ABSTRACT
Incomplete data are inevitable in quality-of-life (QoL) studies involving cancer patients, since a proportion of patients will not complete the trial. Traditional analyses assume that such data are missing at random, but this assumption may be incorrect. We therefore applied mixed-effects statistical models to determine the effects of non-random missing data. Models were applied to results from a large-scale, randomised study of epoetin alfa versus placebo in 375 patients receiving non-platinum-based chemotherapy.
Notes
* Source: Fairclough DL, Gagnon DD, Zagari MJ, et al. Evaluation of quality of life in a clinical trial with non-random dropout: the effect of epoetin alfa in anaemic cancer patients. Qual Life Res. 2003;12:1013-27