Statistical Test for Ordered Categorical Data in Clinical Trials
In clinical trials, subjects are often classified into ordered categories (eg, worsening, no change, and improvement) based on their post-treatment clinical response changes from baseline. The usual approach in using a multinomial model for assessing treatment effects between treatment groups is not efficient because it demands intense computation to evaluate the corresponding probabilities across these ordered categories. In this study, we propose to model the response probabilities via a parametric form, and the comparison of these probabilities is translated into the comparison of the associated model parameters. Maximum likelihood estimates are then derived and the required sample size for achieving a desired statistical power at a prespecified level of significance is also obtained. A simulation study is performed to evaluate finite sample size performance. An example concerning the evaluation of the efficacy of a test treatment for subjects with moderate to severe Crohn disease that is refractory to steroids and immunosuppressants is presented to illustrate the proposed method.
Key WordsClinical trials Likelihood ratio test Multinomial model Ordered categorical data Sample size determination
Unable to display preview. Download preview PDF.
- 3.Cohen A, Sackrowitz HB. Testing whether treatment is “better” than control with ordered categorical data: definitions and complete class theorem. Stat Decis. 2000;68:1–25.Google Scholar
- 4.McCullagh P. Regression models for ordinal data. J R Stat Soc B. 1980;42:109–142.Google Scholar
- 5.Silvapulle MJ, Sen PK. Constrained Statistical Inference. New York: Wiley; 2005.Google Scholar
- 6.Cohen A. One-sided tests for ordered categorical data. In: Chow SC, ed. Encyclopedia of Biopharmaceuticat Statistics, New York: Taylor and Francis: 2006.Google Scholar
- 7.Agresti A. Categorical Data Analysis. New York: Wiley; 1990.Google Scholar