A quality-of-life-oriented endpoint for comparing therapies

R. D. Gelber, R. S. Gelman, A. Goldhirsch

Research output: Contribution to journalArticlepeer-review


An endpoint, time without symptoms of disease and toxicity of treatment (TWiST), is defined to provide a single measure of length and quality of survival. Time with subjective side effects of treatment and time with unpleasant symptoms of disease are subtracted from overall survival time to calculate TWiST for each patient. The purpose of this paper is to describe the construction of this endpoint, and to elaborate on its interpretation for patient care decision-making. Estimating the distribution of TWiST using actuarial methods is shown by simulation studies to be biased as a result of induced dependency between TWiST and its censoring distribution. Considering the distribution of TWiST accumulated within a specified time from start of therapy, L, allows one to reduce this bias by substituting estimated TWiST for censored values and provides a method to evaluate the 'payback' period for early toxic effects. Quantile distance plots provide graphical representations for treatment comparisons. The analysis of Ludwig Trial III evaluating toxic adjuvant therapies versus a no-treatment control group for postmenopausal women with node-positive breast cancer illustrates the methodology.

Original languageEnglish
Pages (from-to)781-795
Number of pages15
Issue number3
Publication statusPublished - 1989

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Agricultural and Biological Sciences (miscellaneous)
  • Applied Mathematics
  • Statistics and Probability
  • Public Health, Environmental and Occupational Health


Dive into the research topics of 'A quality-of-life-oriented endpoint for comparing therapies'. Together they form a unique fingerprint.

Cite this