Challenges for learning health systems
- Speaker: Dr Niels Peek (Health eResearch Centre)
- Host: Goran Nenadic
- 18th November 2015 at 14:00 in Kilburn L.T. 1.4
Health systems are under pressure to deliver better care for more people from fewer resources. The worldwide economic crisis has shrunk the resources available for healthcare but the growth in demand for care services continues unabated. There is an opportunity to create smarter, "learning" health systems by utilizing the information infrastructure provided by electronic health records (EHR) systems. The data this is collected through this infrastructure can be used to efficiently evaluate healthcare services in real time, compare the effectiveness of different treatments in real-world populations, and discover novel disease subtypes for high-resolution medicine. Moreover, EHR systems can be used as a delivery platform to give feedback and advice to clinicians at the point of care. However, there are many challenges that need to be tackled before we can actually realise this vision. The data that are routinely collected in EHRs are a by-product of healthcare: Making sense of these data requires new tools that reach beyond traditional analytical concepts. Furthermore, clinical computerised decision support systems have mostly led to "alert fatigue" rather than improving care. In this presentation I will describe the current state of the art in this field, and address existing scientific, analytic and engineering challenges.