Key points are not available for this paper at this time.
Abstract Prognostic factor evaluation is a frequent and desirable investigative activity in clinical trials. Nevertheless, a number of errors in the evaluation of such factors occur quite commonly. Many of these problems are due to inappropriate statistical interpretation of the analyses that are conducted. A detailed discussion of several problem areas is given. In the area of childhood ALL trials, issues related to the appropriate choice of disease end‐points for analysis of prognostic factors are discussed.
Harland N. Sather (Wed,) studied this question.