August 24, 2025Open Access

Algorithmic Fairness of QPrediction Cardiometabolic Risk Prediction Models

View Full Paper

Ask AI

Bookmark

Authors

IHInchuen Huynh DUDenise Utochkin AKAlexandros Katsiferis

Discussion

Member takes

Overview

This analysis reveals significant biases in risk predictions for cardiovascular disease, diabetes, and stroke across diverse demographic groups, suggesting the need for improved fairness in clinical models.

Key Points

All tested models showed systematic overprediction of risks, particularly for demographic subgroups, indicating significant calibration issues.
In terms of discrimination, disparities in true positive and true negative rates were noted across ethnicity and socioeconomic factors.
Assessment utilized UK Biobank data, comparing predicted risks against observed cumulative incidence rates and calculating survival Brier scores for calibration.
Inclusion of social determinants improved model performance, yet deploying these models in clinical settings may shift focus onto patients' responsibilities for health outcomes.

Cite This Study

Huynh et al. (2025) studied this question.

synapsesocial.com/papers/68af79a47567bf4f94ff167b https://doi.org/10.1101/2025.08.20.25333669

View Full Paper

Ask AI

Bookmark

August 24, 2025Open Access

Algorithmic Fairness of QPrediction Cardiometabolic Risk Prediction Models

View Full Paper

Ask AI

Bookmark

Authors

IHInchuen Huynh DUDenise Utochkin AKAlexandros Katsiferis

Discussion

Member takes

Overview

Key Points

All tested models showed systematic overprediction of risks, particularly for demographic subgroups, indicating significant calibration issues.
In terms of discrimination, disparities in true positive and true negative rates were noted across ethnicity and socioeconomic factors.
Assessment utilized UK Biobank data, comparing predicted risks against observed cumulative incidence rates and calculating survival Brier scores for calibration.
Inclusion of social determinants improved model performance, yet deploying these models in clinical settings may shift focus onto patients' responsibilities for health outcomes.

Cite This Study

Huynh et al. (2025) studied this question.

synapsesocial.com/papers/68af79a47567bf4f94ff167b https://doi.org/10.1101/2025.08.20.25333669

Also Consider

Synapse has enriched 2 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark