Jonathan Cook is an Associate Professor based at the Centre for Statistics in Medicine at the University of Oxford and a Senior Statistician within the Oxford Clinical Trials Research Unit (OCTRU). He has been involved in trials for over 15 years mainly in the area of surgical trials. Jonathan is also an Editorial Board Member for Pilot and Feasibility Studies.

Here, he answers questions on his recent article published in the journal on the appropriateness of the term ‘internal pilot’.

What is an ‘internal pilot’ within a clinical trial?

An internal pilot can be defined as the early part of the planned recruitment period of a clinical (usually) randomized trial.

Why have you raised concerns with the label ‘internal pilot’ and what alternatives do you recommend?

… it often felt to me that this terminology failed to help people think more clearly about the topic and current practice.

There are a couple of main reasons. First, having been involved in a few workshop and courses over the last few years where the concept of internal and external pilots has been introduced and discussed, it often felt to me that this terminology failed to help people think more clearly about the topic and current practice. I think this is partly due to the use of the label “internal pilot” to describe something occurring within the main trial goes against the natural meaning of the word “pilot”.

The other reason is that the increased focus on a one-off assessment seems to have led to a downplaying of the critical role of regular monitoring of progress throughout the recruitment period.

We suggest an alternative way of categorizing approaches to address recruitment uncertainties in the commentary and suggest avoiding using the language of “internal” and “external” pilot studies.

What kind of discussions do you hope to stir for both trialists and funders?

Hopefully it will get people to think about the current practice – I’m sure not all will agree with what we say but hopefully it will lead to further reflection on practice.

Hopefully it will get people to think about the current practice – I’m sure not all will agree with what we say but hopefully it will lead to further reflection on practice.

In particular I think we need more research looking at how we undertake routine monitoring of recruitment. We tend to say little about this in trial protocols and reports, and can give the impression that this is not a core consideration when running a trial whereas it is a vital part of successfully conducting a randomized trial, and in particular meeting the recruitment target.

How do you envision the journal Pilot and Feasibility Studies could help with the situation?

There is a need to think more in-depth about what type of routine monitoring should go on in every trial and how we can make this more effective. Additionally, we think trial reports should be more open about this monitoring and the modifications made to a trial during its conduct than tends to be the case at present.

These aspects need further work exploring how best to do this and Pilot and Feasibility Studies is the natural home for publishing such work and engaging with the community.