In epidemiological research, large datasets are essential to reliably capture small variations among comparative groups or detect new unsuspected associations. Although large databases of web-search information, social media, airline traffic and telephone records are already widely used to capture social trends, large databases in medical research are just emerging. With the universal use of electronic medical records underway, vast amounts of health-related information will become available for biomedical research. Accepting such new research tools - based on the analysis of large pre-existing datasets rather than hypothesis-driven, in-depth prospective study - will require a new mindset in clinical research, as data might be 'messy' and only associations, but not causality, can be detected. In spite of such limitations, the utilization of these new resources for medical research harbours great potential for advancing knowledge about digestive diseases.
ASJC Scopus subject areas