Consideration of Sociodemographics in Machine Learning-Driven Sepsis Risk Prediction
Authors' conclusion
The inclusion of sociodemographic data and stratified assessment of performance-essential steps in developing equitable risk prediction tools-are possible but have yet to be consistently adopted.
Study context
- Population
- Studies of machine learning (ML)/artificial intelligence (AI) prediction of sepsis, sepsis-related outcomes or sepsis treatment in adults, extracted from reviews published January 1, 2023–June 30, 2024studies from 13 reviews
- Participants
- 120 participants
- Outcome
- Reporting of sample sociodemographics, sociodemographics as predictors, stratification by sociodemographics or fairness metrics, and lack of sociodemographic consideration as a limitation
Methodological detailsmeta-research, two reviewers, predefined extraction forms, 13 of 96 review studies (14%) met inclusion criteria: 6 systematic and 7 narrative reviews; 120 of 170 extracted studies (71%) included
- Meta-research
- Two reviewers
- Predefined extraction forms
- 13 of 96 review studies (14%) met inclusion criteria: 6 systematic and 7 narrative reviews; 120 of 170 extracted studies (71%) included
Pooled result
As reported in the abstract
Reporting of sociodemographics
- Reported a measure of geography or where data was collected
- 99 of 120 (83%)
- Reported sex/gender
- 80 (67%)
- Reported race/ethnicity
- 24 (20%)
- Reported other sociodemographics
- 4 (3%)
- Stratified performance results by sociodemographics
- 3 (2%)
- Reported formal fairness metrics
- 0
- Reported lack of geographic heterogeneity as a limitation
- 39/120, 33%
Original publication
Consideration of Sociodemographics in Machine Learning-Driven Sepsis Risk Prediction. Crit Care Med. Published 9 June 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →