Five lessons from the #MyHealthRecord discussion for #PrecisionMedicine
By Dr. Declan Kuch*, Assoc. Prof. Matthew Kearnes*, Dr. Georgia Miller*, Dr. Walter Muskovic*
With Labor announcing a Senate inquiry into the ‘whole My Health Record system’, it’s worth considering what has been learnt so far in public discussions, which have brought the Federal Health Minister Greg Hunt into the spotlight in quite a different way to his normal announcements this year. Back in January, Hunt launched a major report from the Australian Council of Learned Academies on ‘The Future of Precision Medicine in Australia’ that included contributions from all major scientific, humanities and social science perspectives. It’s an interesting, well written overview of this emerging field. One way of understanding Precision Medicine is that it builds on advances in ‘-omics’ methods (see section 2.5 of the report) to build large datasets with a view to developing new diagnostic and therapeutic techniques. Minister Hunt has spent much of the year announcing funding in line with this vision, especially through the ‘Genomic Health’ mission, as well as funding for specific conditions, including cerebral palsy, severe epilepsy, kidney disease.
With only a few notable exceptions, public discussion of Precision Medicine and its potential has been completely disconnected from the My Health Record discussion. Instead, the focus has been on multiple dimensions of privacy (especially stigma, domestic violence risks); the breadth of organisations and individuals who can access the document; multiple concerns about insurance companies accessing records through the My Health Record database; and the clinical utility or otherwise of using data based on billing, and record summaries organised as PDFs. Experts in systems management, procurement or coding point to ways the problems with security, access and control are ‘locked-in’, tempering others’ calls to ‘fix the problems’ through amending legislation or penalties.
There are lessons in all this for the successful rollout of Precision Medicine, which hinges in large part on the widespread aggregation of biological materials. A combination of careful public consultation and a more open-ended design ethos could have avoided many of these problems, which are likely to have profound flow on effects.
Lesson 1: Make clear and compelling case for how the records will make us healthier
The ‘All of Us’ programme in the United States — presented as ‘a historic effort to gather data from one million or more people living in the United States to accelerate research and improve health’, associated with the US Precision Medicine Initiative set up on the final months of the Obama presidency — made a compelling case for the need for a large, voluntary biobank where new diseases are diagnosed, and treatments developed. The ‘sales pitch’ was a very broad one: give us your tissue samples, and data relevant to social and environmental determinants of health like fitness tracking data, and approved scientists will mine the data to make you better; crucially, considering America’s patchy history with biomedical research and minority populations, this ask has been framed as a deeply inclusive and participatory one. The benefits of the My Health Record are less clear: the website focuses on the portability of data without any reference to providing infrastructure for researchers or other government agencies, despite these featuring very prominently in the legislation.
This problem of how different constituencies understand trade-offs is compounded by insurers and pharmaceutical companies publicly seeking the records for uses that do not align with the interests of citizens. ‘Managing costs’ for insurance companies, for example, may mean excluding people from health or life insurance products based on genetic mutations. What is completely lost is any narrative about progress — will this system help us be healthier? How?
Lesson 2: No issue, no (effective) data platform.
“Digital Health is the penicillin of our time, with Precision Medicine and genomics offering opportunities to cure previously incurable diseases and deliver better life saving medicine”. — Greg Hunt.
The rhetoric about Precision Medicine seems boundless — another magic bullet to cure in a single hit. However, what the debates over the Australian My Health Records reveal is that databases of ‘health data’ are not inherently useful. They require careful crafting concerning agreed standards for data inputs, ethical access and are most successful when responding to pressing issues in the community. One could extend Noortje Marres quip about ‘no issue, no public’ in this way — data is quite literally an act of giving, not a passive unit with inherent power.
Understanding the relations between experts, patients, devices and government in successful health database projects like the Hip and Knee Replacement Register demonstrates this etymological point. This registry was developed due to shared professional concerns about poor devices being approved and has successfully led to regulatory changes.
Expert commentary suggests that the My Health Record system was constrained by its political deadlines. Graham Grieve has called for a restructure of the system, arguing that the architecture lacks coherence: “Enterprises don’t automate their systems by collecting a huge pile of PDF documents.”
Furthermore, without justification for how and why government agency and third-party access defaults are set, journalists and members of the public are less likely to trade off their privacy to such a wide range of actors, many of whom have interests counter to their own health. The fate of the UK’s ‘care.data’ project, led by the Australian Digital Health Agency’s Tim Kelsey, bears out this point: the project was abandoned in July 2016 over concerns patient data could be sold to insurance and pharmaceutical companies.
Lesson 3: May a thousand databases bloom? Need for a diverse economy of health data
Data cooperatives like Midata.coop and other non-traditional ways of aggregating data are likely to be an important part of the emerging horizon of Precision Medicine. Decentralisation has advantages of spreading the risk of hacking and enabling more democratic decision-making, as well as flow of benefits. Open Application Programming Interfaces (APIs) can enable information exchange between platforms, mitigating a key security concerns around hacking. An ecosystem of democratically controlled, patient-centred databases would gel with calls for transforming both our understanding of the ‘the economy’ and its structure.
Lesson 4: Ownership matters to both the ethics of the program and ‘public acceptance’
Trust isn’t enough. Trust is active too — we trust healthcare institutions to care for us, but this trust is not absolute and does not automatically extend to ‘digital health’. Publics across the globe understand private ownership of healthcare institutions act according to their incentives, which seldom align with societal goals of equitable access to care.
The success of Precision Medicine hinges on public dialogue about the promise of better health. In this context, ownership over health data goes beyond the legal question of who technically owns the rights to patient records. How do we have a discussion about the collective ownership of health in ways that acknowledge the power of gifts to bond a society, for example? How could an appreciation of the diverse forms of social bonds that make up good health become the starting point for a discourse of innovation, rather than seen as a barrier to technological progress?
Lesson 5: ‘Health’ is an increasingly slippery concept in a post-genomic age
A key understanding of Precision Medicine is its emphasis on the ways it brings social and environmental data to bear on our understanding of disease risk. Thus, new databases of ‘healthy’ populations, such as the Medical Genome Reference Bank, are helping to shed light on how risky gene mutations manifest as disease.
Concerns over tumour sequencing data being added to the My Health Record through systems designed by the Garvan Institute’s corporate arm, ‘Genome.One’, show both the power of genomic data in the public imagination and need for appropriate protections. Australia remains one of the final jurisdictions without legislation forbidding insurers from denying coverage based on genetic conditions. This should be an easy step for the government, one that hopefully saves us from another ‘Downfall’ parody.
Conclusion
The societal implications of the scientific developments around Precision Medicine are profound and require a wide-ranging public discussion about what it means to be healthy or diseased, who is the baseline against which claims of health are to be measured, and who should benefit from the scientific research into health and disease. These discussions require a sociological imagination — one that ties cases to populations and societal institutions; and can enable a sober analysis of the tradeoffs precision medicine poses.
*UNSW, Sydney; Australian Research Council Centre of Excellence in Convergent Bio-Nano Science and Technology.
