Real World Data analytics in health

There is a great opportunity in the exploitation of healthcare data to foster translational research, drug discovery and to improve patient care. By exploiting patient level data such as the one available in EHR and clinical registries (“real world data”, RWD) we aim at discovering disease comorbidity patterns and the temporal evolution of diseases (a.k.a. “disease trajectories”). We apply different statistical and machine learning approaches in order to discover patterns. In addition, we also take into account the medications used as therapies, in order to discover if the appearance of a disease can be attributed to a drug treatment (drug side effects). We compare the results obtained by mining RWD with the ones obtained with molecular information (by analyzing DisGeNET, PsyGeNET and other sources of omics data) in order to identify those disease associations that could be explained by shared genetics/molecular basis. Other application of RWD analytics is for the discovery of patient subgroups or patient stratification.