What is the TRIPOD+AI checklist?

TRIPOD+AI is the reporting guideline for studies that develop or evaluate a clinical prediction model, whether the model uses regression or machine learning. Its checklist has 27 items, 52 with sub-items, running from the title to the discussion, plus a separate 13-item checklist for abstracts. It was published in the BMJ in April 2024 by Collins and colleagues.

Does TRIPOD+AI replace TRIPOD 2015?

Yes. The authors state that the TRIPOD+AI checklist supersedes the TRIPOD 2015 checklist, which should no longer be used. Use TRIPOD+AI even if your model is a logistic or Cox regression and involves no artificial intelligence: the “+AI” means the guidance now also covers machine learning methods.

What do D and E mean in the TRIPOD+AI checklist?

D marks items that apply only to studies developing a prediction model, such as the choice of predictors (9a) and the full model specification (22). E marks items that apply only to studies evaluating a model, such as model updating (12f, 24). Items without a label apply to both. Choose your study type above and the tool hides the items that do not apply.

What is new in TRIPOD+AI compared with TRIPOD 2015?

TRIPOD+AI adds or expands items on health inequalities (3c), data preparation (7), class imbalance (13), fairness (14), open science (18a to 18f: funding, conflicts of interest, protocol, registration, data sharing and code sharing), patient and public involvement (19), and usability of the model in current care (27a to 27c). Fairness is also built into several other items, and the wording covers machine learning as well as regression.

What is the TRIPOD+AI for Abstracts checklist?

A 13-item checklist for the abstract of a prediction model study, which item 2 of the main checklist points to. It covers the title, background, objectives, data sources, eligibility and setting, outcome, model building and validation, performance measures, participants and events, predictors, performance estimates, interpretation and registration. Items 7 and 10 apply only to development studies. Open it with the “Abstract (13 items)” tab.

How do I fill in the TRIPOD+AI checklist?

For each item, write where your manuscript reports it, such as “Methods, Sample size, p. 6; Supplement S2”. If an item does not apply, for example class imbalance methods in a study that used none, mark it not applicable and say why. Download the Word file and upload it as a supplementary file with your submission.

Which TRIPOD checklist should I use for clustered data or a systematic review?

Use TRIPOD+AI as the main checklist. For models developed or validated in clustered data, such as several hospitals, also consult TRIPOD-Cluster. Systematic reviews and meta-analyses of prediction model studies use TRIPOD-SRMA, and study protocols use TRIPOD-P. For risk of bias in prediction model studies, use PROBAST rather than a reporting checklist.

Is TRIPOD+AI a quality assessment tool?

No. TRIPOD+AI says what to report, not how to develop or evaluate a model, and it is not a quality appraisal tool. A well-reported model can still be poorly developed. To judge the risk of bias of a prediction model study, the authors refer readers to PROBAST.

Can I reuse the TRIPOD+AI checklist?

Yes. The TRIPOD+AI statement was published open access under the Creative Commons Attribution (CC BY 4.0) licence, so the checklist can be copied and adapted as long as you cite Collins et al., BMJ 2024;385:e078378.

Are my answers stored anywhere?

No. The checklist runs in your browser and nothing is sent to a server. Your answers are kept in this browser’s local storage so they are still there when you come back; use “Clear answers” to remove them.

TRIPOD+AI Checklist: Fillable Word Template

Choose whether your study develops a prediction model, evaluates one, or both. Then mark each TRIPOD+AI item as reported or not applicable and note where it appears in your manuscript. The abstract tab covers TRIPOD+AI for Abstracts. Download the completed checklist as a Word file for your submission. Your answers are saved in this browser only.

Manuscript title (optional)

Study type

Development and evaluationModel developmentModel evaluation

Items for development only (D) or evaluation only (E) are hidden when they do not apply. Studies that develop and evaluate a model report every item.

