Table 2 Utility of Gibbs Framework and suggested improvements

Integration of CC into a systematic approach to the stages of development of a trial intervention

Create definitions for all terms used in the Framework to improve applicability e.g. expand 'peer educator' definition to include patient and staff profiles

Prompts rationale-based judgements around the levels of CC and not only binary Yes/No output

Formulate a preceding preamble to guide formulation of the target population in advance of applying the Gibbs framework

Scoring is per question permitting within-question judgments rather than an overall score for the tool (this mirrors approaches to study risk of bias assessment, e.g. the Cochrane ROB tool)

Develop an elaboration and explanation document for each question as used in CONSORT with worked examples to illustrate differences between CC levels 1 and 2

The Gibbs framework may need to be applied across multiple documents related to a single research study to fully address CC e.g. protocol, data sources, and final manuscript

Consideration to be given that a cluster trial (or other non-individually randomised trial designs) may have more than one level of target population so multiple applications of frameworks may be required targeted at the different levels of the trial population