Enriching knowledge and enhancing care through health data

Annual Conference, November 2018

We are delighted to announce that our third annual conference took place in Gothenburg, Sweden on 19-20 November. The theme of the conference was Realising the value of health data ~ showcasing pratical examples from research and healthcare. The multi-stakeholder nature of the participants led to interesting perspectives and discussion on how we are realising the value of health data.

A big thank you to all those involved in making this event such a success.


Conference highlights

Card image cap


Andrew Roddam


Chuck Friedman

University of Michigan

Card image cap


Data Quality Service for Hospitals


Card image cap



more in the future


Card image cap

IMI EHDEN project




 30 speakers, panellists and tutors, including patient representatives
 A keynote panel debates the optimal positioning of Real World Evidence in decision making
 We had 135 participants from a mixture of pharma, academia, health industry and IT and patient representatives, creating a multi-stakeholder audience, with multi-stakeholder speakers.


Pre-conference sessions

There were two parallel sessions.

 An introduction to the new IMI project EHDEN. This was led by Nigel Hughes and Peter Rijnbeek.
 i~HD ran a GDPR workshop, led by Nathan Lea, Peter Singleton, Brendan Barnes and Nikolaus Forgó.


Conference day 1

Dipak Kalra, President, launched the conference with a welcome address and a brief synopsis of i~HD. Our vision is to better enable the delivery of high-quality and innovative health care, more efficient research, increased efficiency of our health systems and ultimately improved patient outcomes - all through obtaining and sharing greater value from health data. i~HD is ideally placed to promote and enable this vision, as a neutral body with a multi-stakeholder membership.

At this year’s conference i~HD has put emphasis on demonstrating value to healthcare, to patients, to society and for research, from the use of health data.

Dipak Kalra, President i-HD


Nigel Hughes chaired the first afternoon session: Realising the value of health data for research: Pragmatic Clinical Trials.

Andrew Roddam, VP & Head Real World Evidence and Epidemiology, GSK
Andrew Roddam was the opening keynote speaker, presenting on the role of the pragmatic clinical trial in industry decision making. He emphasised the importance of first determining the specific research questions that need to be addressed when developing a new medicine. He described controlled trials as narrow windows: they do not include patients with co-morbidities or concurrent medications. Using PCTs we can conduct studies on more representative patients and make decisions in near to real time. In the future pragmatic studies may be a basis for regulatory decisions.

Mats Sundgren spoke about Pragmatic Clinical Trials: “Real World Evidence’s Hidden Gem”. From recent evidence PCTs can reduce research costs and provide valid insights into the wider usability of a drug, all with sufficient scientific quality. Rob Thwaites presented the IMI GetReal Toolbox for incorporating PCTs into effectiveness planning: the real world evidence navigator and Pragmagic. These tools can bring together all the ideas and questions that need to be addressed in designing a PCT. Tjeerd van Staa told the audience about the challenges of data quality and completeness in randomised trials (pragmatic and conventional). Currently there are variations in study design and different methodologies for using RWD. There are also variations in the collection of data, recording practices, data flows vary and even conventional trials (RCTs) suffer from uncertainty in the data. Heiner Bucher presented on registry based pragmatic randomized controlled trials for nationwide evaluations of health care interventions. He told the audience how his department had used a pragmatic randomised parallel group trial with data from primary care physicians who were the top antibiotic prescribers. Using personalised prescription feedback they observed these high prescription rates decreasing.


Nigel Hughes, Andrew Roddam, Mats Sundgren, Heiner Bucher, Rob Thwaites and Tjeerd van Staa.


There then followed a lively panel discussion, with all the speakers from this session, about pragmatic clinical trials, their unrealised potential and the challenges in conducting them. The discussion included topics on the benefit that a common search capability would bring if this could allow for real time querying across PCT sites, the relative importance of data quality to the decisions that can be made on the basis of PCT results, and how healthcare professionals and patients can be better incentivised to participate in PCTs.

John Wise chaired the next session on Realising the value of health data for research. He briefly explained the Pistoia Alliance concept and road map, including the project HELM and the production of an ISO standard.

Peter Rijnbeek spoke about the Observational Health Data Sciences and Informatics (OHDSI) Initiative and tools. He discussed the journey from raw real world data to obtaining the highest-quality evidence needed from it. There are OHDSI symposia occurring all over the world, sharing their experience of mapping data to a common data model and then analysing it. Miriam Sturkenboom was next to speak on the subject of transforming big health data into evidence to support vaccine benefit and risk monitoring, with examples from the IMI-ADVANCE project. She highlighted that we require the best evidence, at the right time, to support correct decision making on vaccines. The risks and benefits should be generated from the real data on large populations. Nigel Hughes presented on the journey from EMIF to EHDEN: scaling up the big data ecosystem across Europe. These are a related pair of IMI projects working with real world data and conducting observational real world studies. These are the approaches needed for 21st century research. Julian Isla, spoke of creating digital patients. Dravet syndrome, the condition of his son, was the inspiration for founding the organisation for Dravet syndrome: Foundation Twenty-nine. He stated that other industries are recording data and using it to construct feedback loops, but this is not yet happening in health. This is an opportunity. We need to change the way we capture and use information, and especially to harness the power of patients to contribute. Lars Lindsköld presented the Swedish national initiative to support the usage of systematic health data in personalised medicine. There is an emerging understanding that patients are different and therefore a need for studies to be more pragmatic.


Peter Rijnbeek, Miriam Sturkenboom, Nigel Hughes, Julian Isla and Lars Lindsköld.


A speakers panel discussion session followed about the potential benefits of large scale health data for research and for patients. This included topics on the impact wider adoption of a common data model might have, the impact of mapping data to a standard has on data quality, how to engender the trust of data sources and patients in sharing data, and the importance of a federated model of data access for trust.

