Guidelines for clinical trial protocols for interventions involving artificial intelligence: the SPIRIT-AI extension

The SPIRIT 2013 declaration aims to improve the exhaustiveness of clinical trial protocol reports by providing evidence-based recommendations for the minimum set of elements that must be addressed. This guide has been fundamental in promoting the transparent evaluation of new interventions. More recently, it has been increasingly recognized that interventions with artificial intelligence (AI) must undergo rigorous and prospective evaluation to demonstrate their impact on medical outcomes. The SPIRIT-AI extension (Standard Protocol Items: Recommendations for Interventional Trials-Artificial Intelligence, by its acronym in English) is a new guideline for the reporting of clinical trial protocols that evaluate interventions with an AI component. This guideline was developed in parallel with its complementary declaration for clinical trial reports: CONSORT-AI (Consolidated Standards of Reporting Trials-Artificial Intelligence). Both guidelines were developed through a staged consensus process that included literature review and consultation with experts to generate 26 candidate items, which were consulted by an international group of multiple interested parties in a two-stage Delphi survey ( 103 interested parties), agreed in a consensus meeting (31 interested parties) and refined through a pilot checklist (34 participants). The expansion of SPIRIT-AI includes 15 new elements that are considered sufficiently important for clinical trial protocols with AI interventions. These new items must be reported routinely in addition to the central items of SPIRIT 2013. SPIRIT-AI recommends that researchers provide clear descriptions of the AI intervention, including the instructions and skills necessary for its use, the environment into which it will be integrated the AI intervention, the considerations for the management of input and output data, the interaction between the human being and the AI and the analysis of error cases. SPIRIT-AI will help promote transparency and exhaustiveness of clinical trial protocols for AI interventions. Its use will help editors and reviewers, as well as readers in general, to understand, interpret and critically value the design and risk of a future clinical trial.