Randomised Clinical Trials. David Machin
Чтение книги онлайн.
Читать онлайн книгу Randomised Clinical Trials - David Machin страница 27
![Randomised Clinical Trials - David Machin Randomised Clinical Trials - David Machin](/cover_pre950308.jpg)
2.10.3 Choice of design
On the basis of model (2.1), the two fundamental issues in trial design to consider are as follows:
1 What levels of the design variable x to choose?
2 How many subjects should we recruit?
In this model, with x = 0 for the standard or control treatment, S, y0 = β0 and this represents the ‘true’ or population mean of this group, labelled μS. Alternatively, with x = 1, the sum y1 = β0 + βTreat represents μT the population mean of group T. Thus, the difference, δ, between those receiving S and T, is y1 − y0 = β0 + βTreat − β0 = βTreat. From this βTreat = μT − μS and β0 = μS. Population means are estimated by sample means, in this case
In this example, Equation (2.1) describes the two‐group clinical trial of this chapter but, as we will see in other chapters, we can extend this model to describe more complex clinical trial designs although this basic type of equation encapsulates the essential structure of all our analyses.
2.10.4 Randomisation
In essence, one purpose of any experiment is to estimate the parameters of a statistical model analogous to that of Equation (2.1). Thus, we conduct a trial in order to collect data with this purpose in mind. We would like to believe that the estimates we obtain in some way reflect the true or population parameter values. In principle, if we repeated the study many times, then we would anticipate that these estimates would form a distribution that is centred on the true parameter value. If this is the case, our method of estimation is unbiased. For example, in a clinical trial comparing two treatments, the parameter βTreat corresponds to the true difference (if any) in efficacy between them, and the object of the trial is to obtain an unbiased estimate of this. The method of selecting which of the eligible patients is to be included in the trial does not affect this, but the way in which those patients who are recruited to the trial are then allocated to which particular treatment does. As we discussed in Chapter 1, of fundamental importance to the design of any clinical trial is the random allocation of subjects to the alternative treatments. Randomisation also provides a sound basis for the ensuing hypothesis testing by the use of statistical tests of significance.
2.11 Guidelines
When planning a clinical trial of whatever size or complexity, it is very important to establish the ground rules. Although many of these are common‐sense based, it is very useful to have access to clear guidelines for all stages of the process. Such guidelines have been published covering a wide range of topics ranging from regulatory requirements to what a statistical referee should look for when reviewing a clinical trial manuscript for publication.
We will refer to some of these guidelines at the relevant stages later in the book. However, readers are warned that many of these are constantly being updated and it is always useful to check if the ones referred to are the most current.
2.12 Further reading
1 O’Cathain A, Croot L, Duncan E, Rousseau N, Sworn K, Turner KM, Yardley L and Hoddinott P (2019). Guidance on how to develop complex interventions to improve health and healthcare. BMJ Open, 9, e029954.
Fuller details are provided by www.mrc.ac.uk/complexinterventionguidance or the European Medicines Agency, Spark building, Orlyplein 24, 1043 DP Amsterdam, The Netherlands, http://www.emea.eu.
1 ICH E9 (R1) (2018). Statistical Principles for Clinical Trials. CPMP/ICH/363/96.
This contains, for example, a clear statement of what distinguishes an ITT from a per‐protocol analysis.
1 ICH E10 (2001). Choice of Control Group in Clinical Trials. CPMP/ICH/364/96.
Contains information concerning what are regarded as suitable control groups in many situations.
CHAPTER 3 The Trial Protocol
It is a fundamental requirement to develop a formal protocol for any clinical trial and we describe in general terms the subject matter of such a protocol. The chapter focuses on the content common to all trial protocols such as the background to the trial, the basic design, the type and number potential subjects to recruit, informed consent, details of the intervention options, and other practicalities including the forms required for recording the data. We illustrate each section by extracts from activated protocols used for a variety of trials in different areas.
We also emphasise the need to check local regulations concerned with the conduct of trials and reference is made to some published guidelines to help in the protocol development process.
3.1 Introduction
Day (2007) defines a protocol as:
A written document describing all the important details of how a study will be conducted. It will generally include details of the products being used, a rationale for the study, what procedures will be carried out on subjects in the study, how many subjects will be studied, the design of the study and how the data will be analysed.
In our context, ‘study’ is replaced by ‘trial’ and commonly ‘products’ will be replaced by ‘treatments’ or ‘interventions’. In particular, a clinical trial protocol is a document that will include sections addressing those design features we have highlighted in Chapter 2, as they relate to the specific trial in question.
As we have pointed out, the questions posed in a clinical trial must be important and so it is vital that the research team identify all the scientific, clinical and practical elements, together with the resources and skills necessary in order to achieve clear answers. This is no small task even when embarking on a clinical trial of moderate size and complexity. A fundamental part of this