Value from health data: European opportunity to catalyse progress in digital health

Privacy-preserving federated machine learning on FAIR health data: A real-world application

Objective

This paper introduces a privacy-preserving federated machine learning (ML) architecture built upon Findable, Accessible, Interoperable, and Reusable (FAIR) health data. It aims to devise an architecture for executing classification algorithms in a federated manner, enabling collaborative model-building among health data owners without sharing their datasets.

Materials and methods

Utilizing an agent-based architecture, a privacy-preserving federated ML algorithm was developed to create a global predictive model from various local models. This involved formally defining the algorithm in two steps: data preparation and federated model training on FAIR health data and constructing the architecture with multiple components facilitating algorithm execution. The solution was validated by five healthcare organizations using their specific health datasets.

Results

Five organizations transformed their datasets into Health Level 7 Fast Healthcare Interoperability Resources via a common FAIRification workflow and software set, thereby generating FAIR datasets. Each organization deployed a Federated ML Agent within its secure network, connected to a cloud-based Federated ML Manager. System testing was conducted on a use case aiming to predict 30-day readmission risk for chronic obstructive pulmonary disease patients and the federated model achieved an accuracy rate of 87%.

Discussion

The paper demonstrated a practical application of privacy-preserving federated ML among five distinct healthcare entities, highlighting the value of FAIR health data in machine learning when utilized in a federated manner that ensures privacy protection without sharing data.

Conclusion

This solution effectively leverages FAIR datasets from multiple healthcare organizations for federated ML while safeguarding sensitive health datasets, meeting legislative privacy and security requirements.

Medical history predicts phenome-wide disease onset and enables the rapid response to emerging health threats

The COVID-19 pandemic exposed a global deficiency of systematic, data-driven guidance to identify high-risk individuals. Here, we illustrate the utility of routinely recorded medical history to predict the risk for 1883 diseases across clinical specialties and support the rapid response to emerging health threats such as COVID-19. We developed a neural network to learn from health records of 502,460 UK Biobank. Importantly, we observed discriminative improvements over basic demographic predictors for 1774 (94.3%) endpoints. After transferring the unmodified risk models to the All of US cohort, we replicated these improvements for 1347 (89.8%) of 1500 investigated endpoints, demonstrating generalizability across healthcare systems and historically underrepresented groups. Ultimately, we showed how this approach could have been used to identify individuals vulnerable to severe COVID-19. Our study demonstrates the potential of medical history to support guidance for emerging pandemics by systematically estimating risk for thousands of diseases at once at minimal cost.

Challenges of artificial intelligence in medicine and dermatology

Artificial intelligence (AI) in medicine and dermatology brings additional challenges related to bias, transparency, ethics, security, and inequality. Bias in AI algorithms can arise from biased training data or decision-making processes, leading to disparities in health care outcomes. Addressing bias requires careful examination of the data used to train AI models and implementation of strategies to mitigate bias during algorithm development. Transparency is another critical challenge, as AI systems often operate as black boxes, making it difficult to understand how decisions are reached. Ensuring transparency in AI algorithms is vital to gaining trust from both patients and health care providers. Ethical considerations arise when using AI in health care, including issues such as informed consent, privacy, and the responsibility for the decisions made by AI systems. It is essential to establish clear guidelines and frameworks that govern the ethical use of AI, including maintaining patient autonomy and protecting sensitive health information. Security is a significant concern in AI systems, as they rely on vast amounts of sensitive patient data. Protecting these data from unauthorized access, breaches, or malicious attacks is paramount to maintaining patient privacy and trust in AI technologies. Lastly, the potential for inequality arises if AI technologies are not accessible to all populations, leading to a digital divide in health care. Efforts should be made to ensure that AI solutions are affordable, accessible, and tailored to the needs of diverse communities, mitigating the risk of exacerbating existing health care disparities. Addressing these challenges is crucial for AI's responsible and equitable integration in medicine and dermatology.

Public Preferences for Digital Health Data Sharing: Discrete Choice Experiment Study in 12 European Countries

BACKGROUND

With new technologies, health data can be collected in a variety of different clinical, research, and public health contexts, and then can be used for a range of new purposes. Establishing the public's views about digital health data sharing is essential for policy makers to develop effective harmonization initiatives for digital health data governance at the European level.

