Quality of life studies of the Australian New Zealand Breast Cancer Trials Group: approaches to missing data.

The purpose of this paper is to describe the degree of compliance with quality of life measures in two clinical trials conducted by the Australian New Zealand Breast Cancer Trials Group comparing different chemotherapy policies for metastatic breast cancer. Quality of life was assessed by the patient using linear analogue scales and by the physician using the Spitzer QLI. Compliance was generally good, ranging from 66 per cent to 79 per cent in the earlier study, and from 63 per cent to 97 per cent in the later study. Compliance with physician rated quality of life was consistently slightly better than for patient self-assessment. The results of physician and patient assessments were generally consistent, but there was a systematic bias toward lower quality of life (as assessed by the physician) in patients who failed to comply with self-assessment. Our conclusions were that quality of life can be assessed in large scale multi-institution clinical trials in metastatic breast cancer. The results are important in assessing treatment comparisons. Missing data cannot be assumed to be similar to those available. Optimal assessment of quality of life therefore requires careful prospective attention to complete data collection.