Since the publication of two landmarks reports by the Institute of Medicine 2000 and 2001, quality improvement and patient safety have dominated the healthcare research agenda. The reports led to wide awareness that preventable medical errors are responsible for substantial numbers of adverse events in routine care, leading to increased costs and poor efficiency of the healthcare system. In the U.S. between 44,000 to 98,000 people die each year as a result
of preventable medical errors. This is more than the numbers of deaths caused by motor vehicle crashes (43,000) and by incidents involving firearms (20,000). The Institute of Medicine attributed these errors not to failures of individual healthcare professionals, but to inabilities of the healthcare system to manage the growing complexity of care, inadequate translation of knowledge into practice, and failures to apply new technology safely and appropriately. The Institute of Medicine called for a comprehensive effort by researchers, healthcare providers, governments, and care consumers, to reduce the number of errors by 50% within five years. Similar targets for improvement were formulated in many other countries. Although not officially quantified, it is broadly agreed that these ambitious goals have yet to be met.
Over the past few years, there has been increasing interest in, and evidence, of using intelligent data analysis methods in patient safety research and practice. A number of methods have been developed for identifying adverse drug events in post-marketing surveillance as well as near-real time detection of such events in spontaneous report databases. However, there is an urgent need to new approaches that address the limitations in such data, especially the lack of robust denominators, reporting accuracy and data quality. These analytic methods should address ongoing clinical needs, such as pharmacovigilance, real-time drug interaction detection, and provider performance, guideline compliance, and safety audits conducted in a variety of patient care settings. In addition, there is a pressing need in the clinical research domain for these analytic methods, especially with the increasing interest in distributed research networks and other approaches to sharing clinical data on a wide scale. Of particular interest is the detection of adverse events that may be rare or unexpected. IDAMAP 2011 will provide the opportunity for researchers and developers to present their work which addresses these critical issues.
The IDAMAP workshop series is devoted to computational methods for data analysis in medicine, biology and pharmacology that present results of analysis in the form communicable to domain experts and that somehow exploit knowledge of the problem domain. Such knowledge may be available at different stages of the data-analysis and model-building process. Typical methods include data visualization, data exploration, machine learning, and data mining.
Gathering in an informal setting, workshop participants will have the opportunity to meet and discuss selected topics in an atmosphere which fosters the active exchange of ideas among researchers and practitioners. The workshop is intended to be a genuinely interactive event and not a mini-conference, thus ample time will be allotted for general discussion.
The IDAMAP workshops are organized in collaboration with the working group on Intelligent Data Analysis & Data Mining of the International Medical Informatics Association, and the working group on Knowledge Discovery & Data Mining of the American Medical Informatics Association.
A selection of revised and expanded papers from each workshop will appear in the Methods of Information in Medicine journal.
Niels Peek, University of Amsterdam, The Netherlands
John Holmes, University of Pennsylvania, Philadelphia, PA, USA
Allan Tucker, Brunel University, London, UK
Riccardo Bellazzi, University of Pavia, Italy