Discrete Data Analysis, Failure Time Data Analysis. Better Assessments of Biological and Pharmaceutical Agents

Introduction

The last decades have witnessed a dramatic improvement in the methods of drug evaluation, and, therefore, our ability to use biological or pharmaceutical agents which will benefit risk ratio, can be better assessed. While these changes have had an immense impact on the professional day to day lives of all those involved in human research, there are still growing expectations for education information and reflections in a more demanding environment. This chapter will review the statistical analysis of qualitative data, otherwise called discrete data, and summarizes for that purpose the 2015 lectures given to the master’s student European diploma of pharmaceutical medicine at the European College Pharmaceutical Medicine, Claude Bernard University Lyon France.

Clinical investigators, like one of the authors of the current work (TC), are, usually, involved in everyday clinical practice, and, in addition, in clinical research. The job of statisticians, like the other author of the current work (AZ), as chief statistics at his university department, is different. Apart from educational tasks, he is not only involved in research activities, but also in research lines, like the ones given underneath:

clinical epidemiology (cardiovascular diseases,familiar hypercholesterolemia, cardiogenetics)

population epidemiology (early exposure, ethnicity) epidemiology infectious diseases (hiv, tuberculosis, malaria) biomarker & test evaluation biostatistics (high dimensionality, causal effects) bioinformatics (knowledge bases, e-science, dna sequencing) systems medicine (mathematical modeling)

systematic review (Dutch Cochrane center, intervention, diagnostic test accuracy).

© Springer International Publishing Switzerland 2017

TJ. Cleophas, A.H. Zwinderman, Understanding Clinical Data Analysis,

DOI 10.1007/978-3-319-39586-9_5

The current chapter will, particularly, focus on discrete data and failure-time data, and will use for the purpose recently published global studies in the above fields of expertise. The analyses of quantitative data will be covered in the Chap. 6. Discrete data can answer many questions in clinical trials. Basic methods, but also relatively novel subjects will be addressed, like one sample tests for multiple crossovers such as the Cochrane’s Q tests, and the methods for assessing failure-time data analysis, otherwise called time-to-event analysis.