What is secondary use of health data and why is it important? When the ODI's ‘Secondary use of health data in Europe’ project was commissioned in autumn 2020, there wasn't a common definition or framework available. Over the last year, the ODI has developed and launched a report which clearly defines the secondary use of health data and provides stakeholders with a set of topic resources that are among the most popular in the world.
Mark Boyd, Principal Policy Associate and lead author of the report, charts the journey and the next steps.
Health data in Europe
In recent years, the European Commission (EC) has reoriented policy and strategy goals across European economic sectors with the aim of establishing Europe’s global leadership in the data economy. In the European Strategy for Data (2020), the EC highlighted Europe's unique position, contrasted with the ‘government surveillance … without sufficient safeguards for individuals’ model of China and the free-market, low-regulated model of the United States. Europe's strategy notes that it is seeking a middle ground: ‘balancing the flow and wide use of data, while preserving high privacy, security, safety and ethical standards.’ Foundational to enabling this data economy is Europe's 2018 data privacy legislation, the General Data Protection Regulation (GDPR), which defines individual data rights including how data can be shared and reused.
As a key part of this strategic focus on a data economy, Europe has proposed the creation of Data Spaces: a new regional structure for sector-based data sharing, including the health sector, transport sector, energy sector, industrial sector, agriculture sector, finance sector, public sector, skills, and the Green New Deal.
Because of the range and speed of data and digital opportunities emerging around health data, the European Health Data Space (EHDS) is one of the first of the European Data Spaces to be created by the EC. The challenge for the EC is to demonstrate that GDPR does not prevent innovative uses of health data, so that Europe does not fall behind in global investment in health R&D.
Scores and outcomes
When we began work on this project, one of the only frameworks we found had been published in 2007 by the American Medical Informatics Association. In addition, we couldn't even locate a clear definition of 'secondary use of health data' that described what sort of data was involved and what the potential benefits of its use could be.
For our own approach, we believed that a quantitative ranking quadrant and qualitative sharing of good practices would best support each country to benchmark against peers in different ways, such as by population size or by gross domestic product (GDP).
We found that an advantage of a quantitative approach is it gives a focus for discussion and a spur for timely and concrete action. But a disadvantage of a quantitative approach is that it doesn’t always capture the ‘on-the-ground’ reality of policy implementation, and so doesn’t necessarily tell the full story. The qualitative approach is a necessary counterpoint to this. But we found that most important of all – and the most effective use of our research and analysis findings – is fostering conversations and encouraging a focus on the specific success indicators where a country can improve.
In practice, we saw less ‘elbow-jostling’ around rankings and instead a desire by all, even those policymakers whose countries were already scoring highly, to make sure health data was used to get the best patient outcomes for everyone. Sharing good practices and highlighting what individual countries and EU member states were doing in this area also fed into the EC’s ‘virtuous cycle’ of policy implementation and influence, as country-level activities created templates and evidence that can be adapted by other countries or by the EC as a whole.
Around the table, around the region
In our analysis and follow-up work sharing the results with different countries across Europe, we were surprised by the diversity of approaches to how policymakers were using this work to support the better use of health data in their country:
- In Austria, recent amendments to the Research Organisation Act allowed stakeholders to collaborate on industry responses to health data use, drawing on this policy framework
- In Belgium, the research was the first-of-its-kind catalyst for bringing broad data expertise together around the table; and the mapping of ecosystem roles helped to ensure and shape multi-stakeholder representation on the newly-formed Belgium Health Data Authority
- In Greece, the country’s low score was used by a collaborative partnership between the Ministries of Digital Government and of Health to work together to identify improvement strategies that could be taken forward under newly available funding for digital health initiatives
- In Hungary, internal teams within Roche used the report findings and research in design thinking workshops to identify new project opportunities
- In the Netherlands, the findings supported the local Chamber of Commerce to engage with industry groups and commit to enhancing the digital health tech industry locally
- In Romania, the research supported multi-stakeholder partnerships including government, industry and patient groups to create concrete action plans that addressed health inequalities, particularly under initiatives aligned with Europe's Beating Cancer Plan
Several countries previously had challenges in fostering cross-collaborative forums on health data, but with our clear policy framework that shared ranking of countries and accessible interactive tools, there was renewed interest to work together. For example, the Ministries of Health in Austria and Greece used this project to prompt discussion in new forums about the secondary use of health data, across ministries of digital transformation and external stakeholders
Looking ahead
The ODI’s 'working in the open' approach gives stakeholders and policy analysts several different ways to engage with our research and test our assumptions or adjust their strategy roadmaps based on what elements they think are most important.
Because the individual country datasheets are available alongside the tool and report, readers can download these to experiment with how country scores could be improved if good practices were applied to particular policy framework areas in their country. This data could also be used to create new analyses and new digital tools for others. Participation of diverse communities in developing Europe’s health data ecosystem will help to make it more trustworthy. And as new data sources come on stream (including anonymised data), and as new digital tools are developed and new use cases are identified, we’d love to see our policy framework tested against these new challenges and opportunities.
We would also love to expand our analysis and test a policy framework for secondary use of health data globally, including a version focussing on North America or South-East Asia, for contrasting cultural and legislative approaches to data rights and healthcare innovation than Europe. We recognise that the EHDS approach is likely to be globally influential, in a similar way to how GDPR has influenced global approaches to data rights through a first mover advantage, and so we’d also love to explore developing a version suitable for emerging health data ecosystems in low- and middle-income countries, that is tailored to also support wider ambitions for their health data ecosystem development.