Big data and outcomes: opportunities and challenges
Currently, decisions on treatments and pricing are based largely on data from clinical trials. This data is inevitably limited in scope and does not fully reflect the situation of all patients with a given disease. Yet immense amounts of data are generated daily by researchers as well as clinicians and patients themselves. If we could harness this ‘big data’, it could revolutionise research and healthcare and help us move towards more sustainable healthcare systems focused on outcomes for patients. However, delivering ‘big data for better outcomes’ is far from easy. On the big data side, bringing diverse data sources and formats together, linking them up and analysing them is far from easy due to technical, legal and ethical issues. On the outcomes side, identifying and defining outcomes for patients that are meaningful, measurable and relevant for all stakeholders in healthcare is also a challenge.
IMI’s Big Data for Better Outcomes (BD4BO) programme aims to address these challenges. It brings together all stakeholders, including patients, academic researchers, healthcare policy makers, regulators, healthcare providers, payers, and the pharmaceutical industry.
Overall coordination of the programme is the responsibility of the DO>IT project, a coordination and support action set up for this task. Leading the work on data integration is the EHDEN (European Health Data and Evidence Network) project.
Meanwhile, a series of projects focused on specific therapeutic areas are putting the concept of ‘big data for better outcomes’ into practice in the fields of cardiovascular disease, haematological malignancies (blood cancers), Alzheimer’s disease, and prostate cancer.
By making it easier to tap into ‘big data’, BD4BO will make research more efficient, as researchers will be able to re-use existing data instead of having to generate new data. The sheer volume of data involved will make it easier for researchers to uncover new insights into diseases and treatments, and accelerate the development of innovative, more effective medicines.
For healthcare systems, the tools and resources created by BD4BO, as well as the networks of experts in the projects, will make it easier to identify of which treatments work best (and which do not) for different groups of patients.
For patients, BD4BO will help to ensure that treatments are designed and selected on the basis of outcomes that matter to patients. It will also increase the likelihood of patients receiving a treatment that works for them.
For the pharmaceutical industry, BD4BO will supply greater knowledge of how patients experience their condition, and which outcomes really matter, allowing them to better target potential treatments and demonstrate efficacy in real-life conditions.
Ultimately, BD4BO will contribute to the evolution of healthcare systems, making them more sustainable and crucially, more focused on outcomes for patients.
The DO>IT project was launched to provide a coordination platform for the BD4BO programme, exploiting synergies across the projects and maximising its impact on healthcare systems. Specially, DO>IT will aggregate learnings and disseminate findings from the projects; develop minimum data privacy standards; engage with key stakeholders; and recommend areas for future collaborative research.
Alzheimer’s disease is on the rise in our ageing population, and new, effective treatments are urgently needed. ROADMAP aims to deliver methods and tools that will allow the scalable, transferable integration of data on patient outcomes in the real world. The tools will be developed and tested through pilot projects and will lay the foundations for a Europe-wide platform on real world evidence in Alzheimer’s disease. The project will also deliver tools for patient engagement and address the ethical, legal and social implications of adopting a real world evidence approach to Alzheimer’s disease.
HARMONY and HARMONY PLUS
Blood cancers, or haematologic cancers (e.g. leukaemia, lymphoma and myeloma), affect the production and function of blood cells and account for about one third of cancer cases in children and about one third of cancer deaths. As many blood cancers are rare, and healthcare practice varies across EU, a lack of data on relevant outcomes represents a challenge for clinicians, researchers, and decision-makers alike. HARMONY aims to use ‘big data’ to deliver information that will help to improve the care of patients with these diseases. More broadly, the project will result in a pan-European network of stakeholders with expertise in this disease area.
HARMONY PLUS builds on HARMONY's work by adding additional blood cancers to the list of diseases included in the initiative, and by converting the HARMONY platform into an integrated services platform that could serve clinicians to improve their decision making process and support clinical trial design, among other things.
Cardiovascular disease (CVD) is a major killer in Europe, accounting for 45% of all deaths. BigData@Heart focuses on three types of CVD. It aims to develop new definitions of diseases and outcomes; informatics platforms that link, visualise and harmonise different data sources; data science techniques; and guidelines on the cross-border use of big data resources. In the long term, the project expects to have an impact on our understanding of heart disease, the discovery of new targets for treatments, and progress towards personalised treatments for CVD.
Prostate cancer accounts for 9 % of all cancer deaths in men. Currently, it is hard to predict which patients will respond best to different treatments, and which patients can be managed safely without undergoing treatment. PIONEER is using big data to address key knowledge gaps related to the screening, diagnosis and treatment of prostate cancer patients. To do this, they will standardise and integrate existing ‘big data’ from sources such as clinical trials and electronic health records into a single, innovative data platform. The project will draw on this database to identify ways to improve prostate cancer outcomes and health system efficiency. Ultimately, the project results should feed back into clinical centres so that patients can benefit from the best possible care.
