MétaCan
Menu
Back to cohort
Record W4414190713 · doi:10.1101/2025.09.12.25335172

Creating and evaluating a decision support tool to reduce critically low hemoglobin and blood transfusions in-hospital

2025· preprint· en· W4414190713 on OpenAlexaffabout
Michael Fralick, Stephanie Lee, Meggie Debnath, Derek Beaton, B Jones, Vera Dounaevskaia, Yuna Lee, Michael Colacci, Orly Bogler, Frank Rudzicz, Muhammad Mamdani

Bibliographic record

VenuemedRxiv · 2025
Typepreprint
Languageen
FieldBusiness, Management and Accounting
TopicBlood donation and transfusion practices
Canadian institutionsDalhousie UniversityToronto Public HealthUniversity of Toronto
Fundersnot available
KeywordsHemoglobinProspective cohort studyClinical decision support systemCohort studyRetrospective cohort studyAnticoagulantCritically illBlood transfusion

Abstract

fetched live from OpenAlex

ABSTRACT Background Hospitalized patients have multiple risk factors for bleeding including critical illness, interventional procedures, frequent phlebotomy, and use of anticoagulant and antiplatelet medications. It is unknown if a decision support tool to identify inpatients at highest risk of bleeding can prevent critically low hemoglobin and its associated sequelae, such as requiring blood transfusion. Objective To create a decision support tool to identify hospitalized patients at high risk of bleeding sequelae (“high risk”) and to evaluate the impact of this tool on rates of critically low hemoglobin and blood transfusion. Methods We conducted a cohort study of patients hospitalized under general internal medicine at a tertiary-care teaching hospital in Toronto, Ontario. We defined “high risk” as patients with a hemoglobin < 86 g/L, or an absolute decrease in hemoglobin of ≥ 30 g/L, or a platelet count < 50 x 10 9 /L. We then created, implemented, and prospectively evaluated a decision support tool to identify these high-risk patients and alert their clinical team. Our primary outcome was the percentage of patients who received a blood transfusion. Results Our retrospective pre-deployment phase included 6,401 hospitalizations and our prospective phase included 4,274 hospitalizations. The median age of patients was 67 years (IQR 52,80), 43% were female, and the median length of stay was 5 days (IQR 2,10). Overall, 9% had a hemoglobin of 70 g/L or lower and 10% received a blood transfusion in hospital. After model implementation, the median timing of alerts was 1 day after admission (IQR 1,4). The most common trigger for an alert was a hemoglobin < 86 g/L (N=624, 76%), followed by a decrease in hemoglobin ≥ 30 g/L (N=131, 16%), and a platelet count < 50 x 10 9 /L (N=37, 5%). Deployment of the decision support tool was associated with a reduction in the primary outcome (OR 0.78, 95% CI 0.64, 0.96), which was driven by a reduction in red blood cell transfusion (OR 0.78, 95% CI 0.64, 0.96) as opposed to platelet transfusion (OR 1.10, 95% CI 0.46, 2.91). We also observed a reduction in our secondary outcome of critically low hemoglobin (OR 0.81, 95% CI 0.65, 0.99). Interpretation Our decision support tool was associated with a modest reduction in our primary outcome of blood transfusions and secondary outcome of critically low hemoglobin. Because our study was single-centre, we believe future larger studies are needed to validate our findings.

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)0.024
metaresearch head score (Gemma)0.071
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Non-randomized trial · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.024
Threshold uncertainty score0.128

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0240.071
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0030.001
Science and technology studies0.0010.001
Scholarly communication0.0040.002
Open science0.0020.002
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0030.001

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 head0.019
GPT teacher head0.304
Teacher spread0.285 · 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 designNon-randomized trial
Domainnot available
GenreEmpirical

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

Citations0
Published2025
Admission routes2
Has abstractyes

Explore more

Same venuemedRxivSame topicBlood donation and transfusion practicesFrench-language works237,207