Clinical trials and survival curves: the shape of things to come

• Published in: Acta haematologica Vol. 105; no. 3; p. 188
• Main Author: Elfenbein, G J
• Format: Journal Article
• Language: English
• Published: Switzerland 01.01.2001
• Subjects: Clinical Trials as Topic - mortality; Clinical Trials as Topic - trends; Humans; Neoplasms - mortality; Neoplasms - therapy; Statistics as Topic; Survival Rate
• ISSN: 0001-5792
• DOI: 10.1159/000046563

Clinical trials remain the standard for introducing new treatments into the medical armamentarium. There is a wealth of information contained in survival curves that are the ultimate outcomes reported in most oncology clinical trials. Survival curves report interesting information about the disease and its response to treatment. Unfortunately, and all too often, the limitations of survival curves are not adequately presented in publications and alternative interpretations for the data are not meticulously delineated by authors. The danger inherent in any publication of survival is that the data are not sufficiently 'mature' to support the conclusions that are drawn. This would be of little consequence if medical decisions were not based upon reading the latest publication that is acclaimed to be definitive and settle the question 'once and for all.' Better understanding of survival curves and how they shape the future of clinical practice may reduce the dangers of these pitfalls.