In the design of clinical trials, a clear definition of disease is essential for enrollment of a homogeneous study population with a higher likelihood of demonstrating a benefit of an intervention. A definition that is applicable to standard clinical practice enhances the ability of clinicians to apply results of the clinical trial to patient care. Use of a universally accepted definition allows valid comparisons across multiple studies. Sepsis, the acute respiratory distress syndrome, and ventilator-associated pneumonia are examples of conditions for which universal definitions developed by panels of experts have facilitated the design of successful clinical trials. However, implementation of the results of some of these studies has been complicated by a lack of understanding or acceptance of disease definitions or by their overly inclusive nature. For example, the presence of Systemic Inflammatory Response Syndrome (SIRS) will identify most patients with sepsis, however, a significant number of patients with those clinical findings will have other underlying processes. Approved definitions for VAP are cumbersome, and adherence to those definitions in the design of clinical trials is poor. This has led to confusion regarding the accuracy of diagnostic tests and poor acceptance of evidence based guidelines by clinicians. When investigators and clinicians do not adhere to common definitions of disease, results of clinical trials may be applied inappropriately or ignored altogether. More specific identifiers of critical illnesses using specific biochemical or genetic markers are being explored. This approach may also be useful for staging disease.