The Discipline of Statistics

The discipline of Statistics is a well-developed and powerful discipline that involves much more than simply number crunching. Crunching numbers is certainly part of it, but, unless those numbers have been carefully and meaningfully collected, crunching them is not going to provide any useful information. 

In this book we have adopted the broad operational definition of Statistics provided by Turner (2007). In this definition. Statistics is regarded as a multi-faceted scientific discipline that comprises the following activities  

This functional definition makes clear that the discipline of Statistics is Indeed multi-faceted and essential throughout clinical trials. It is critical at the start of the clinical trial process so that a study can be designed appropriately to facilitate the collection of optimum quality data, which then need to be organized and managed correctly. These data are then described and analyzed, and the numerical results of these analyses are interpreted in the context of the particular study. Finally, the numerical results of the analyses and the authors interpretation of these results are presented to regulatory agencies to request permission to market the drug, and published in clinical communications to provide Information to physicians. 

This functional definition of Statistics may well contain some concepts and terms with which you are not familiar at this point, and that is fine. This book s goal is to make you familiar with these terms and concepts so that you will understand the statistical processes and procedures that are used in clinical trials. Individual chapters address different parts of this definition. However, it is important for us to emphasize here that the individual aspects presented in the chapters are really seamless components of one overall experimental approach to gaining knowledge, the discipline of Statistics. These components act together to ensure that high-quality data acquisition, correct analysis, and appropriate interpretations provide optimal answers to good research questions. 

Throughout the book, statistical considerations are presented conceptually rather than computationally. It is hoped that, by the end of the book, the word statistics may appear less mysterious, irrelevant, or threatening to readers for whom the very mention of the word conjures up these or similar feelings. 

---

The discipline of Statistics is employed in clinical trials to investigate a new drug s safety and efficacy. This investigation may provide compelling evidence that the drug is safe and will induce a biological response that improves patients well-being. 

Consider two examples from previous chapters that illustrate this. In Chapter 9, discussions of sample-size estimation emphasized that the process is indeed one of estimation rather than pure calculation. A calculation is certainly executed, but the values that are placed into the appropriate formula are chosen by the sponsor. On each occasion, the sponsor must consider the influences of the choices that are made and make the most appropriate decision in the specific context of that trial. In Chapter 11, equivalence and noninferiority designs were discussed. In addition to the calculations that are involved using the data collected in a trial, equivalence or noninferiority margins must be established before the trial commences. Their choice is a clinical choice, not a statistical choice, and subjectivity is necessarily involved in this choice. Thus, the discipline of Statistics certainly involves using informed judgments. Statistics really is an art as well as a science, a sentiment well expressed by Katz (2001) as cited in Chapter 13. 

The discipline of statistics is the study of effective methods of data collection, data summarization, and (data based, quantitative) inference making in a framework that explicitly recognizes the reality of nonnegligible variation in real-world processes and measurements. 

MODERN BUSINESS PROCESS IMPROVEMENT AND THE DISCIPLINE OF STATISTICS... 

The process of research in chemical systems is one of developing and testing different models for process behavior. Whether empirical or mechanistic models are involved, the discipline of statistics provides data-based tools for discrimination between competing possible models, parameter estimation, and model verification for use in this enterprise. In the case where empirical models are used, techniques associated with linear regression (linear least squares) are used, whereas in mechanistic modeling contexts nonlinear regression (nonlinear least squares) techniques most often are needed. In either case, the statistical tools are applied most fruitfully in iterative strategies. 

It is common to start a textbook with a definition of the book s topic, ffowever, in the case of the discipline of Statistics (indicated in this book with a capital "S"), it is difficult to find a universally accepted definition. Different textbooks are written for different target audiences, and their goals can therefore be quite different. So, before going any further, it is appropriate to identify our target audiences, and to provide you with a definition of Statistics that is informative and meaningful in the context of this book. 

The discipline of Statistics is discussed in the context of pharmaceutical clinical trials because two of the primary target audiences for this book are students of pharmacy and students of clinical research. A clinical trial can be defined as an experiment testing a medical treatment on humans (Piantadosi, 2005). In this definition, the words "human" and "clinical" are closely linked. However, this definition also makes it clear that clinical trials can be performed to test a variety of medical treatments. In addition to pharmaceutical trials, the focus of this book, clinical trials are conducted to test medical devices (see Becker and Whyte, 2006) and some surgical practices. 

A couple of conceptual points are useful here. First, there is much more to the discipline of Statistics than "obligatory calculations" at the end of a study. Rather, the discipline of Statistics is an integral component throughout the entire... 

Many decisions have to be made throughout the drug development process. A lot of these decisions concern whether it is prudent to continue to the next phase of development, as outlined in Chapter 2. To allow rational decisions to be made we need reliable, quantitative information upon which to base our decisions. In a very real sense the main contribution of the discipline of Statistics in research endeavors is providing numerical representations of information that facilitate good decision-making. 

This book, therefore, is a self-contained introduction to the discipline of Statistics and its use in therapeutic confirmatory clinical trials. The first part of the book provides introductory comments about the discipline of Statistics and lays the foundations for our later discussions in the context of pharmaceutical trials. Chapter 2 presents an overview of the process of new drug development and the role of clinical trials in this process. The categories of different types of clinical trials are identified and discussed so that you will understand thetypes of data collectedin them. 

Providing the evidence This is the domain of clinical research and, in the case of drug development, the province of clinical trialists. The discipline of Statistics is a central and critical component of the planning, conduct, analysis, and interpretation of clinical trials. 

Sponsors typically publish the results of important clinical trials in clinical communications in medical journals, and present the results at scientific conferences. As for the preparation of regulatory documentation, scientific communications depend heavily on the discipline of Statistics. Piantadosi (2005) made the following comment about publishing clinical communications ... 

While the format of clinical communications is different from the format of clinical study reports that are submitted to regulatory agencies, the discipline of Statistics provides the basis for the approach taken. A typical format for a clinical communication is ... 

It is precise. There will be a precise answer -yes it does, or no it doesn t. We will have to define the term "more" in more detail shortly, but in the meantime please trust us that this can indeed be done in a precise manner by using the discipline of Statistics. 

Like the word significance, the concept of probability is used in everyday language as well as in the discipline of Statistics. As Turner (2007) noted, the statement "I ll probably be there on Saturday" involves a probabilistic statement, but there is no degree of quantification (if you know the individual making this statement, past experience may lead you to have an informed opinion concerning the relative meaning of "probably," but this is a subjective judgment). 

In this book, set in the context of drug development, we have taught you how to conduct an array of statistical analyses. While teaching such computational skills is appropriate in a statistics textbook, we also hope that we have been successful in providing you with a conceptual understanding of and appreciation for the contribution of the discipline of Statistics to the development of pharmaceutical drugs that may improve the health of your family members, your friends, and yourself. 

---