Quality of Life Research

, Volume 1, Issue 3, pp 219–224

Statistical analysis of longitudinal quality of life data with missing measurements

Authors

  • A. H. Zwinderman
    • Department of Medical StatisticsUniversity of Leiden
Statistical Review

DOI: 10.1007/BF00635621

Cite this article as:
Zwinderman, A.H. Qual Life Res (1992) 1: 219. doi:10.1007/BF00635621

Abstract

The statistical analysis of longitudinal quality of life data in the presence of missing data is discussed. In cancer trials missing data are generated due to the fact that patients die, drop out, or are censored. These missing data are problematic in the monitoring of the quality of life during the trial. However, by means of assuming that the cause of the missing data lies in the observed history of the patients and not in their unobserved future, the missing data are ignorable. Consequently, all available data can be used to estimate quality of life change patterns with time. The computations that are required are illustrated with real quality of life data and three commonly used computer packages for statistical analysis.

Key words

Longitudinal analysismissing dataquality of life

Copyright information

© Rapid Communications of Oxford Ltd 1992