Are chemotherapy patients' HRQoL importance weights consistent with linear scoring rules? A stated-choice approach

Objective: To compare a linear scoring rule with the subjective importance of different domain and symptom levels of the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-C30 (EORTC QLQ-C30) among patients undergoing chemotherapy. Methods: Using a stated-choice or choice-format conjoint analysis survey instrument, we elicited patient preferences for varying levels of physical, role, social, emotional, and cognitive function along with chemotherapy-related side effects and financial difficulties. A total of 375 patients completed the questionnaire: 159 breast cancer, 117 colorectal cancer, 99 non-small-cell lung cancer; and 21 with unknown tumor type. Constrained maximum likelihood estimates were used to estimate relative importance weights for each level of each domain and symptom. Results: Summary HRQoL measures generally presume that differences among Likert categories are equally important to patients within and across outcomes. Our results indicate strong non-linearities both within and across domain and symptom categories. Improvements from severe pain to mild pain, severe fatigue to no fatigue, and severe social limitations to moderate social limitations are all about twice as important as no work to limited work in the Role domain. Conclusions: Our results indicate large differences in the impact of individual domains and symptoms on patient perceptions of well-being. Most cancer patients are likely to be less concerned about specific symptoms than the impact of those symptoms on their ability to function physically, socially, and in their daily roles.