Healthcare data has the potential to transform our understanding of health, disease and outcomes, yet it is currently scattered across multiple institutions and countries, stored in different formats, and subject to different rules. This makes it very difficult to fully utilise this data to benefit patients. The goal of EHDEN is to make the large-scale analysis of health data in Europe a reality. The project aims to do this by building a federated data network of allowing access to the data of 100 million EU citizens standardised to a common data model. At the heart of the project will be a group of trained, certified small and medium-sized enterprises (SMEs) responsible for transforming the data owned by hospitals to the common data model. The data will remain under complete control of the original data owner, thereby ensuring ethical and local data privacy rules are respected.
Achievements & News
IMI projects are presenting results on remote clinical trials, patient information, health data, and real world evidence at the upcoming...
Prostate cancer is the second most common cancer in men and its diagnosis - and treatment - can take a...
The project hopes the CODE-EHR framework will boost confidence in the results of studies using real world data, and so...
IMI project HARMONY is proposing a new, more accurate way of assessing the prognosis of patients with multiple myeloma.
COVID-19 vaccines do not raise the risk of developing four rare neurological disorders, but an increased risk was seen in...
The EHDEN project has announced the creation of a new organisation that will support research carried out by partners in the EHDEN network and the wider research community beyond the lifetime of the IMI-funded project.###
The EHDEN project has a vision of the future wherein real-world data gathered during routine patient care can be used by the healthcare community to quickly generate high-quality insights that can improve care. Through a network of 98 data partners, the project has harmonised more than 400 million health records, transforming data from disparate databases to a ‘common data model’. This valuable resource has already been used to generate evidence on COVID19 vaccines and potential treatments, diabetes, and other indications.
The new organisation is being launched as a not-for-profit incorporated in the Netherlands, transitioning the project into a long-term sustainable operation from 2024. It will support studies into more diseases and new treatments through study-a-thons and research programmes, continue work on methodological and technical developments of the data and analytics infrastructure, ensure new data partners can join, and further grow the ‘EHDEN Academy’ training programme.
Find out more
- Read the article in full
Experts from IMI’s Big Data for Better Outcomes (BD4BO) programme have some recommendations on how to make the European Health Data Space really work for patients, doctors and researchers.###
The BD4BO programme was set up to explore new, improved and ethical ways for researchers to unleash the collective power of healthcare datasets, with the ultimate aim of improving real-life outcomes for patients.
In response to the European Commissions’ announcement of their intention to put in place a European Health Data Space (EHDS), experts from BD4BO have offered up the following recommendations based on their extensive experience working out the right tools and rules for harnessing and exploiting big data via a multi-themed portfolio of research projects.
- Showing results can build trust
- Use what already exists
- Getting everyone on board will require big investments
- Ensure an interoperable and flexible architecture
- Meaningful codes of conduct will strengthen citizens’ rights
Find out more
The beginning of the coronavirus pandemic led people to stay away from emergency departments (EDs) in the UK, recent data analysis suggests, including those with urgent heart problems. And this in turn may have led to a spike in mortality, according to a new study supported by IMI project BigData@Heart. ###
The study, which was published in the journal Circulation: Cardiovascular Quality and Outcomes, analysed data gathered between March and April 2020 from the UK’s government statistics body, as well as a public registry that tracks emergency department visits, and ran a script to estimate the number of excess deaths. They estimate that during the initial phase of the pandemic, there was a drop of about 2 750 visits per week (a 35% decrease) for suspected cardiac disease.
Lead author Dr Michail Katsoulis said: "Our analysis suggested that one cardiac death might have been prevented or delayed for every additional 12 ED visits for suspected cardiac conditions." They arrived at the conclusion that there may have been as many as 232 more deaths per week early in the pandemic, compared to the pre-pandemic period.
As for the reasons people stayed away, no doubt fear of infection played a role. On top of that, public health messaging focused on encouraging people to put as little pressure on the UK’s national health service as possible.
Find out more
- Read the article in full
Over 1 000 people have signed up to the EHDEN Academy, which provides the global scientific community with free educational resources on real world data and real world evidence.###
The aim of EHDEN is to make the large-scale analysis of health data in Europe a reality. At the heart of the project is a growing group of trained, certified small and medium-sized enterprises (SMEs) responsible for transforming data owned by hospitals, and others, to a common data model.
The training materials designed for these SMEs formed the basis of the EHDEN Academy, expanding to methods, skills and tools for working with real world data to generate real world evidence. The Academy was launched in April 2020 and has now attracted over 1 000 enrollees from 46 countries worldwide.