Robert vander Stichele, finished the first day with his presentation on challenging the hierarchy of evidence. He deconstructed the pyramid of evidence…. He turned it upside down and suggested that there is not one single ideal pinnacle, if evidence from all sources is considered.

And then… dinner on the 29th floor of the Gothia Towers!


Conference day 2

The second day began with a Panel Debate moderated by Dipak Kalra. Tomas Andersson Julian Isla, Wim Goettsch, and Miriam Sturkenboom each gave brief presentations on the acceptance of Real World Evidence in decision making. They then debated the topic and responded to challenging questions from the audience. The areas discussed included the important patient groups, such as those with rare diseases, pregnant women and young children in whom classical RCTs cannot ethically be conducted, the difficulties and approaches to handling poor data quality and potential biases in the data including geographical variations in standards of care across health systems, the potential for the greater involvement of patients in enriching the available real world data and in evidence generation, and the lack of good examples of RWE impacting on decision making and how RWE can translate into accelerated availability of medicines and better patient outcomes. A fuller report of this panel debate will be published later.


Tomas Andersson, Miriam Sturkenboom, Wim Goettsch and Julian Isla.


The next session, chaired by Dipak Kalra, featured speakers focusing on the use of data for direct learning in healthcare, reflecting the patient, clinician, academic clinical research and healthcare system perspectives. Maximising patient outcomes was a running thread through this session.

Kaisa Immonen gave the first presentation, on bringing the value of health data to patients. We need the health system to stop disempowering patients; then they will naturally become empowered. She discussed the importance of data to patients. We need to make use of patient data and acknowledge its relevance. Rosna Mortuza spoke of improving health outcomes through health data. ICHOM are promoting a culture of adoption of outcomes data sets. Through outcomes measurement we can catalyse the transformation to value based health care. Stephanie Steels presented the evaluation of a learning health care system in action: Connected Health Cities in the North of England. This programme was established to deliver better health outcomes for patients and communities by analysing routinely captured health data and introducing new evidence-based care pathways: a learning health system. She described some key transferrable lessons (success factors): key transferable lessons: good governance structures, communication, patient and public involvement and multidisciplinary team work. Hanna Svensson, spoke about data management, a necessity for patients. She discussed being a patient with diabetes since she was young: self-care and healthcare need to synthesise their data for better health outcomes. Currently the patient experiences a lack of co-operation and communication. Healthcare needs to listen to patients and patients need to be in the driver’s seat. Juuso Blomster presented the experience and practical examples from utilising electronic health records in academic research at the Turku University Hospital. He illustrated how EHR data can be extracted directly into a research database and used to run a clinical study.


Hanna Svensson, Kaisa Immonen, Juuso Blomster, Rosna Mortuza and Stephanie Steels


The presenters from this session were involved in a panel discussion about how we can engage and serve patients through co-creation of value from their data and grow their trust and confidence. The discussion included how to best enable the empowerment of patients through their data, the importance of nurturing leadership and skills amongst patients, the co-design of patient relevant outcomes and the need to refocus information systems in healthcare to collect the most relevant data to achieve those outcomes.

Chuck Friedman, Professor of Learning Health Sciences, University of Michigan; Editor-in-Chief, Learning Health Systems Journal
Keynote, Chuck Friedman was unable to travel at the last minute, but joined us remotely to give his dynamic keynote presentation on the global landscape of Learning Health Systems. A health system that can study, learn and improve is the goal: every participating person’s characteristics and experience are available to learn from. Best practice knowledge is immediately available to support decision making. Improvement is continuous through ongoing study.

An infrastructure needs to enable this to happen at scale. Chuck conveyed how the LHS is now going global, with initiatives at varying scales from single organisations through to international organisations.

We then followed with presentations about i~HD taskforces at work.

Nathan Lea, GDPR Task Force lead, spoke of the goals for i~HD when ensuring public trust for reusing EHRs. Everyone is talking about the GDPR, and there are uncertainties when it comes to reusing health data for research. i~HD launched the GDPR Task Force earlier this year, offering GDPR tutorials, and will next launch the i~HD GDPR Observatory to monitor national adoption experiences.

Robert vander Stichele spoke on using quality indicators of pharmacotherapy for checking the quality of clinical data for secondary use: data quality with regards to the prescription. There is a high prevalence of potentially inappropriate prescriptions in community dwellings of older adults, such as nursing homes. Appropriate prescribing indicators can be used to identify polypharmacy, and to detect underuse and misuse of drugs, provided that the EHRs have good data quality.

Presenting the data quality assessment with the team in Barcelona and Valencia. Pascal Coorevits (i~HD Data Quality Task Force lead), Miguel-Angel Mayer, Marta Durá-Hernández, Carlos Sáez gave a group presentation. They discussed the assessment of data quality at a hospital in Barcelona. The assessments included presence, value range checking and multi-variant rules. The hospital results pointed to specific kinds of data that were most often incomplete, identified areas lacking correctness and consistency, and uncovered the data quality impact of a new automated coding system introduced in emergencies on a certain date.

Pascal Coorevits i~HD Data Quality Task Force lead, announced to the audience the launch of the i~HD Data Quality Service for Hospitals (DQS4H). This service is unique. It is a pragmatic, evidence based, flexible, focussed, staged, holistic and extendable data quality service. Pascal was pleased to announce that this service is now available to any European hospital.


Pascal Coorevits


“Wouldn’t it be great to have great data and to make great use of data?”


Dipak Kalra concluded the conference and spoke of the future plans of i~HD, which include communicating the value of health data to society. This is centred around a new EFPIA project, with the EPF as the anchor: i~HD will compile the case studies and other facts about the reuses of data that have been done in trustworthy ways.

No matching content found.

Published on: 30 November 2018