14-15 November, 2016: The European Medicines Agency invited experts from the European Union and the USA to discuss 5 key perspectives to speak on Big Data at a workshop in London aimed at identifying opportunities from 'Real World and other "Big Data" to improve development of new medicines and surveillance of licensed medicines for safety, risk and effectiveness.
The 5 key 'stakeholder' perspectives? Patients and the public, health professionals, academia, regulators and policy makers, industry (both health sector and software/hardware) and payers (considering a change in strategy to payment for health impact rather than sales, as exemplified by the Health Impact Fund).
A pragmatic definition from Lu and his colleagues states that "Big data analytics (BDA) applications are a new category of software applications that process large amounts of data using scalable parallel processing infrastructure to obtain hidden value." There are many potential applications from planning for public transport flows to using large health record datasets to improve patient safety such as in the US FDA-Harvard Sentinel partnership.
The FAIR principles for Big Data, Finding, Accessing, Interoperability and Reuse of Big Data, have both general and special challenges and potential benefits when applied to healthcare.
For example, in a recent issue of Nature Reviews Cardiology, Rumsfeld and colleagues from Colorado and Boston outline 8 potential applications of big data analytics to improve cardiovascular care, including "predictive modelling for risk and resource use, population management, drug and medical device safety surveillance, disease and treatment heterogeneity, precision medicine and clinical decision support, quality of care and performance measurement, and public health and research applications".
The EMA note: "Rapid developments in technology have led to the generation of vast volumes of data, which have the capability to transform the way the benefit-risk of medicinal products is assessed over their entire life cycle. However, it is recognised there are multiple challenges in the exploitation of these data.
"These range from the fundamental need to establish methods to enable the access to, integration and analysis of heterogeneous datasets to understanding the limitations in its use. Importantly, robust and transparent mechanisms to protect patient confidentiality are key to secure patient trust. It is important for the European Medicines Agency and the European Union medicines regulatory network to gather information on the latest developments in big data from the perspective of all stakeholders in order to identity how and when the multitude of data sources may contribute to medicinal product development, authorisation and post-marketing surveillance."
Some key themes:
- Patrick Ryan on which patients chose which treatments
- Sophie Louveaux discussing new EU regulation of data, meaningful consent and processing sensitive health data
- David Martin addressing challenges in Big Data analytics from FDA and PPP perspectives
- Julian Isla from the Dravet patient charity on making the patient the centre in digital health
- Baroness Helene Hayman on ethics, governance and public confidence
- Ronald Brand from the University of Leiden on informed consent v. opt out
- Nicolas Tatonetti from Columbia University, NY on data mining for medical discovery
- Nico Gaviola from Google on cloud data for safer medicines
See more on key threads and discussion points including on the European Open Science Cloud, new EU General Data Protection Regulations - from May 2018, replacing Directive 95/46, machine-learning for chemogenomics, challenges to implementing applications to precision medicines, access to the OHDSI community, social media to find new adverse drug event signals, FDA case studies using then Sentinel-HMO-Harvard collaboration, opening access to the 28 EU independent national health care systems and more in due course when talks are made available on the EMA website for public access.
The 5 key 'stakeholder' perspectives? Patients and the public, health professionals, academia, regulators and policy makers, industry (both health sector and software/hardware) and payers (considering a change in strategy to payment for health impact rather than sales, as exemplified by the Health Impact Fund).A pragmatic definition from Lu and his colleagues states that "Big data analytics (BDA) applications are a new category of software applications that process large amounts of data using scalable parallel processing infrastructure to obtain hidden value." There are many potential applications from planning for public transport flows to using large health record datasets to improve patient safety such as in the US FDA-Harvard Sentinel partnership.
The FAIR principles for Big Data, Finding, Accessing, Interoperability and Reuse of Big Data, have both general and special challenges and potential benefits when applied to healthcare.
For example, in a recent issue of Nature Reviews Cardiology, Rumsfeld and colleagues from Colorado and Boston outline 8 potential applications of big data analytics to improve cardiovascular care, including "predictive modelling for risk and resource use, population management, drug and medical device safety surveillance, disease and treatment heterogeneity, precision medicine and clinical decision support, quality of care and performance measurement, and public health and research applications".
The EMA note: "Rapid developments in technology have led to the generation of vast volumes of data, which have the capability to transform the way the benefit-risk of medicinal products is assessed over their entire life cycle. However, it is recognised there are multiple challenges in the exploitation of these data.
"These range from the fundamental need to establish methods to enable the access to, integration and analysis of heterogeneous datasets to understanding the limitations in its use. Importantly, robust and transparent mechanisms to protect patient confidentiality are key to secure patient trust. It is important for the European Medicines Agency and the European Union medicines regulatory network to gather information on the latest developments in big data from the perspective of all stakeholders in order to identity how and when the multitude of data sources may contribute to medicinal product development, authorisation and post-marketing surveillance."
Some key themes:
- Patrick Ryan on which patients chose which treatments
- Sophie Louveaux discussing new EU regulation of data, meaningful consent and processing sensitive health data
- David Martin addressing challenges in Big Data analytics from FDA and PPP perspectives
- Julian Isla from the Dravet patient charity on making the patient the centre in digital health
- Baroness Helene Hayman on ethics, governance and public confidence
- Ronald Brand from the University of Leiden on informed consent v. opt out
- Nicolas Tatonetti from Columbia University, NY on data mining for medical discovery
- Nico Gaviola from Google on cloud data for safer medicines
See more on key threads and discussion points including on the European Open Science Cloud, new EU General Data Protection Regulations - from May 2018, replacing Directive 95/46, machine-learning for chemogenomics, challenges to implementing applications to precision medicines, access to the OHDSI community, social media to find new adverse drug event signals, FDA case studies using then Sentinel-HMO-Harvard collaboration, opening access to the 28 EU independent national health care systems and more in due course when talks are made available on the EMA website for public access.
