Webinar: 'The role of a European Health Data Space in a pandemic'

Speaker comments

• 

  Ioana-Maria Gligor
  Head of Unit, European Reference Networks and Digital Health, DG SANTE

  See the full presentation

  Key remarks 

  • Outline the work of the EC with its data strategy to create a health data space and advance recommendations made in February 2019 for cross EU border data flow.
  • The European Provide an overview of 3 workshops being held with Member States on processing primary and secondary health data to develop a study an recommend regulatory action (in particular CoC between Member States and Stakeholders).
• 

  Nigel Hughes
  Scientific Director Observational Health Data Analytics/Epidemiology, Janssen Research and Development AND Project Leader, European Health Data & Evidence Network (EHDEN)

  Key remarks 

  • Real world data is a challenge normally but is even more complicated during a pandemic. Health systems aren’t ready today to fully reap the benefits of data. There is unclarity on the interpretation of public and legitimate interest in the GDPR. Divergences exist on the legal basis for health data processing as well as the safeguards to apply. Cross-border research suffers from this fragmentation. A research initiative based on re-using health data for public/legitimate interest under the GDPR can be allowed in some countries but not in others, which require consent. This makes difficult for healthcare providers to share data with private entities, and for them to share precious clinical research data with partners in initiatives like IMI. We suffer from a lack of an adequate interpretation layer of GDPR, which impacts on the implementation layer, leaving ambiguity for all parties concerned, and we need a ‘plain language’ approach to respond to this
  • Most often we learn post hoc, but we need to have relevant and representative data to inform a whole cascade of decisions, from planning to clinical, today and not in many tomorrows. We also need to be prepared for the next wave of SARS-CoV-2/COVID-19 and inevitable epidemics/pandemics of other pathogens
  • International initiatives, such as South Korea’s COVID-19 Open Data Initiative, or the OHDSI COVID-19 study-a-thon need to be emulated across Europe asap, with a collective approach by Member States to at least aggregate relevant and representative data going forward vs. piecemeal initiatives and studies. We have not been in a position to utilise learning from the past, inclusive of e.g. understanding how best to repurpose drugs for COVID-19 therapy (and which became more politicised in the absence of and/or reliable data)
  • Trust is key for the success of research based on patient data. In the existing, fragmented data use context, a Code of Conduct would provide the necessary reassurances to patients on data use governance and safeguards. This would generate trust. It’s also crucial that patients are informed about the potential of health data re-use to drive health innovation, and that public-private partnerships advance much of today’s research. A good patient-awareness example is Data Saves Lives. It aims to create a multi-stakeholder dialogue on the responsible use of health data and good practices across Europe, which in turn will create a common basis for working appropriately with health data
  • Together with a Code of Conduct, we need federated data models to provide a new paradigm for the discovery and analysis of health data in Europe by building a large-scale, federated network of data sources standardized to a common data model. This is key to overcome health data access barriers in Europe. The European Health Data & Evidence Network (EHDEN) under an IMI aims to advance this paradigm. As health data is diversifying beyond EHR data per se, e.g. digital biomarkers, wearables, and the like, we need a wider framework to work with more holistic datasets within an appropriate code of conduct.
• 

  Nazar Rasul
  Head of Technology & Innovation Strategy, Siemens Healthineers

  See the full presentation

  Key remarks  

  • Expanding precision medicine – We believe that medicine will become more precise. However, medicine today is generally based on a “one-size-fits-all” practice, and where targeted therapies are possible, it is impractical to scale. The goal of expanding precision medicine is to provide the right treatment at the right time for every patient. Tailoring treatment starts with a highly-specific diagnosis without unwarranted variation. Based on data integrated from existing sources, adding genomics and radiomics enables a holistic understanding of the individual. These unique characteristics steer the personalization of treatment. A precise understanding of a patient’s condition is the most effective approach to deliver outcomes favorable to all stakeholders.
  • Improve diagnostic accuracy – Generating and integrating highly-specific and quantitative data lays the groundwork for accurate diagnosis. Amending the diagnostic process with new types of biomarkers, such as genomics and radiomics, and lifestyle components, creates a holistic understanding of the individual and the disease. Providing this information that has been curated for relevance, at the point of clinical decision-making, improves diagnostic accuracy.
  • Reduce unwarranted variations – Building best-practice standards along the continuum of care is a prerequisite for consistent, evidence-based care. The key levers for reducing variations are adapting to patients’ individual needs and reducing the subjectivity of operators and physicians. Correspondingly, automation and assisted decision-making based on reproducible technology activate these levers to create consistent diagnostic results. Steady, iterative improvements using this approach effectively reduce unwarranted variations.
  • Personalize when it matters – When personalizing treatment strategies, a precise understanding of a patient’s condition at the point of decision is essential. Information from imaging technologies, lab testing, and genomics delivers specifics of both the patient and the disease that are necessary for tailoring therapy. Further, monitoring treatment efficacy ensures continuous delivery of appropriate care. Personalizing treatment when it matters delivers safer, more effective therapy while saving costs.
  • Advance therapy outcomes – Many opportunities to enhance therapy precision lie in the integration of imaging and treatment delivery. Therapeutic procedures from surgery to radiation oncology rely on integrated real-time or multimodal image guidance and robotic assistance. This facilitates minimally and non-invasive interventions that support faster recovery and create fewer complications. This enables healthcare providers to advance therapy outcomes and make treatment safer, faster and less costly.
  • Data ownership – patients apps, owned and controlled by each individual citizen
  • New forms of collaboration between various stakeholders (research between HC providers and industry (MedTech, digital, biotech, etc.)
  • Decentralized data handling and storing with clear rules and guidance
  • Federated learning for use of AI in health and clinical decision making
• 