OBJECTIVE

This study investigated public preferences for digital health data sharing.

METHODS

A discrete choice experiment survey was administered to a sample of European residents in 12 European countries (Austria, Denmark, France, Germany, Iceland, Ireland, Italy, the Netherlands, Norway, Spain, Sweden, and the United Kingdom) from August 2020 to August 2021. Respondents answered whether hypothetical situations of data sharing were acceptable for them. Each hypothetical scenario was defined by 5 attributes ("data collector," "data user," "reason for data use," "information on data sharing and consent," and "availability of review process"), which had 3 to 4 attribute levels each. A latent class model was run across the whole data set and separately for different European regions (Northern, Central, and Southern Europe). Attribute relative importance was calculated for each latent class's pooled and regional data sets.

RESULTS

A total of 5015 completed surveys were analyzed. In general, the most important attribute for respondents was the availability of information and consent during health data sharing. In the latent class model, 4 classes of preference patterns were identified. While respondents in 2 classes strongly expressed their preferences for data sharing with opposing positions, respondents in the other 2 classes preferred not to share their data, but attribute levels of the situation could have had an impact on their preferences. Respondents generally found the following to be the most acceptable: a national authority or academic research project as the data user; being informed and asked to consent; and a review process for data transfer and use, or transfer only. On the other hand, collection of their data by a technological company and data use for commercial communication were the least acceptable. There was preference heterogeneity across Europe and within European regions.

CONCLUSIONS

This study showed the importance of transparency in data use and oversight of health-related data sharing for European respondents. Regional and intraregional preference heterogeneity for "data collector," "data user," "reason," "type of consent," and "review" calls for governance solutions that would grant data subjects the ability to control their digital health data being shared within different contexts. These results suggest that the use of data without consent will demand weighty and exceptional reasons. An interactive and dynamic informed consent model combined with oversight mechanisms may be a solution for policy initiatives aiming to harmonize health data use across Europe.

FAIR, ethical, and coordinated data sharing for COVID-19 response: a scoping review and cross-sectional survey of COVID-19 data sharing platforms and registries

Data sharing is central to the rapid translation of research into advances in clinical medicine and public health practice. In the context of COVID-19, there has been a rush to share data marked by an explosion of population-specific and discipline-specific resources for collecting, curating, and disseminating participant-level data. We conducted a scoping review and cross-sectional survey to identify and describe COVID-19-related platforms and registries that harmonise and share participant-level clinical, omics (eg, genomic and metabolomic data), imaging data, and metadata. We assess how these initiatives map to the best practices for the ethical and equitable management of data and the findable, accessible, interoperable, and reusable (FAIR) principles for data resources. We review gaps and redundancies in COVID-19 data-sharing efforts and provide recommendations to build on existing synergies that align with frameworks for effective and equitable data reuse. We identified 44 COVID-19-related registries and 20 platforms from the scoping review. Data-sharing resources were concentrated in high-income countries and siloed by comorbidity, body system, and data type. Resources for harmonising and sharing clinical data were less likely to implement FAIR principles than those sharing omics or imaging data. Our findings are that more data sharing does not equate to better data sharing, and the semantic and technical interoperability of platforms and registries harmonising and sharing COVID-19-related participant-level data needs to improve to facilitate the global collaboration required to address the COVID-19 crisis.

Talking Ethics Early in Health Data Public Private Partnerships

Data access and data sharing are vital to advance medicine. A growing number of public private partnerships are set up to facilitate data access and sharing, as private and public actors possess highly complementary health data sets and treatment development resources. However, the priorities and incentives of public and private organizations are frequently in conflict. This has complicated partnerships and sparked public concerns around ethical issues such as trust, justice or privacy-in turn raising an important problem in business and data ethics: how can ethical theory inform the practice of public and private partners to mitigate misaligned incentives, and ensure that they can deliver societally beneficial innovation? In this paper, we report on the development of the Swiss Personalized Health Network's ethical guidelines for health data sharing in public private partnerships. We describe the process of identifying ethical issues and engaging core stakeholders to incorporate their practical reality on these issues. Our report highlights core ethical issues in health data public private partnerships and provides strategies for how to overcome these in the Swiss health data context. By agreeing on and formalizing ethical principles and practices at the beginning of a partnership, partners and society can benefit from a relationship built around a mutual commitment to ethical principles. We present this summary in the hope that it will contribute to the global data sharing dialogue.

