Making evidence available: prognostic classification systems

Claims of treatment selectivity can represent an important therapeutic advance: however, most such claims cannot be confirmed with independent data. This implies a management dilemma when confronted with ambiguous information on the risks of therapies and, most of all, insufficient criteria for identifying patients at risk. An accurate assessment of prognosis can help overcome the uncertainties facing the patients. But the identification of prognostic factors is subject to certain caveats. An essential part of prognostic factor analysis is standardization, a fact that is often overlooked. Thus, two different studies identifying the prognostic value of a factor may not be truly comparable, making a meaningful comparison difficult. Therefore, the development of an accurate prognostic system provides a method of conveying clinical experience to others without ambiguity.