COMPARISON OF SEVERAL MODEL-BASED METHODS FOR ANALYZING INCOMPLETE QUALITY-OF-LIFE DATA IN CANCER CLINICAL-TRIALS

This paper considers five methods of analysis of longitudinal assessme nt of health related quality of life (QOL) in two clinical trials of c ancer therapy. The primary difference in the two trials is the proport ion of participants who experience disease progression or death during the period of QOL assessments. The sensitivity of estimation of param eters and hypothesis tests to the potential bias as a consequence of t he assumptions of missing completely at random (MCAR), missing at rand om (MAR) and non-ignorable mechanisms are examined. The methods includ e complete case analysis (MCAR), mixed-effects models (MAR), a joint m ixed-effects and survival model and a pattern-mixture model. Complete case analysis overestimated QOL in both trials. In the adjuvant breast cancer trial, with 15 per cent disease progression, estimates were co nsistent across the remaining four methods. In the advanced non-small- cell lung cancer trial, with 35 per cent mortality, estimates were sen sitive to the missing data assumptions and methods of analysis. (C) 19 98 John Wiley & Sons, Ltd.