ClearMed
BriefingsDrug & Biomarker IndexSaved (0)
All Oncology BriefingsNature Medicine
OncologyReal-World Cohort3 min read

AI-assisted scoping review of code sharing in clinical prediction model research

These findings underscore the need for clearer expectations beyond code availability, including documentation, dependency specification and executable structure.

Published in Nature MedicineObservational cohort study
3/5 Observational
Evidence Meter · 3/5

Real-World Observation

Finds meaningful patterns in human health records or cohorts, though observational data cannot prove direct cause.

Source

Real-World Observation

3/5

Strengthening these practices may improve the usability of clinical prediction model studies and support their deployment in real-world clinical settings.

What Changes?

Oncology · Real-World Cohort

Current Standard

Conventional oncology practice relies on standard-of-care monitoring and existing regimens prior to this observational cohort study.

New Finding

Repository assessment showed substantial heterogeneity in reproducibility features.

3 Key Takeaways

Plain English Summary
  1. 01Core Finding

    Analytical code supports independent assessment of these processes, yet its availability in the literature remains limited.

  2. 02Study Evidence

    Tested in a observational cohort study at the real-world cohort stage.

  3. 03Practical Meaning

    Observational cohort findings reveal real-world links, though lifestyle or unmeasured factors cannot be fully ruled out without an RCT.

What This Means for You

Repository assessment showed substantial heterogeneity in reproducibility features. These findings underscore the need for clearer expectations beyond code availability, including documentation, dependency specification and executable structure.

Strengthening these practices may improve the usability of clinical prediction model studies and support their deployment in real-world clinical settings.

The Discovery

Clinical prediction models are increasingly deployed to support diagnostic and prognostic decisions, making reproducibility essential for assessing their reliability and generalizability. Analytical code supports independent assessment of these processes, yet its availability in the literature remains limited. This scoping review quantifies current practices in sharing analytical code to inform the development of TRIPOD-Code, a reporting guideline for code availability and reproducibility. Here a large-language-model-assisted pipeline was developed to screen articles citing TRIPOD or TRIPOD+AI, extract repository links and assess retrieved repositories against 14 predefined reproducibility-related features. Among 3,967 articles, 482 (12.2%) included code-sharing statements. Sharing prevalence varied widely by journal and country.

How It Works, Simply

This oncology research focuses on AI model. Researchers evaluated it using an observational cohort study at the real-world cohort stage of medical research.

3 Key Takeaways

Clinical Takeaways
  1. 01Core Finding

    Analytical code supports independent assessment of these processes, yet its availability in the literature remains limited.

  2. 02Study Evidence

    Tested in a observational cohort study at the real-world cohort stage.

  3. 03Practical Meaning

    Observational cohort findings reveal real-world links, though lifestyle or unmeasured factors cannot be fully ruled out without an RCT.

01

Biological Mechanism & Study Design

Study Design & Cohort

Observational cohort study

Observational cohort study · Real-World Cohort

Biological Pathway & Modality

AI model

Intervention: AI-assisted scoping review of code sharing in clinical prediction model

02

Experimental Findings & Endpoints

Clinical prediction models are increasingly deployed to support diagnostic and prognostic decisions, making reproducibility essential for assessing their reliability and generalizability.

Analytical code supports independent assessment of these processes, yet its availability in the literature remains limited. This scoping review quantifies current practices in sharing analytical code to inform the development of TRIPOD-Code, a reporting guideline for code availability and reproducibility.

Here a large-language-model-assisted pipeline was developed to screen articles citing TRIPOD or TRIPOD+AI, extract repository links and assess retrieved repositories against 14 predefined reproducibility-related features. Among 3,967 articles, 482 (12.2%) included code-sharing statements.

Sharing prevalence varied widely by journal and country.

03

Clinical Translation & Practice Context

Repository assessment showed substantial heterogeneity in reproducibility features.

These findings underscore the need for clearer expectations beyond code availability, including documentation, dependency specification and executable structure.

