Determination of utility scores for control of chemotherapy-induced nausea or vomiting - CALGB 309801

Utility scores for temporary health states are not well defined. We investigated two techniques to define utilities for chemotherapy-induced nausea and vomiting. Patients receiving cyclic chemotherapy evaluated various hypothetical health states, each assuming 2-year overall survival with 1 year of chemotherapy followed by 1 year of good health. Health states included perfect health, no nausea/vomiting, limited nausea, limited vomiting, limited nausea/vomiting, and continuous nausea/vomiting. Subjects scored each health state with a rating scale and a standard gamble, using perfect health and continuous nausea/vomiting as positive and negative anchors. In addition, continuous nausea/vomiting was compared with a standard gamble between perfect health and immediate death. The study included 96 evaluable subjects. With perfect health and continuous nausea/vomiting anchors, rating scale scores for intermediate health states ranged from 43-84 whereas standard gamble scores clustered at approximately 55. However, using a standard gamble between perfect health and death, continuous nausea/vomiting had a utility score of 53. The decrement in utility for severe nausea/vomiting can be well characterized, but utility scores for less extreme health states of nausea/vomiting are not as distinct. Although the rating scale suggests some differentiation, the standard gamble reveals clustering of intermediate health state scores. Evaluation of the effect of varying duration and intensity of nausea/vomiting on standard gamble utility scores will provide greater insight into the impact of such temporary health states on quality of life. © 2009 Elsevier Inc. All rights reserved.

Duke Scholars

Author Kimberly Lynn Blackwell Medicine, Medical Oncology

Author James Emmett Herndon II Biostatistics & Bioinformatics, Division of Biostatistics