Today, the academy features 9 courses and a faculty of 10 expert instructors. The courses are freely available on demand and represent an important source of training for anyone who generates and uses data, works with it, or is involved in the development and use of analytical tools.
Find out more
- Read the article in full
EHDEN is offering to harmonise organisations’ clinical data to a standard model, while preserving patients’ privacy. This will make it easier to aggregate and jointly analyse data from different sources, something that is essential if we are to stop the outbreak and save lives.###
Since the start of the COVID-19 outbreak, hospitals and clinics around the world have captured vast amounts of data on the disease in patients’ healthcare records. This data, like most healthcare data, is stored in separate databases and different formats, and this drastically limits its usefulness for research. Now EHDEN is offering its expertise to organisations that have amassed data from COVID-19 patients and want help converting it to a standardised data format so that it can be used as part of wider studies on the disease. They have launched a ‘Call for data partners’ through which organisations with clinical data on COVID-19 can apply to benefit from this offer. Organisations have until 14 May to apply; a panel of bioinformatics experts will review the applications as they are received.
Successful applicants will receive a grant from EHDEN of up to EUR 50 000 to support the work they will need to carry out to assist with the data harmonisation process. They will also be invited to participate in COVID-19 research studies to accelerate our understanding of how to combat this pandemic and improve patient outcomes, as well as the ongoing study-a-thon.
‘We will provide financial and technical support to assist with harmonising your data to the OMOP common data model, with a view to linking Data Partners together and with researchers for quicker open science on SARS-CoV-2/COVID-19,’ the project says.
The first two datasets have been added to the PIONEER big data platform, a major milestone for the IMI project. The researchers want to use real-world data from well-known prostate cancer studies to answer some vital questions about the disease. ###The first question they hope the data will help answer concerns the kind of variables that affect the prognosis for prostate cancer patients.
The two datasets will be followed by more; other partners are poised to add another 22 anonymised datasets to the platform, and PIONEER hopes that other data custodians, both private and public, will be encouraged to contribute. The ultimate aim is to improve the health and social care for all prostate cancer patients and their families.
Analytics will be run over the lifetime of the project to account for new and updated datasets. Both the OMOP/OHDSI platform for population-based registries and epidemiological research and the PIONEER omics analysis platform for cohort and clinical trial data were set up by The Hyve, a partner in the project.
IMI’s big data projects will promote their activities at the BioDataWorld Congress 2019 in Basel, Switzerland, on 4-5 December 2019. The projects will share their approach to improving health outcomes### in the following sessions:
- The pro’s and con’s of the different IMI projects approaches. IT and high-tech aspects (Roundtable 27, 4 December at 11:40)
- The need for data communities in Europe, and why should you care? (Roundtable 25, 5 December at 11:40)
- It is now or never! Connecting data communities with clinical research communities (Lunchtime panel, 5 December at 12:30)
- Connecting the dots to benefit patients: This is how we do it, but we need you too! (IMI Showcase, 5 December at 15:10)
The projects participating in these sessions are: HARMONY (big data and blood cancers), EHDEN (health data and evidence network), BigData@Heart (big data and heart disease), PIONEER (big data and prostate cancer), and ROADMAP (big data and Alzheimer’s disease). All projects are part of IMI’s Big Data for Better Outcomes programme.
IMI’s EHDEN project has dramatically demonstrated the power of using clinical data in research by replicating, during a five-day ‘study-a-thon’, the results of a systematic review covering 20 years of research, and a multi-year clinical trial. The findings, which focus on the pros and cons of different types of knee replacement surgery, are published in the journal Lancet Rheumatology. ###They show that it is possible to harness clinical data (such as electronic health records) from different sources and use it to generate information that could help patients and doctors to make better decisions about their care.
‘Randomised controlled trials remain the gold standard for establishing efficacy,’ the EHDEN team concludes in the paper. ‘However, we feel that this study shows the value of real world evidence for complementing the evidence produced from randomised trials.’
‘This is particularly relevant for less common outcomes including safety and adverse reactions,’ adds one of the lead authors of the paper, Daniel Prieto-Alhambra of the University of Oxford, noting that the EHDEN team is keen to build on this experience. ‘Future study-a-thon’s are planned during the duration of EHDEN. Watch this space!’
IMI’s EHDEN project is launching a call for data custodians to apply for funding to map their health data to the Observational Medical Outcomes Partnership common data model (OMOP CDM).### EHDEN’s ambitious goal is to standardise more than 100 million patient records across Europe from different geographic areas and different data sources. Mapping this data to the OMOP CDM will facilitate their use for a variety of purposes, enhancing and accelerating research and healthcare decision-making for global benefit. In addition to funding, EHDEN data partners will benefit from being part of a thriving academic / medical network; more opportunities to participate in international studies; and improved interoperability and visibility of their data, among other things.
