Skip to main content

Box 3 ACE guideline scope and general principles

From: The adaptive designs CONSORT extension (ACE) statement: a checklist with explanation and elaboration guideline for reporting randomised trials that use an adaptive design

1. It applies to all randomised clinical trials using an adaptive design (AD), as defined in Box 1.
2. It excludes randomised clinical trials that change aspects of an ongoing trial based entirely on external information [130] or with internal pilots focusing solely on feasibility and processes (such as recruitment, intervention delivery, and data completeness) [131].
3. It covers general reporting principles to make it applicable to a wide range of current and future ADs and trial adaptations.
4. It is not intended to promote or discourage the use of any specific type of AD, trial adaptation, or frequentist or Bayesian statistical methods. These choices should be driven by the scientific research questions, the goals behind the use of the proposed AD features, and practical considerations [22].
5. It aims to promote transparent and adequate reporting of AD randomised trials to maximise their potential benefits and improve the interpretability of their results and their reproducibility, without impeding their appropriate use or stifling design innovation. Therefore, the guideline does not specifically address the appropriateness of adaptive statistical methods.
6. It presents the minimum requirements that should be reported but we also encourage authors to report additional information that may enhance the interpretation of trial findings.
7. Access to information is most important regardless of the source and form of publication. For example, use of appendices and citation of accessible material (such as protocols, statistical analysis plans (SAPs), or related publications) is often sufficient.
8. The order in which researchers report information does not necessarily need to follow the order of the checklist.
9. The guideline does not primarily address specific reporting needs for non-randomised ADs (such as phase I dose escalation studies, phase II single-arm designs). However, some principles covered here may still apply to such trials.