Big Data – How to Realize the Promise

Abstract

The increasing volume and complexity of data now being captured across multiple settings and devices offers the opportunity to deliver a better characterization of diseases, treatments, and the performance of medicinal products in individual healthcare systems. Such data sources, commonly labeled as big data, are generally large, accumulating rapidly, and incorporate multiple data types and forms. Determining the acceptability of these data to support regulatory decisions demands an understanding of data provenance and quality in addition to confirming the validity of new approaches and methods for processing and analyzing these data. The Heads of Agencies and the European Medicines Agency Joint Big Data Taskforce was established to consider these issues from the regulatory perspective. This review reflects the thinking from its first phase and describes the big data landscape from a regulatory perspective and the challenges to be addressed in order that regulators can know when and how to have confidence in the evidence generated from big datasets.

From Real‐World Patient Data to Individualized Treatment Effects Using Machine Learning: Current and Future Methods to Address Underlying Challenges

“…14 However, it is important to not only make predictions from EHRs, but also understand what the best way is to act and give treatments. As also described by Beaulieu-Jones et al 9 and Cave et al, 15 this requires understanding causality and being able to quantify causal relationships using the available observational data. In this context, machine learning methods for causal inference can be used to learn the effects of treatments from observational data and subsequently provide clinicians with actionable intelligence for making treatment decisions.…”

“…We have yet to harness the full potential of observational data for clinical decision support. 15 Although the simple case of static, binary treatments has been well-studied in the causal inference literature, EHRs contain information about significantly more complex treatment scenarios. In order to design more robust and effective machine learning methods for personalized treatment recommendations, it is vital that we gain a deeper theoretical understanding of the challenges and limitations of modeling multiple treatment options, combinations, and treatment dosages from observational data-in both the cross-sectional and time-series settings.…”

Using RWE research to extend clinical trials in diabetes: An example with implications for the future

“…Another consideration is the growing availability of RWD, both in breadth and depth. More and more measurements about patient activities and status are available in electronic form, and this electronic data can be linked with others to form a data source that is customized for the needs of the research question . As data sources expand from the periodic capture of routine transactions with healthcare providers to continuous monitoring through connected devices and wearables, RWD improves in its representativeness and granularity, making it possible to study aspects of treatment on patients that were previously unimaginable.…”

Data Science in Clinical Pharmacology and Drug Development for Improving Health Outcomes in Patients

“…11 Given the range of opportunities, we must also understand the issues and challenges in how RWD should be used [12][13][14] and ensure that appropriate validation is undertaken for new methods. 15 RWD is also used commonly in the context of the explosion of data available from -omics, continuous, ambulatory (usually in the real world) patient-monitoring technology, including wearables and other high-capacity data capture and analytical methods (often referred to as Big Data). The potential of Big Data in drug development is of interest to regulators, 15 and clinical pharmacologists are well placed to guide integration of Big Data analyses with modeling 16 as a way to use, analyze, and interpret the data.…”

“…There are also many opportunities where RWD can benefit clinical pharmacologists, including streamlining or even replacing clinical trials in a few instances, informing on difficult‐to‐study populations, such as children, or rare diseases, drug repurposing, pharmacovigilance, pharmacokinetic/pharmacodynamic modeling, and physiologically‐based pharmacokinetic modeling . Given the range of opportunities, we must also understand the issues and challenges in how RWD should be used and ensure that appropriate validation is undertaken for new methods …”