Dl. Fairclough et al., COMPARISON OF SEVERAL MODEL-BASED METHODS FOR ANALYZING INCOMPLETE QUALITY-OF-LIFE DATA IN CANCER CLINICAL-TRIALS, Statistics in medicine, 17(5-7), 1998, pp. 781-796
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.