0 of 52 items done

0 reported · 0 not applicable · 52 open

Show open items only

Title

1
Title
Identify the study as developing or evaluating the performance of a multivariable prediction model, the target population, and the outcome to be predicted

Abstract

2
Abstract
See TRIPOD+AI for Abstracts checklist

Introduction

3a
Background
Explain the healthcare context (including whether diagnostic or prognostic) and rationale for developing or evaluating the prediction model, including references to existing models
3b
Background
Describe the target population and the intended purpose of the prediction model in the context of the care pathway, including its intended users (eg, healthcare professionals, patients, public)
3c
Background
Describe any known health inequalities between sociodemographic groups
4
Objectives
Specify the study objectives, including whether the study describes the development or validation of a prediction model (or both)

Methods

5a
Data
Describe the sources of data separately for the development and evaluation datasets (eg, randomised trial, cohort, routine care or registry data), the rationale for using these data, and representativeness of the data
5b
Data
Specify the dates of the collected participant data, including start and end of participant accrual; and, if applicable, end of follow-up
6a
Participants
Specify key elements of the study setting (eg, primary care, secondary care, general population) including the number and location of centres
6b
Participants
Describe the eligibility criteria for study participants
6c
Participants
Give details of any treatments received, and how they were handled during model development or evaluation, if relevant
7
Data preparation
Describe any data pre-processing and quality checking, including whether this was similar across relevant sociodemographic groups
8a
Outcome
Clearly define the outcome that is being predicted and the time horizon, including how and when assessed, the rationale for choosing this outcome, and whether the method of outcome assessment is consistent across sociodemographic groups
8b
Outcome
If outcome assessment requires subjective interpretation, describe the qualifications and demographic characteristics of the outcome assessors
8c
Outcome
Report any actions to blind assessment of the outcome to be predicted
9a
PredictorsModel development only
Describe the choice of initial predictors (eg, literature, previous models, all available predictors) and any pre-selection of predictors before model building
9b
Predictors
Clearly define all predictors, including how and when they were measured (and any actions to blind assessment of predictors for the outcome and other predictors)
9c
Predictors
If predictor measurement requires subjective interpretation, describe the qualifications and demographic characteristics of the predictor assessors
10
Sample size
Explain how the study size was arrived at (separately for development and evaluation), and justify that the study size was sufficient to answer the research question. Include details of any sample size calculation
11
Missing data
Describe how missing data were handled. Provide reasons for omitting any data
12a
Analytical methodsModel development only
Describe how the data were used (eg, for development and evaluation of model performance) in the analysis, including whether the data were partitioned, considering any sample size requirements
12b
Analytical methodsModel development only
Depending on the type of model, describe how predictors were handled in the analyses (functional form, rescaling, transformation, or any standardisation)
12c
Analytical methodsModel development only
Specify the type of model, rationale†, all model building steps, including any hyperparameter tuning, and method for internal validation
12d
Analytical methods
Describe if and how any heterogeneity in estimates of model parameter values and model performance was handled and quantified across clusters (eg, hospitals, countries). See TRIPOD-Cluster for additional considerations‡
12e
Analytical methods
Specify all measures and plots used (and their rationale) to evaluate model performance (eg, discrimination, calibration, clinical utility) and, if relevant, to compare multiple models
12f
Analytical methodsModel evaluation only
Describe any model updating (eg, recalibration) arising from the model evaluation, either overall or for particular sociodemographic groups or settings
12g
Analytical methodsModel evaluation only
For model evaluation, describe how the model predictions were calculated (eg, formula, code, object, application programming interface)
13
Class imbalance
If class imbalance methods were used, state why and how this was done, and any subsequent methods to recalibrate the model or the model predictions
14
Fairness
Describe any approaches that were used to address model fairness and their rationale
15
Model outputModel development only
Specify the output of the prediction model (eg, probabilities, classification). Provide details and rationale for any classification and how the thresholds were identified
16
Training versus evaluation
Identify any differences between the development and evaluation data in healthcare setting, eligibility criteria, outcome, and predictors
17
Ethical approval
Name the institutional research board or ethics committee that approved the study and describe the participant informed consent or the ethics committee waiver of informed consent

Open science

18a
Funding
Give the source of funding and the role of the funders for the present study
18b
Conflicts of interest
Declare any conflicts of interest and financial disclosures for all authors
18c
Protocol
Indicate where the study protocol can be accessed or state that a protocol was not prepared
18d
Registration
Provide registration information for the study, including register name and registration number, or state that the study was not registered
18e
Data sharing
Provide details of the availability of the study data
18f
Code sharing
Provide details of the availability of the analytical code§

Patient and public involvement

19
Patient and public involvement
Provide details of any patient and public involvement during the design, conduct, reporting, interpretation, or dissemination of the study or state no involvement