The call is currently in a draft form, and data custodians are invited to review the draft and provide feedback by 15 August. The EHDEN team will review feedback received and open the call for applications from 1 to 15 September.
IMI’s HARMONY project has captured data on 45 000 patients with blood cancers, meaning it is almost half way to achieving its goal of collecting data on at least 100 000 patients during the lifetime of the project. The data, which comes from multiple sources such as clinical trials and registries, is gathered in the project’s Big Data Platform. HARMONY researchers are already mining it to answer research questions such as whether one specific treatment improves outcomes in patients with aggressive multiple myeloma, and which subgroups of patients with myelodysplastic syndromes benefit from certain treatments. ‘Mining this unique Big Data Platform to address pressing research questions will give us better insight in the molecular landscape of blood cancers and the prognostic value of disease related variables, hence increasing our understanding of their pathophysiology. Making use of big data analytics in blood cancer research will lead us faster to identifying novel drug targets,’ says Bruno Costa of Celgene, representing the EFPIA members in HARMONY. ‘Ultimately, our goal is that, together, we can accelerate drug development, regulatory evaluation, access appraisal, and treatment strategies to improve the care of patients with these blood cancers.’
Partners in blood cancer alliance HARMONY have uploaded the first datasets into the HARMONY Big Data Platform. This is a major milestone for the large alliance that aims to improve the treatment of haematologic malignancies. This first batch of data focuses on acute myeloid leukaemia (AML), a form of blood cancer that is most common in the over 75s. ###There are effective treatments for AML, but many elderly and weaker patients are not strong enough to tolerate the intensive chemotherapy regimens. Improved insight into the molecular basis of the disease is essential to develop better treatments for these patients. This requires data from thousands of patients. Therefore, HARMONY is bringing together datasets from all over Europe. The datasets that were recently uploaded into the Big Data Platform are from the German Austrian AML Study Group (AMLSG); the Haemato Oncology Foundation for Adults in the Netherlands (HOVON); and Novartis’s RATIFY trial. More HARMONY Partners are ready to transfer their data to the platform as well. ‘With such positive efforts in sharing aggregated data of hundreds of patients, the hope is that other data custodians across industry and academia will also be inspired to join the HARMONY Alliance’, said HARMONY project lead Mirko Vukcevic from Novartis. ‘In the future, the aspiration is that the HARMONY Alliance model will allow us to create a blueprint that can be applied to future projects in other disease areas, ultimately benefiting many more patients in Europe and well beyond.’ HARMONY is a unique network of more than 80 public-private organizations. It is part of IMI’s Big Data for Better Outcomes programme.
IMI’s EHDEN project is looking for small and medium-sized enterprises (SMEs) to get involved in the project’s efforts to map and convert health data from diverse sources to a common model. The project has launched a pilot call for SMEs who want to get training and certification in this exciting new field. ###EHDEN’s ambitious goal is to standardise 100 million patient records across Europe covering different data source types, including hospitals, registries, and population databases. Transforming this data to the Observational Medical Outcomes Partnership (OMOP) common data model (CDM) is an immense task that EHDEN hopes will be carried out by a community of SMEs. SMEs selected through this pilot call will receive training in this and a certification that will highlight their expertise in this area. ‘We expect that those who receive training and certification through EHDEN will actually go on to perform mappings to help build the network,’ the project writes. ‘We also hope that they will become active members of the community, sharing ideas and possibly helping to develop or improve data mappings and analysis tools.’ According to the project, SMEs will also benefit from access to a larger market, the opportunity to build expertise in and contribute to a relatively new field, and the support of the wider EHDEN community. Finally, certified SMEs will receive invitations to additional training sessions as well as hackathons to improve the open source tools and other events.
- Details of how to apply can be found on the EHDEN website.
- Deadline for applications: 1 May 2019
DO>IT, the coordination project of the Big Data for Better Outcomes (BD4BO) programme, has delivered a toolkit to support the other BD4BO projects in the identification, selection and measurement of outcomes. The BD4BO programme currently has projects focusing on Alzheimer’s disease, blood cancers, prostate cancer, and heart disease. ###The new toolkit represents a practical guide which will help the projects to adopt a standardised approach when developing core outcome sets (COS) in their disease areas. The toolkit proposes six main stages for developing a COS, from scoping to dissemination, with a focus on stakeholder input across all stages to ensure a wide range of perspectives are taken into account. Whilst the toolkit highlights any existing best practice for developing COS, it also presents a range of methodological options which BD4BO projects can consider depending on the scope of the work and resources available. Each stage includes decision-making flowcharts, summaries of key considerations and case studies to highlight the key factors and considerations when developing COS. These typically reflect aspects that are of importance to BD4BO projects around the use of data from a range of sources from ‘real world' settings in addition to clinical trials.
Details of all project participants can be found on the individual project factsheets.