Strengthening these practices may improve the usability of clinical prediction model studies and support their deployment in real-world clinical settings.

03

The Bottom Line · Clinical Pearl

“Strengthening these practices may improve the usability of clinical prediction model studies and support their deployment in real-world clinical settings.”

Published in Nature Medicine · Current Stage: Real-World Cohort

Oncology · Real-World Cohort
Study Type
Observational cohort study

Observational cohort study

Evidence Score
3/5 · Real-World Observation
Evidence Meter · 3/5

Real-World Observation

Finds meaningful patterns in human health records or cohorts, though observational data cannot prove direct cause.

Finds meaningful patterns in human health records or cohorts, though observational data cannot prove direct cause.

Who It Applies To
Observational cohort study in oncology care
Intervention & Mechanism
AI-assisted scoping review of code sharing in clinical prediction model

AI model

Key Limitation
Observational cohort findings reveal real-world links, though lifestyle or unmeasured factors cannot be fully ruled out without an RCT.

Translational Stage

LEVEL 2 · Real-World Cohort

Classified as Observational cohort study along the clinical evidence continuum

  1. Lab & Discovery
  2. Early Trials
  3. Human Evidence
  4. 4Real-World CohortActive

Indexed Clinical Entities

Clinical AI & Deep Learning Models

Original Publication & AMA Citation

AI-assisted scoping review of code sharing in clinical prediction model research. Nature Medicine. Published 2026;Oct 9. Accessed November 15, 2026. https://pubmed.ncbi.nlm.nih.gov/42855584/

Nature Medicine

https://pubmed.ncbi.nlm.nih.gov/42855584/

Read Original Paper

Continue Reading

Related in Oncology

View all Oncology briefings →
ONCOLOGY · CLINICAL PRACTICE

Neoadjuvant short-course radiotherapy followed by camrelizumab and chemotherapy versus standard of care for locally advanced rectal cancer (UNION): a prespecified interim analysis of a multicentre, open-label, randomised, phase 3 trial

In this interim analysis, short-course radiotherapy followed by camrelizumab and CAPOX significantly improved 3-year event-free survival in patients with locally advanced rectal cancer, with a safety profile similar to long-course chemoradiotherapy followed by CAPOX. Comparison of this promising new treatment strategy with contemporary total neoadjuvant therapy strategies is warranted.

5/5 Gold Standard
Evidence Meter · 5/5

Gold-Standard Human Evidence

Backed by randomized controlled human trials or a formal meta-analysis pooling multiple clinical studies.

1 min readThe Lancet Oncology
ONCOLOGY · REAL-WORLD COHORT

Curative-intent management of de novo metastatic nasopharyngeal carcinoma: an evidence-based consensus guideline from the International Collaborating Oncology Organisations

This guideline provides a practical framework for standardisation of clinical management of de novo metastatic nasopharyngeal carcinoma and identification of key evidence gaps for future research. We recommend that curative-intent treatment should be offered to medically fit patients with no more than three metastatic lesions in two or fewer organs at baseline or following response to systemic therapy; multimodality treatment (comprising immunochemotherapy, locoregional radiotherapy, metastasis-directed therapy, and maintenance therapy) is recommended.

3/5 Observational
Evidence Meter · 3/5

Real-World Observation

Finds meaningful patterns in human health records or cohorts, though observational data cannot prove direct cause.

1 min readThe Lancet Oncology
ONCOLOGY · REAL-WORLD COHORT

Development of the Children's Oncology Group long-term follow-up guideline for cancer genetics evaluation in childhood cancer survivors

Raising awareness of this guideline is crucial to enhancing the integration of genetics into survivorship care and improving health outcomes for childhood cancer survivors with CPS. This Policy Review details the importance, development, and anticipated impact of this new guideline.

3/5 Observational
Evidence Meter · 3/5

Real-World Observation

Finds meaningful patterns in human health records or cohorts, though observational data cannot prove direct cause.

1 min readThe Lancet Oncology