Results

20a
Participants
Describe the flow of participants through the study, including the number of participants with and without the outcome and, if applicable, a summary of the follow-up time. A diagram may be helpful
20b
Participants
Report the characteristics overall and, where applicable, for each data source or setting, including the key dates, key predictors (including demographics), treatments received, sample size, number of outcome events, follow-up time, and amount of missing data. A table may be helpful. Report any differences across key demographic groups
20c
ParticipantsModel evaluation only
For model evaluation, show a comparison with the development data of the distribution of important predictors (demographics, predictors, and outcome)
21
Model development
Specify the number of participants and outcome events in each analysis (eg, for model development, hyperparameter tuning, model evaluation)
22
Model specificationModel development only
Provide details of the full prediction model (eg, formula, code, object, application programming interface) to allow predictions in new individuals and to enable third party evaluation and implementation, including any restrictions to access or reuse (eg, freely available, proprietary)¶
23a
Model performance
Report model performance estimates with confidence intervals, including for any key subgroups (eg, sociodemographic). Consider plots to aid presentation
23b
Model performance
If examined, report results of any heterogeneity in model performance across clusters. See TRIPOD-Cluster for additional details‡
24
Model updatingModel evaluation only
Report the results from any model updating, including the updated model and subsequent performance

Discussion

25
Interpretation
Give an overall interpretation of the main results, including issues of fairness in the context of the objectives and previous studies
26
Limitations
Discuss any limitations of the study (such as a non-representative sample, sample size, overfitting, missing data) and their effects on any biases, statistical uncertainty, and generalisability
27a
Usability of the model in the context of current careModel development only
Describe how poor quality or unavailable input data (eg, predictor values) should be assessed and handled when implementing the prediction model
27b
Usability of the model in the context of current careModel development only
Specify whether users will be required to interact in the handling of the input data or use of the model, and what level of expertise is required of users
27c
Usability of the model in the context of current care
Discuss any next steps for future research, with a specific view to applicability and generalisability of the model

D = items relevant only to the development of a prediction model; E = items relating solely to the evaluation of a prediction model; items without a label apply to both development and evaluation.

† Separately for all model building approaches.

‡ TRIPOD-Cluster is a checklist of reporting recommendations for studies developing or validating models that explicitly account for clustering or explore heterogeneity in model performance (eg, at different hospitals or centres).

§ Relates to the analysis code, for example, any data cleaning, feature engineering, model building, and evaluation.

¶ Relates to the code to implement the model to get estimates of risk for a new individual.

Your answers stay in this browser. Nothing is sent to a server.

What the TRIPOD+AI checklist covers

TRIPOD+AI, Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis, is the reporting guideline for diagnostic and prognostic prediction model studies. It covers studies that develop a model, evaluate (validate) an existing model, or both, and it applies whether the model is a regression model or a machine learning method such as a random forest or a neural network.

The statement was published in the BMJ in 2024 and replaces the TRIPOD 2015 checklist. Items marked D apply only to model development and items marked E only to evaluation; the tool above hides the ones that do not fit your study type. For a walk-through of the items reviewers check most, read our guide to TRIPOD+AI reporting.

The sections at a glance

Sections of the TRIPOD+AI checklist and their items
Section	Items	What to report
Title and abstract	1–2	Name the study as developing or evaluating a prediction model, with the population and outcome; follow TRIPOD+AI for Abstracts.
Introduction	3–4	Healthcare context, intended use and users, known health inequalities, and objectives.
Methods	5–17	Data sources and dates, participants, data preparation, outcome, predictors, sample size, missing data, analysis, class imbalance, fairness, model output, development versus evaluation data, ethics.
Open science	18a–18f	Funding, conflicts of interest, protocol, registration, data sharing and code sharing.
Patient and public involvement	19	Any involvement in design, conduct, reporting or dissemination, or a statement that there was none.
Results	20–24	Participant flow and characteristics, events per analysis, the full model specification, performance with confidence intervals, and model updating.
Discussion	25–27	Interpretation including fairness, limitations, and how the model would be used in current care.

How to complete and submit it

Choose the study type. Development, evaluation, or both. Studies that develop a model and then evaluate it in new data report every item.
Give a location for every item. Prediction model papers put a lot in supplements, such as the full model (item 22) or code (18f), so name the supplement as well as the section.
Explain anything not applicable. “Not applicable: no class imbalance methods used” is clearer than an empty cell.
Complete the abstract checklist on the second tab if the journal asks for it, and download each Word file.
Say that you followed TRIPOD+AI in the methods and cite the statement.

