Big Data and scientific research
Big Data has been called, in recent years, “an invaluable source of value,” the new gold, but merely collecting the data, while taking advantage of the best technologies available on the market, does not guarantee that it will yield valuable information and, more importantly, extract useful knowledge for advances in health.
Health data are only useful if they can be transformed into meaningful information, and this requires high-quality data sets, communications between IT systems, and standard data formats that can be easily processed.
The digitization of medicine promises great progress for global health; digital data is expected to transform medicine.
However, most of today’s medical data lacks interoperability: data are hidden in isolated silos and incompatible systems that are difficult to exchange, analyze and interpret. And this slows down medical progress because the technologies that rely on these data cannot be used to their full potential.
The use of interoperable formats for real-world data, Real World Data (RWD), patient health status data routinely collected in clinical practice or, increasingly, via mobile apps in patients’ daily lives, opens up various opportunities for researchers.
The digitization of medicine and the tremendous enhancement of computing capabilities have now made it possible to collect health data electronically, in large quantities, very quickly and at the same time that the service is delivered. It is therefore possible to use data from large numbers of subjects representative of the real population to describe the impact of care, to predict outcomes and response to various treatments at relatively low cost, because they are produced in the daily routine and not collected only from ad hoc studies.
Real World Data (RWD) are collected in routine clinical practice through a variety of sources such as electronic health records, administrative flows related to hospitalizations, drug prescriptions or outpatient services, drug and disease registries, data collected from other sources such as mobile devices and telehealth.
Real-world data are also valuable for artificial intelligence and learning methods.
They allow patterns and correlations to be found in high-dimensional datasets that can help researchers identify new research hypotheses that can later be studied in a more focused way in controlled clinical trials (these controlled studies will always remain important for ruling out bias and identifying causal relationships).
More generally, if health data are structured according to international standards, the data are much easier to analyze and the effort required to prepare the data for analysis is reduced, allowing the research process to be accelerated.
Similarly, interoperable data can ensure that an analysis is performed on different data sources, using data from different institutions or countries. This is very important for scientific research where it is often necessary to bring together multiple multidisciplinary research teams to achieve meaningful analysis and to pass the results into clinical practice.
In summary, interoperability can generate new medical knowledge by enabling more efficient analysis of existing data sources, promote evidence-based medical practices, and accelerate their implementation in public health policies.