ABSTRACT
A new endpoint for clinical trials, the time without symptoms of disease (TWiST), was defined and heralded the integration of quality-of-life into primary endpoints for clinical trials. This is particularly important for evaluating therapies which lead to only small survival differences, but have markedly different toxicities. The method consisted of subtracting time with toxicity from time to relapse in making comparisons. Subsequent work looked at survival (Goldhirsch et al. 1989) and gave non-zero coefficients (in fact 0.5 in this paper) to the toxicity time and also the time after relapse when making comparisons. This led to the so-called Q-TWiST analysis. Actuarial methods were extended to analyse this type of data.
