Quality of life as a clinical trial endpoint: Determining the appropriate interval for repeated assessments in patients with advanced lung cancer

One of the unresolved design issues for clinical trials with quality of life (QOL) as an endpoint is the frequency of measurement in patients with stage III and IV lung cancer. In a retrospective review of clinical trials, the QOL interval varied widely from 1 to 12 weeks during treatment. During follow-up, the interval was generally 2 to 3 months or not at all. The purpose of this methodological study was to determine an appropriate interval for QOL serial measurement based on prospectively collected data. The 20 patients for this study were part of a phase I/II study using combination chemotherapy in patients with advanced non-small cell lung cancer (NSCLC). They were typical of patients in lung cancer clinical trials, with a median age of 67 (interquartile range: 58, 72) years, the majority were male (13, 65%), and a baseline median Karnofsky performance status was 80 (interquartile range: 70%, 90%). The primary instrument, developed in 1985, was the Lung Cancer Symptom Scale (LCSS) patient form, a 9-item self-report and site-specific QOL measure. The method, outcome, and implication of these findings to research are presented for establishing a method for obtaining an appropriate serial measurement interval for QOL during therapy in clinical trials. Based on the findings of this study, an every 3-week QOL assessment for patients with advanced NSCLC provides data similar to more frequent evaluation (94% of data preserved compared to twice-weekly assessment, 95% confidence interval, 86-98%, p=0.05). Less frequent assessment (every 4 or every 6 weeks) retained less than 85% of the data, which is the recommended minimum adequacy rate. Retaining a high percentage of QOL information may lessen the effect of measurement bias due to patient attrition and may give more validity to QOL studies.