A Data Transformation Methodology to Create Findable, Accessible, Interoperable, and Reusable Health Data: Software Design, Development, and Evaluation Study

BACKGROUND

Sharing health data is challenging because of several technical, ethical, and regulatory issues. The Findable, Accessible, Interoperable, and Reusable (FAIR) guiding principles have been conceptualized to enable data interoperability. Many studies provide implementation guidelines, assessment metrics, and software to achieve FAIR-compliant data, especially for health data sets. Health Level 7 (HL7) Fast Healthcare Interoperability Resources (FHIR) is a health data content modeling and exchange standard.

OBJECTIVE

Our goal was to devise a new methodology to extract, transform, and load existing health data sets into HL7 FHIR repositories in line with FAIR principles, develop a Data Curation Tool to implement the methodology, and evaluate it on health data sets from 2 different but complementary institutions. We aimed to increase the level of compliance with FAIR principles of existing health data sets through standardization and facilitate health data sharing by eliminating the associated technical barriers.

METHODS

Our approach automatically processes the capabilities of a given FHIR end point and directs the user while configuring mappings according to the rules enforced by FHIR profile definitions. Code system mappings can be configured for terminology translations through automatic use of FHIR resources. The validity of the created FHIR resources can be automatically checked, and the software does not allow invalid resources to be persisted. At each stage of our data transformation methodology, we used particular FHIR-based techniques so that the resulting data set could be evaluated as FAIR. We performed a data-centric evaluation of our methodology on health data sets from 2 different institutions.

RESULTS

Through an intuitive graphical user interface, users are prompted to configure the mappings into FHIR resource types with respect to the restrictions of selected profiles. Once the mappings are developed, our approach can syntactically and semantically transform existing health data sets into HL7 FHIR without loss of data utility according to our privacy-concerned criteria. In addition to the mapped resource types, behind the scenes, we create additional FHIR resources to satisfy several FAIR criteria. According to the data maturity indicators and evaluation methods of the FAIR Data Maturity Model, we achieved the maximum level (level 5) for being Findable, Accessible, and Interoperable and level 3 for being Reusable.

CONCLUSIONS

We developed and extensively evaluated our data transformation approach to unlock the value of existing health data residing in disparate data silos to make them available for sharing according to the FAIR principles. We showed that our method can successfully transform existing health data sets into HL7 FHIR without loss of data utility, and the result is FAIR in terms of the FAIR Data Maturity Model. We support institutional migration to HL7 FHIR, which not only leads to FAIR data sharing but also eases the integration with different research networks.

Towards the European Health Data Space (EHDS) ecosystem: A survey research on future health data scenarios

OBJECTIVE

The European Health Data Space (EHDS) aims to provide better exchange and expand access to health data across Europe. In this way, the EHDS will support healthcare delivery (known as the "primary use of data") and facilitate access to health data for research and policy-making purposes (known as the "secondary use of data"). To achieve this goal, we need to build the required ecosystem of the EHDS with all healthcare stakeholders.

MATERIALS AND METHODS

We conducted a survey research study to explore the health informaticians' recommendations on future health data scenarios shaping the EHDS ecosystem. We created an anonymous- online questionnaire and disseminated it through wide international networks of health informaticians. In addition, we conducted a workshop during the Medical Informatics Europe Conference (MIE2022) and invited the attendees to complete the questionnaire during the workshop.

RESULTS

We received 43 responses to our questionnaire from 15 European Union (EU) countries and 7 non-EU countries. Most respondents described the current health data scenario in their countries as a traditional healthcare system with moderate growth (25.6 %, n = 11). The second selected scenario was the reinventing healthcare scenario in a data-driven one-world framework (23.3 %, n = 10).

DISCUSSION

The results of this work are matched with the findings of the recently published study on digital health implementation in the EU (conducted by the French government in April 2022). This also reflects the current ongoing efforts in the EU countries to deploy national infrastructure for health data management, exchange, and sharing.

CONCLUSIONS

Upon the respondents' recommendations, there is a strong need to support the health democratization scenarios in Europe, as the main driver for building the EHDS ecosystem.

