Real World Evidence
As the name itself suggests, the definition “real world evidence” (RWE) indicates scientific evidence derived from the real world, i.e. the output of analysis of data coming from a patient’s daily life (“real world data”, RWD).
RWD are originated from numerous sources, among which electronic health records, administrative logs (e.g. for reimbursements or activity monitoring), pathology registers and non-clinical sources capable of providing information on the state of health (e.g., mobile and/or wearable devices).
An RWE-based approach favors research from several points of view, improving the planning of processes from the clinical, management and financial point of view.
An example of the above described situation comes from the issues stemming from inadequate planning, which affect clinical trials causing delays- or even, in extreme cases, the failure of the set objectives.
- 11% of the involved investigators enroll no patient at all
- 48% of the investigators underperform with respect to the trial’s requirements (low number of enrolled patients)
- 80% of the enrolment timelines for Phase II-IV studies are not met
- As a consequence, an extension of the timelines (sometimes even requiring twice the initially estimated time) to allow meeting the enrolment quota
- On average, about half the patients get to the completion of the study
It is clear how the management of these unforeseen situations affects not only the scientific aspect of a trial, but also the financial one, by significantly impacting the size of the investment necessary to meet the set goals.
A data driven approach to the feasibility analysis for trials allows the execution of estimations that are more precise and helps containing the above listed issues.
This analysis (patients eligibility assessment) is traditionally carried out through surveys for potential investigators. Instruments exist that allow the optimization of the evaluation process.
The availability of research data warehouses (DWH)- i.e. an aggregator of data originating from different sources- facilitates a fast, accurate and reliable assessment.
The creation of DWH networks based on a common data model (CDM) bringing together different centers favors a federated approach to research: the sharing of aggregated data and the implementation of semi-automated procedures that can work in all the nodes of the network
The exploitation of standard terminologies, the use of a common data model and the alignment of the data sources to the FAIR principles, all of which are covered in other insights on this site, are essential elements to ensure interoperability within a federated infrastructure.
