Categorizing a prognostic variable: Review of methods, code for easy implementation and applications to decision-making about cancer treatments

Categorizing prognostic variables is essential for their use in clinical de cision-making. Often a single cutpoint that stratifies patients into high-r isk and low-risk categories is sought. These categories may be used for mak ing treatment recommendations, determining study eligibility, or to control for varying patient prognoses in the design of a clinical trial. Methods used to categorize variables include: biological determination (mos t desirable but often unavailable); arbitrary selection of a cutpoint at th e median value; graphical examination of the data for a threshold effect; a nd exploration of all observed values for the one which best separates the risk groups according to a chi-squared test. The last method, called the mi nimum p-value approach, involves multiple testing which inflates the type I error rates. Several methods for adjusting the inflated p-values have been proposed but remain infrequently used. Exploratory methods for categorization and the minimum p-value approach wit h its various p-value corrections are reviewed, and code for their easy imp lementation is provided. The combined use of these methods is recommended, and demonstrated in the context of two cancer-related examples which highli ght a variety of the issues involved in the categorization of prognostic va riables. Copyright (C) 2000 John Wiley & Sons, Ltd.