Efficiently generating medical insights – Advanced analytics on longitudinal health records
The increasing pressures faced by life sciences companies and the rise of big data have led to a new approach for the generation of medical evidence in recent years: mining longitudinal electronic medical records (EMRs).
This change is brought to life here with two recent IBM projects carried out with leading European life sciences companies.
In these projects, advanced analytics was applied to the IBM Explorys data set, which consists of real life data from millions of patients. Both examples illustrate the power of a novel method that allows medical evidence to be efficiently generated when mining EMRs: Agile Advanced Analytics (AAA). The technique helps pave the way towards personalized and outcome-based healthcare.
Generating medical evidence with advanced analytics has become essential.
The healthcare industry is in the midst of a value-and patient-driven transformation. Government and private payers are exerting pressure to restrain drug prices based on outcomes.
Subsequently, life sciences companies are increasingly being asked to justify their prices and to develop innovative models that take healthcare outcomes into account. At the same time, growing costs and increasing complexity are associated with the development of medical products. Recent estimates of bringing a molecular entity to market reach up to USD 2.6 billion and about half of Phase III drug candidates still have to be dropped.
As part of the medical development process, modern clinical trials have been conducted for decades in order to study the safety and efficacy of medical treatments.
However, in recent years a number of challenges typically associated with clinical trials have been exacerbated:
Rising trial costs: Research has identified increases in the cost of trials that impact the cost of bringing new products to market.
Discrepancies between trials and real life: Ensuring that the conditions of a clinical trial are representative of those encountered in real life are challenging.
Lack of exploratory design: Often design for purpose is in the foreground, hampering the ability to identify unmet medical needs. […]