; (iv) infrastructural impediments to data sharing, which is divided into research and clinical settings; (v) international data sharing in research; (vi) “siloed data” in the clinical setting; (vii) costs related to sharing data in research and biobanking; (viii) maintaining patient and participant privacy; (ix) economic and academic (dis)incentives for sharing; (x) lack of consistent governance arrangements across organisations; and (xi) international data protection guidelines. 4. Data Use and Outcomes, such as: (i) secondary use of genomic data, including discussion of open data, secondary analysis of clinical data, and the restrictions that cold storage places on secondary use of data; (ii) return of results in biobank and research contexts; (iii) public wariness of genomics perceived by genomic counsellors and consumer advocates; (iv) discrimination and disadvantage in genomic testing broadly; (v) clinical understanding of genomic data and information; and (vi) limitations to clinical care using genomic data and access to genomics-related diagnosis and treatment. To develop a national genomic data governance framework, we recommend that an ethics framework be constructed in order to address the differing needs of genomic data stakeholders. Such an ethics framework should, at minimum, address the following issues identified in this research. Recommendation 1: Learning from the Indigenous Data Sovereignty Movement. A national genomic data governance framework should learn from and engage with the Indigenous Data Sovereignty movement by (a) seeking to incorporate the principle of self-determination of Indigenous peoples into data governance and (b) by reflecting on how these principles may find relevance for marginalised cultural groups within Australia more broadly. Recommendation 2: Equitable Consent. A national genomic data governance framework should develop a concept of equitable consent to address (a) the tension between clinical and research uses of consent, and (b) the use of digital consent approaches, such as dynamic consent, in research. Recommendation 3: Strategic Interoperability. A national genomic data governance framework should offer principles for enabling inter-organisational and inter-state agreement on when and how genomic data can be accessed by third parties. Such principles should aim to foster organisational interoperability through recognising the differential interests of genomic data custodians across different contexts. Recommendation 4: Genomic Data as Public Goods. A national genomic data governance framework should present principles to ensure that uses of genomic data premised on the ‘public good’ or uses for ‘public health’ are both democratically decided, with participation from marginalised social groups who may benefit from greater representation in genomic science and medicine.