Statistical trial design

ISSUES IN STATISTICAL TRIAL DESIGN Multicenter Trials... 

An important difference between the statistical mixture design techniques popular in HPLC and the PRISMA model is that the former yields a computed optimum solvent composition id>ile the latter relies on a structured trial and error approach, which is readily adaptable to TLC. Solvent changes and re-equilibration in HPLC can be quite time consuming, so that it becomes attractive to ainimize the number of experiments, while for TLC, experiments can be performed in parallel and time constraints are less significant. Changes in solvent strength are also more rapidly adjusted empirically within the PRISMA model when theoretical considerations are found inadequate or require modification due to differences in the experimental approach. 

The E9 discusses the statistical issues in the design and conduct of a clinical trial. It details trial design, trial conduct, and data analysis and reporting. Although most useful... 

Adequate resolution of the components of a mixture in the shortest possible time is nearly always a principal goal. Establishing the optimum conditions by trial and error is inefficient and relies heavily on the expertise of the analyst. The development of computer-controlled HPLC systems has enabled systematic automated optimization techniques, based on statistical experimental design and mathematical resolution functions, to be exploited. The basic choices of column (stationary phase) and detector are made first followed by an investigation of the mobile phase composition and possibly other parameters. This can be done manually but computer-controlled optimization has the advantage of releasing the analyst for other... 

Many of the important statistical aspects of clinical trials design are dealt with in detail in Chapter 6 of this volume (Nigel Baber and John Sweatman) and so here only specific statistical aspects are discussed. 

When using such a statistical significance test, it is important to recognise that this generally has low power in a trial designed to detect the main effect of treatment/... 

Three comments are appropriate here. First, consideration of the traditional clinical trial design that has been the focus of attention up until this chapter is extremely worthwhile and instructive It has facilitated the introduction of fundamental design, methodology, and statistical concepts, and it will be an influential player in pharmaceutical drug development for many years to come. Second, the simple observation that the adaptive design may seem different does not in itself make it less valid, less valuable, or less important. Third, statistical approaches that are suitable for adaptive designs are, as yet, less well developed than they are for other study designs. 

Buncher, C.R. and Tsay, J-Y., 2006a, Clinical trial designs. In Buncher, C.R. and Tsay, J-Y., (Eds), Statistics in the pharmaceutical industry, 3rd Edition, Chapman Hall/CRC, 79-90. 

Other aspects of trial design including randomization, acceptability of sham operation in a placebo arm, appropriate statistical approaches, as well... 

Obviously, not only are the fundamental features of the study design determined, in part, by statistical models, but the trial design in turn dictates the analytical models to be used. Most plans, particularly those for a multicenter study, should allow blocking or stratification of the data so that differences in individual study centers can be accounted for in the analysis. 

When the patients who received a drug candidate have the disease manifestations completely eradicated and experience no other effects while patients treated with a placebo have a continuation of the disease process, the evaluation is not difficult. However, that is rarely, if ever, the case, and evaluation requires detailed statistical analysis of the collected data. An ICH guide-line covers statistical issues related to the scope of clinical trials, design techniques to minimize bias, types of clinical trial designs, conduct considerations, data analysis for efficacy, evaluation of safety and tolerance, and reporting. 

Another difficult aspect in the design of open-label studies is how one assesses those patients who withdraw from the study. The reasons for withdrawal can be at least as varied as in double-blind studies (intolerability, administrative difficulties, coincidental emergent disease or concomitant therapies, etc.). However, in addition, in an open-label design, patients may develop an opinion on the superiority of one or other treatment for reasons that may or may not be explicit. If completion of a course of therapy is one end point of the study, then all withdrawals can be accounted treatment failures, and the statistical handling is fairly straightforward. However, if there is another end point, and if withdrawals are imbalanced between the treatment groups and unrelated to product intolerability, then the situation becomes a lot more clouded. Under these latter conditions, the entire trial may have to be abandoned when it becomes apparent that the trial design cannot answer the hypothesis under test one way or the other. 

We have already seen through a number of examples the interplay between sample size, variability and the performance of the statistical procedures employed to analyze the data. The sample size determines the amount of information that will be available at the end of the trial. Therefore, the determination of an adequate sample size is one of the most important aspects of the trial design. A trial accumulating inadequate amount of information is hopelessly flawed, as it will not enable the researcher to answer the questions the trial is intended to answer. 

---