Semantic wikis as flexible database interfaces for biomedical applications

Several challenges prevent extracting knowledge from biomedical resources, including data heterogeneity and the difficulty to obtain and collaborate on data and annotations by medical doctors. Therefore, flexibility in their representation and interconnection is required; it is also essential to be able to interact easily with such data. In recent years, semantic tools have been developed: semantic wikis are collections of wiki pages that can be annotated with properties and so combine flexibility and expressiveness, two desirable aspects when modeling databases, especially in the dynamic biomedical domain. However, semantics and collaborative analysis of biomedical data is still an unsolved challenge. The aim of this work is to create a tool for easing the design and the setup of semantic databases and to give the possibility to enrich them with biostatistical applications. As a side effect, this will also make them reproducible, fostering their application by other research groups. A command-line software has been developed for creating all structures required by Semantic MediaWiki. Besides, a way to expose statistical analyses as R Shiny applications in the interface is provided, along with a facility to export Prolog predicates for reasoning with external tools. The developed software allowed to create a set of biomedical databases for the Neuroscience Department of the University of Padova in a more automated way. They can be extended with additional qualitative and statistical analyses of data, including for instance regressions, geographical distribution of diseases, and clustering. The software is released as open source-code and published under the GPL-3 license at https://github.com/mfalda/tsv2swm .

Reverse Engineering of Digital Measures: Inviting Patients to the Conversation

Background

Digital measures offer an unparalleled opportunity to create a more holistic picture of how people who are patients behave in their real-world environments, thereby establishing a better connection between patients, caregivers, and the clinical evidence used to drive drug development and disease management. Reaching this vision will require achieving a new level of co-creation between the stakeholders who design, develop, use, and make decisions using evidence from digital measures.

Summary

In September 2022, the second in a series of meetings hosted by the Swiss Federal Institute of Technology in Zürich, the Foundation for the National Institutes of Health Biomarkers Consortium, and sponsored by Wellcome Trust, entitled "Reverse Engineering of Digital Measures," was held in Zurich, Switzerland, with a broad range of stakeholders sharing their experience across four case studies to examine how patient centricity is essential in shaping development and validation of digital evidence generation tools.

Key Messages

In this paper, we discuss progress and the remaining barriers to widespread use of digital measures for evidence generation in clinical development and care delivery. We also present key discussion points and takeaways in order to continue discourse and provide a basis for dissemination and outreach to the wider community and other stakeholders. The work presented here shows us a blueprint for how and why the patient voice can be thoughtfully integrated into digital measure development and that continued multistakeholder engagement is critical for further progress.

Sensing the (digital) pulse. Future steps for improving the secondary use of data for research in Switzerland

Introduction

Ensuring that the health data infrastructure and governance permits an efficient secondary use of data for research is a policy priority for many countries. Switzerland is no exception and many initiatives have been launched to improve its health data landscape. The country now stands at an important crossroad, debating the right way forward. We aimed to explore which specific elements of data governance can facilitate - from ethico-legal and socio-cultural perspectives - the sharing and reuse of data for research purposes in Switzerland.

Methods

A modified Delphi methodology was used to collect and structure input from a panel of experts via successive rounds of mediated interaction on the topic of health data governance in Switzerland.

Results

First, we suggested techniques to facilitate data sharing practices, especially when data are shared between researchers or from healthcare institutions to researchers. Second, we identified ways to improve the interaction between data protection law and the reuse of data for research, and the ways of implementing informed consent in this context. Third, we put forth ideas on policy changes, such as the steps necessary to improve coordination between different actors of the data landscape and to win the defensive and risk-adverse attitudes widespread when it comes to health data.

Conclusions

After having engaged with these topics, we highlighted the importance of focusing on non-technical aspects to improve the data-readiness of a country (e.g., attitudes of stakeholders involved) and of having a pro-active debate between the different institutional actors, ethico-legal experts and society at large.

