Alternatively, comparative trials may evolve from questions arising in clinical practice and not from a specific development process. Thus, one may wish to compare different surgical timings, at 6 months or at 1 year of age, for reconstructive surgery in infants with cleft palate as is proposed in the trial conducted by Yeow, Young, Chen, et al. (2019).
Whatever the pathway, the eventual randomised comparative trial to be conducted is clearly a major event as only when this has been conducted will there be reliable (although not necessarily convincing) evidence of the efficacy of the intervention concerned. In certain situations, often for regulatory purposes, a Phase III trial may be followed by a confirmatory trial asking essentially the same question. In addition, following regulatory approval of a product, so‐called Phase IV or postmarketing trials may be initiated with the aim to gain broader experience with using the new product.
1.8 Ethical considerations
For a trial to be ethical, at the time it is designed the ethical review committees will want to be convinced that there is collective uncertainty amongst clinicians as to which treatment is superior or more appropriate for the patients. They will also need to be persuaded that the sample size and other aspects of the study design are such that the trial is likely to provide information sufficient to reduce this uncertainty and thus influence subsequent medical practice if one treatment or the other appears superior.
A clinical trial cannot go forward until the protocol has been through the appropriate ethical review processes, the exact nature of which varies from country to country. These should always include a very thorough review of the scientific aims as well as the more ‘subject’ oriented concerns to protect those who will be recruited to the trial. In brief terms, this implies that if a trial is not scientifically sound – then it should not be judged as ethically acceptable.
1.9 Regulatory requirements
In addition to the more overtly scientific parts of the clinical trials process on which to focus, there are many regulatory requirements which national and international law obliges a trial team to adhere to. For example, the regulations insist that informed consent is obtained from patients entering trials and on the preservation of personal data confidentiality. These regulations are generally referred to as requirements for Good Clinical Practice (GCP) as is described in ICH E6 (R2) (2016). We will refer to specific aspects of GCP as they arise in the text, but readers are cautioned that the specifics are continually being updated. Principles to guide statisticians working on clinical trials have been laid down by ICH E9 (R1) (2018).
If the trial is seeking regulatory approval of (say) a new drug, then all the associated requirements for approval should be reviewed by the trial team before, during and after the trial protocol is being developed to ensure that all aspects are covered so as to avoid the rejection of the application on what might be a technical detail. For example, there may be a regulatory requirement for some additional animal studies to be conducted before approval can be granted. These requirements are summarised in documents such as those provided by the European Medicines Agency (EMEA) and the US Food and Drug Administration (FDA).
In some circumstances, it is a requirement for regulatory approval that a confirmatory trial is conducted. Such a trial is essentially a repeat of an initial one, perhaps in a different or wider patient group or with a wider group of clinical teams involved, but it must follow the essential features of the predecessor design. Clearly, these details should be cross‐checked with the relevant authorities before the protocol is finalised and patients are recruited.
1.10 Focus
As we have illustrated, the size of clinical trials can range from the relatively few to as many as several thousands of subjects being recruited. Consequently, and leaving specific details aside, these will require a range of resources from the relatively modest to the very considerable. It must be emphasised that the size of a clinical trial is determined by the question(s) that are posed, and the resources allocated should reflect the importance of that question. Clearly, a very experienced team is required to launch a large trial but even the design team of an ultimately small sized trial will need access to appropriate personnel including, at a minimum, those with clinical, statistical, data management and organisational skills and often other specialist skills from, for example, pharmacy, pathology and many other specialties. It is important that the design team do not underestimate the scale of the task.
The focus of this book is on the design of (randomised) comparative (usually termed Phase III) trials which are likely to be of relatively modest size. We aim to give clear guidance as to how these may be designed, conducted, (to some extent) analysed and reported. However, it is also important that investigators contributing patients to clinical trials who are perhaps not part of the design team also understand the issues concerned as the very success of the trials depends crucially on their collaboration and understanding of the processes involved.
1.11 Further reading
Although Day (2007) provides a comprehensive list of books about clinical trials the following are particularly useful:
1 Day S (2007). Dictionary for Clinical Trials. (2nd edn). Wiley, Chichester.
2 Fitzpatrick S (2008a). Clinical Trial Design. ICR Publishing, Marlow.
3 Fitzpatrick S (2008b). The Clinical Trial Protocol. ICR Publishing, Marlow.
4 Girling DJ, Parmar MKB, Stenning SP, Stephens RJ and Stewart LA (2003). Clinical Trials in Cancer: Principles and Practice. Oxford University Press, Oxford.
5 Machin D, Campbell MJ, Tan SB and Tan SH (2018). Sample Size Tables for Clinical, Laboratory and Epidemiology Studies. (4th edn). Wiley‐Blackwell, Chichester.
6 Redwood C and Colton T (eds) (2001). Biostatistics in Clinical Trials. Wiley, Chichester.
7 Wang D and Bakhai A (eds) (2006). Clinical Trials: A Practical Guide to Design, Analysis and Reporting. Remedica, London
Hints on how to display medical data in tabular and graphical form are given by:
1 Freeman JV, Walters SJ and Campbell MJ (2008). How to Display Data. BMJ Books, Oxford.
For those specifically interested in health‐related quality of life issues:
1 Fayers PM and Machin D (2016). Quality of Life: The Assessment, Analysis and Interpretation of Patient‐reported Outcomes. (3rd edn). Wiley‐Blackwell, Chichester.
For those specifically interested cluster randomised trials at a more technical level:
1 Campbell MJ and Walters SJ (2014). How to Design, Analyse and Report Cluster Randomised Trials in Medicine and Health Related Research. Wiley, Chichester.
For those requiring a wide view of how randomised trials have impacted on clinical practice over a wide range of diseases and conditions:
1 Machin D, Day S and Green S (eds) (2006). Textbook of Clinical Trials. (2nd edn). Wiley, Chichester.
CHAPTER 2 Design Features
This chapter gives an overview of the general structure of a randomised clinical trial. The key components are highlighted. These include the type of patients or subjects that are likely to be relevant to the objectives of the trial and then, within this group, those who are specifically eligible for the trial in mind, the research question(s) and the choice of design. We emphasise the requirement of fully informed consent before a patient is entered into a trial, the determination as to whether or not the interventions on offer are equally appropriate for the individual concerned, the method of allocation to the alternative interventions and