Serial assessments of Karnofsky performance status (KPS) and of the Functional Living Index--Cancer (FLIC) have been used to estimate the quality of life (QOL) of patients in a prospective, randomized trial of supportive care versus supportive care given with combination chemotherapy to patients with metastatic non-small cell lung cancer. There was a good correlation between KPS and FLIC scores at study entry, thus confirming results originally reported using the FLIC. However, a number of unexpected problems were encountered in data collection and quality control with this QOL assessment instrument. This made it impossible to look for differences in treatment effect on QOL in this clinical trial. It is believed that QOL assessment should be an integral part of cancer clinical trials; however, investigators must acknowledge the difficulties in collecting this type of data. The development of new instruments and the refinement of old ones will facilitate the collection of data for this important aspect of clinical trials research.