MétaCan
Menu
← Back to cohort
Record W4411205597 · doi:10.17269/s41997-025-01059-9

Machine learning used to study risk factors for chronic diseases: A scoping review

2025· review· en· W4411205597 on OpenAlexafffundvenue
Mahek Shergill, Steve Durant, Sharon Birdi, Roxana Rabet, Carolyn Ziegler, Shehzad Ali, David L. Buckeridge, Marzyeh Ghassemi, Jennifer Gibson, Ava John‐Baptiste, Jillian Macklin, Melissa D. McCradden, Kwame McKenzie, Parisa Naraei, Akwasi Owusu‐Bempah, Laura C. Rosella, James Shaw, Ross Upshur, Sharmistha Mishra, Andrew D. Pinto

Bibliographic record

VenueCanadian Journal of Public Health · 2025
Typereview
Languageen
FieldMedicine
TopicArtificial Intelligence in Healthcare and Education
Canadian institutionsTrillium Health CentreMcMaster UniversityWellesley InstituteToronto Metropolitan UniversitySickKids FoundationWestern UniversityUniversity of TorontoMcGill UniversitySt. Michael's HospitalHospital for Sick ChildrenPublic Health OntarioToronto Public Health
FundersCanadian Institutes of Health ResearchUniversity of TorontoCanada Research Chairs
KeywordsComputer scienceMedicineRisk analysis (engineering)Artificial intelligence

Abstract

fetched live from OpenAlex

OBJECTIVES: Machine learning (ML) has received significant attention for its potential to process and learn from vast amounts of data. Our aim was to perform a scoping review to identify studies that used ML to study risk factors for chronic diseases at a population level, notably those that incorporated methods to mitigate algorithmic bias. We focused on ML applications for the most common risk factors for chronic disease: tobacco use, alcohol use, unhealthy eating, physical activity, and psychological stress. METHODS: We searched the peer-reviewed, indexed literature using Medline (Ovid), Embase (Ovid), Cochrane Central Register of Controlled Trials and Cochrane Database of Systematic Reviews (Ovid), Scopus, ACM Digital Library, INSPEC, and Web of Science's Science Citation Index, Social Sciences Citation Index, and Emerging Sources Citation Index. Among the included studies, we examined whether bias was considered and identified strategies employed to mitigate bias. SYNTHESIS: The search identified 10,329 studies, and 20 met our inclusion criteria. The studies we identified used ML for a wide range of goals, from prediction of chronic disease development to automating the classification of data to identifying new associations between risk factors and disease. Nine studies (45%) included some discussion of algorithmic bias. Studies that incorporated a broad array of sociodemographic variables did so primarily to improve the performance of a ML model rather than to mitigate potential harms to populations made vulnerable by social and economic policies. CONCLUSION: This work contributes to our understanding of how ML can be used to advance population and public health.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.043
metaresearch head score (Gemma)0.200
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Systematic review · Consensus signal: Systematic review
GenreCandidate signal: Review · Consensus signal: Review
Teacher disagreement score0.043
Threshold uncertainty score0.228

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0430.200
Meta-epidemiology (narrow)0.0020.001
Meta-epidemiology (broad)0.0070.008
Bibliometrics0.0200.017
Science and technology studies0.0020.002
Scholarly communication0.0060.006
Open science0.0030.003
Research integrity0.0050.004
Insufficient payload (model declined to judge)0.0050.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.425
GPT teacher head0.537
Teacher spread0.111 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSystematic review
Domainnot available
GenreReview

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations5
Published2025
Admission routes3
Has abstractyes

Explore more

Same venueCanadian Journal of Public Health→Same topicArtificial Intelligence in Healthcare and Education→French-language works237,207→