In the shadow of privacy: Overlooked ethical concerns in COVID-19 digital epidemiology

The COVID-19 pandemic witnessed a surge in the use of health data to combat the public health threat. As a result, the use of digital technologies for epidemic surveillance showed great potential to collect vast volumes of data, and thereby respond more effectively to the healthcare challenges. However, the deployment of these technologies raised legitimate concerns over risks to individual privacy. While the ethical and governance debate focused primarily on these concerns, other relevant issues remained in the shadows. Leveraging examples from the COVID-19 pandemic, this perspective article aims to investigate these overlooked issues and their ethical implications. Accordingly, we explore the problem of the digital divide, the role played by tech companies in the public health domain and their power dynamics with the government and public research sector, and the re-use of personal data, especially in the absence of adequate public involvement. Even if individual privacy is ensured, failure to properly engage with these other issues will result in digital epidemiology tools that undermine equity, fairness, public trust, just distribution of benefits, autonomy, and minimization of group harm. On the contrary, a better understanding of these issues, a broader ethical and data governance approach, and meaningful public engagement will encourage adoption of these technologies and the use of personal data for public health research, thus increasing their power to tackle epidemics.

Precision Oncology in Canada: Converting Vision to Reality with Lessons from International Programs

Canada's healthcare system, like others worldwide, is immersed in a process of evolution, attempting to adapt conventional frameworks of health technology assessment (HTA) and funding models to a new landscape of precision medicine in oncology. In particular, the need for real-world evidence in Canada is not matched by the necessary infrastructure and technologies required to integrate genomic and clinical data. Since healthcare systems in many developed nations face similar challenges, we adopted a solutions-based approach and conducted a search of worldwide programs in personalized medicine, with an emphasis on precision oncology. This search strategy included review articles published between 1 January 2016 and 1 March 2021 and hand-searches of their reference lists for relevant publications back to 1 December 2005. Thirty-nine initiatives across 37 countries in Europe, Australasia, Africa, and the Americas had the potential to lead to real-world data (RWD) on the clinical utility of oncology biomarkers. We highlight four initiatives with helpful lessons for Canada: Genomic Medicine France 2025, UNICANCER, the German Medical Informatics Initiative, and CANCER-ID. Among the 35 other programs evaluated, the main themes included the need for collaboration and systems to support data harmonization across multiple jurisdictions. In order to generate RWD in precision oncology that will prove acceptable to HTA bodies, Canada must take a national approach to biomarker strategy and unite all stakeholders at the highest level to overcome jurisdictional and technological barriers.

European Health Data Space—An Opportunity Now to Grasp the Future of Data-Driven Healthcare

The May 2022 proposal from the European commission for a ‘European health data space’ envisages advantages for health from exploiting the growing mass of health data in Europe. However, key stakeholders have identified aspects that demand clarification to ensure success. Data will need to be freed from traditional silos to flow more easily and to cross artificial borders. Wide engagement will be necessary among healthcare professionals, researchers, and the patients and citizens that stand to gain the most but whose trust must be won if they are to allow use or transfer of their data. This paper aims to alert the wider scientific community to the impact the ongoing discussions among lawmakers will have. Based on the literature and the consensus findings of an expert multistakeholder panel organised by the European Alliance for Personalised Medicine (EAPM) in June 2022, it highlights the key issues at the intersection of science and policy, and the potential implications for health research for years, perhaps decades, to come.

The Challenges of Big Data for Research Ethics Committees: A Qualitative Swiss Study

Big data trends in health research challenge the oversight mechanism of the Research Ethics Committees (RECs). The traditional standards of research quality and the mandate of RECs illuminate deficits in facing the computational complexity, methodological novelty, and limited auditability of these approaches. To better understand the challenges facing RECs, we explored the perspectives and attitudes of the members of the seven Swiss Cantonal RECs via semi-structured qualitative interviews. Our interviews reveal limited experience among REC members with the review of big data research, insufficient expertise in data science, and uncertainty about how to mitigate big data research risks. Nonetheless, RECs could strengthen their oversight by training in data science and big data ethics, complementing their role with external experts and ad hoc boards, and introducing precise shared practices.

Follow *the* science? On the marginal role of the social sciences in the COVID-19 pandemic

In this paper, we use the case of the COVID-19 pandemic in Europe to address the question of what kind of knowledge we should incorporate into public health policy. We show that policy-making during the COVID-19 pandemic has been biomedicine-centric in that its evidential basis marginalised input from non-biomedical disciplines. We then argue that in particular the social sciences could contribute essential expertise and evidence to public health policy in times of biomedical emergencies and that we should thus strive for a tighter integration of the social sciences in future evidence-based policy-making. This demand faces challenges on different levels, which we identify and discuss as potential inhibitors for a more pluralistic evidential basis.