Allan Garland, Hayley B. Gershengorn, Ruth Ann Marrie, Nadia Reider, M. Elizabeth Wilcox

Various data sources can be used to conduct research on critical illness and intensive care unit (ICU) use. Most published studies derive from randomized controlled trials, large-scale clinical databases, or retrospective chart reviews. However, few investigators have access to such data sources, or possess the resources to create them. Hospital administrative data, also called health claims data, constitute an important alternative data source which can be used to address a broad range of research questions, including many that would be difficult to study in interventional studies. Such data often contain information that allows identification of ICU care, specific types of critical illness, and ICU-related procedures. The strengths of using administrative data are that some datasets are population-based, cover broad geographic regions, and are large enough to provide high statistical power and precise effect estimates. Linking hospital data to other databases regarding chronic care facilities, home care services or rehabilitation services, for example, can expand the scope of research questions that can be answered. However the limitations of administrative data must be recognized. They are not collected for research purposes, thus data elements may vary in accuracy, and key clinical variables such as ICU-specific physiologic and laboratory data are lacking. Specific efforts should be made to validate the data elements used, as has been done in several world regions. As with any other research question, it is imperative that the analysis plan be carefully defined in advance, and that appropriate attention be paid to potential sources of bias and confounding.