TRIPOD+AI

A 27-item checklist for writing up studies that develop or validate a clinical prediction model, whether it uses regression or machine learning.

What it is¶

TRIPOD+AI tells you what to include when you report a diagnostic or prognostic prediction model study, from the title and abstract through data, analysis, results and discussion. It updates and replaces TRIPOD 2015 and adds items that matter for machine learning, such as how you handled fairness, data preparation and open science. A separate 13-item checklist covers abstracts, and an expanded version explains each item.

In healthcare¶

Built for clinical prediction models in any specialty; use it for model development studies, external validation studies, or papers that do both.

Where to use it in The ELSA Way¶

• Reporting and transparency › International reporting guidelines: Use this checklist when your study develops or externally validates a prediction model, which covers most published clinical AI studies.

• Evaluation planning › Pre-specification: Reading the checklist before you start shows which analysis choices, such as subgroups and performance measures, you will later have to justify. Decide them in advance.

References¶

• Collins GS, Moons KGM, Dhiman P, Riley RD, Beam AL, Van Calster B, et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ. 2024;385:e078378. Collins et al. (2024)

Related tools¶

• PROBAST+AI: A structured tool for judging the quality, risk of bias and applicability of studies that develop or test a clinical prediction model, including AI models.

• Model Cards: A short, structured document that travels with a trained model and tells readers what it is for, how it was tested and where it performs worse.

• CONSORT-AI: An extension of the CONSORT checklist that tells you what to report when a randomised trial tests an intervention with an AI component.