  Bert Verdonck
  Business Leader, Population Health Management International, Philips

  See the full presentation

  Key remarks

  • Scalability: There have been many challenges in the past to exchange data at scale between Members States, across regions and between healthcare providers. There have been many attempts and initiatives (many were EU funded) that achieved success in a pilot stage but were unable to scale up to create significant societal impact. A public data space should create the possibility to scale to national and international levels. The COVID crisis demonstrates the blatant need for such scalable data sharing infrastructures. For example: in The Netherlands Philips worked on a data sharing initiative between intensive care units for patients who had to relocate to other hospitals. But it remains difficult to share information between care centres for recovering patients (university, speciality, first line, home care).
  • Standardisation: AI becomes much more powerful if we have a common, structured language. For example: there was a lack of coordination to rapidly standardise relevant vital signs and symptoms for the COVID pandemic. If these ontologies would have been available, then vital information could have been collected, analysed and shared in a much earlier phase. Philips has proposed a governance or data trust structure to enable the setup of data exchanges. There are many national standardisation initiatives in the healthcare domain, but it remains hard to select and setup an effective governing organisation. The COVID crisis demonstrated that sharing infection data across borders was very difficult and everyone fell back on national systems. We need a pan-European data trust authority, a kind of Interpol for health data. Such EU level data trust could then facilitate the different national systems to ensure data can be shared quickly and effectively.
• 

  Frank Baitman
  Global Head, Compliance and Strategy, Digital Health Technologies, Roche

  Key remarks

  • Provide an international perspective and experiences of data sharing and processing challenges. Provide a perspective from Roche in its activities during the COVID-19 crisis.
  • Provide a perspective on the clarity needed for standards used in data sharing. The focus should be on openness to encourage data to be shared and are technologically neutral.
  • There have been several regimes for data protection. In the US HIPPA (Health Insurance Portability and Accountability Act) predates technology of today but was valuable to communicate to health workers and citizens the need for privacy, protecting data and the penalties for any breaches or infringements. GDPR also plays a valuable role but both regimes are not future proof and there is a need to see where it falls short and where it needs to adapt.
  • Data standards need to be updated to ensure data can cross not only borders in the EU but internationally as the virus sees no borders.
• 

  Cornelia Kutterer
  Senior Director European Government Affairs, AI, Privacy, Digital Policies, Microsoft

  See the full presentation

  Key remarks

  • Brief overview of the digital tools that customers on the frontlines of the COVID-19 response are deploying in Europe.
  • Perspectives on the technical and regulatory obstacles that limited the establishment of health data ecosystems in Europe: silos, lack of legal certainty and trust.
  • Perspectives on the technical solutions as well as the policy and regulatory initiatives that can help establishing a thriving health data culture and ecosystems in view of the European Health Data Space: establishing data interoperability standards, open data initiatives, trust and harmonization of GDPR’s interpretation vis-à-vis the processing of personal health data for secondary use.
  • Lessons learnt from the Covid-19 crisis in view of the establishment of the European Health Data Space
• 

  Mario Romao
  Global Director for Health IT Policy, Intel

  See the full presentation

  Key remarks

  • Outline early findings from a study Intel has sponsored with the Israel Tech Policy Institute to compare regulatory regimes for accessing health data for research. Phase 1 is completed and recommends:
    • Move beyond consent as grounds for non–interventional secondary use of health data (e.g. public interest, research).
    • Protect health data with a mix of de-identification (pseudonymisation, anonymisation, aggregation), processing controls (e.g. data use agreements, reference methodologies) and safeguard procedures as an internal ethical review board.
    • Foster technical solutions to safeguard patient’s data for secondary use.