Problems in Assessing Quality of Life (QL) of Lung Cancer Patients in Clinical Trials

Abstract

If we assume acceptable quality of data, the best method of integrating QL data in treatment evaluation has yet to be defined. For a variety of reasons, multiple cross-sectional comparisons reflect only very grossly what is actually going on. Intraindividual analysis by using the patient as his own control is mandatory. Methods for analyzing longitudinal data, ie, defining patterns or clusters of psychosocial reactions with respect to treatment and disease related variables, have yet to be developed. For this purpose at the current state, descriptive analysis is more appropriate than hypothesis testing based on dubious assumptions. Every scoring system for a comprehensive treatment evaluation is based on arbitrary values and weighing of the observer. These have to be made explicit in reporting of results.

Integration of the patient's quality of life in treatment evaluation in clinical trials is necessary, but its assessment and analysis are at a pioneer stage. Instruments and analysis have to be refined, and cooperation and coordination have to be improved. It has become fashionable to assess quality of life in clinical trials. In our opinion at this time it is more promising to do only a few but carefully designed and meticulously planned studies.