Items 18b and 18e ask for conflicts of interest and data availability. Our conflict of interest statement generator and data availability statement generator write both statements in the wording journals use.

Frequently asked questions

What is the TRIPOD+AI checklist?: TRIPOD+AI is the reporting guideline for studies that develop or evaluate a clinical prediction model, whether the model uses regression or machine learning. Its checklist has 27 items, 52 with sub-items, running from the title to the discussion, plus a separate 13-item checklist for abstracts. It was published in the BMJ in April 2024 by Collins and colleagues.
Does TRIPOD+AI replace TRIPOD 2015?: Yes. The authors state that the TRIPOD+AI checklist supersedes the TRIPOD 2015 checklist, which should no longer be used. Use TRIPOD+AI even if your model is a logistic or Cox regression and involves no artificial intelligence: the “+AI” means the guidance now also covers machine learning methods.
What do D and E mean in the TRIPOD+AI checklist?: D marks items that apply only to studies developing a prediction model, such as the choice of predictors (9a) and the full model specification (22). E marks items that apply only to studies evaluating a model, such as model updating (12f, 24). Items without a label apply to both. Choose your study type above and the tool hides the items that do not apply.
What is new in TRIPOD+AI compared with TRIPOD 2015?: TRIPOD+AI adds or expands items on health inequalities (3c), data preparation (7), class imbalance (13), fairness (14), open science (18a to 18f: funding, conflicts of interest, protocol, registration, data sharing and code sharing), patient and public involvement (19), and usability of the model in current care (27a to 27c). Fairness is also built into several other items, and the wording covers machine learning as well as regression.
What is the TRIPOD+AI for Abstracts checklist?: A 13-item checklist for the abstract of a prediction model study, which item 2 of the main checklist points to. It covers the title, background, objectives, data sources, eligibility and setting, outcome, model building and validation, performance measures, participants and events, predictors, performance estimates, interpretation and registration. Items 7 and 10 apply only to development studies. Open it with the “Abstract (13 items)” tab.
How do I fill in the TRIPOD+AI checklist?: For each item, write where your manuscript reports it, such as “Methods, Sample size, p. 6; Supplement S2”. If an item does not apply, for example class imbalance methods in a study that used none, mark it not applicable and say why. Download the Word file and upload it as a supplementary file with your submission.
Which TRIPOD checklist should I use for clustered data or a systematic review?: Use TRIPOD+AI as the main checklist. For models developed or validated in clustered data, such as several hospitals, also consult TRIPOD-Cluster. Systematic reviews and meta-analyses of prediction model studies use TRIPOD-SRMA, and study protocols use TRIPOD-P. For risk of bias in prediction model studies, use PROBAST rather than a reporting checklist.
Is TRIPOD+AI a quality assessment tool?: No. TRIPOD+AI says what to report, not how to develop or evaluate a model, and it is not a quality appraisal tool. A well-reported model can still be poorly developed. To judge the risk of bias of a prediction model study, the authors refer readers to PROBAST.
Can I reuse the TRIPOD+AI checklist?: Yes. The TRIPOD+AI statement was published open access under the Creative Commons Attribution (CC BY 4.0) licence, so the checklist can be copied and adapted as long as you cite Collins et al., BMJ 2024;385:e078378.
Are my answers stored anywhere?: No. The checklist runs in your browser and nothing is sent to a server. Your answers are kept in this browser’s local storage so they are still there when you come back; use “Clear answers” to remove them.

Sources

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. doi:10.1136/bmj-2023-078378. Distributed under CC BY 4.0.
Collins GS, Reitsma JB, Altman DG, Moons KGM. Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): the TRIPOD statement. Ann Intern Med 2015;162:55-63.
Snell KIE, Levis B, Damen JAA, et al. Transparent reporting of multivariable prediction models for individual prognosis or diagnosis: checklist for systematic reviews and meta-analyses (TRIPOD-SRMA). BMJ 2023;381:e073538.
TRIPOD Group. TRIPOD statement website.
EQUATOR Network. TRIPOD reporting guideline.

Content updated November 15, 2026. This is an independent tool and is not affiliated with the TRIPOD Group. Check your target journal’s instructions for the form it requires.