Statistical Modelling of Intensive Care Data
Sylvie Chevret, Matthieu Resche–Rigon and Romain Pirracchio, Paris Diderot University, France
Presents both basic and advanced methods for statistical modeling of ICU data
The intensive care unit (ICU) is one of the major components of the current health care system. The complex tasks of collecting and analyzing data on performance measures are made easier when clinical information systems are available, but few clinical information systems have attempted to gather clinical information generating full reports that provide a panorama of the ICU performance and detailed data on several domains.
This book presents statistical methods – with special focus on including innovative approaches – that allow handling the specificities of ICU data, enabling practitioners to conduct appropriate analyses of their own data.
Statistical Modelling of Intensive Care Data:
- Covers the various clinical endpoints used in the ICU, notably the different measures of mortality, of durations (lengths of stay, durations of support, etc.), as well as the modelling of severity scores or biological measures over time.
- Deals with data from observational nonrandomized studies, including cohorts that are often used in ICU or assessment of care that could not be easily randomly assigned (such as ICU care itself).
- Considers performance measures in ICU, such as the centre effect and the volume effect, which are generally used to compare ICUs.
Organized in distinct sections which will provide both introduction and advanced understanding according to the level of the reader, this book will prove a valuable resource to either statisticians involved in ICU studies, or ICU physicians who need to model statistical data.