EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Refocusing algorithmic fairness on feature-level bias: A diagnostic approach using dutch EHR data

Cathleen S Parsons iD, Shelley-Ann M Girwar, Sepinoud Azimi iD, Marco R Spruit iD, Marcel R Haas iD

PLOS Digital Health · 26.10.2026 · 5 (10) · e0001105

DOI: 10.1371/journal.pdig.0001105 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Digital Health, online erschienen am 26.10.2026. Verfasst von Cathleen S Parsons und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Engineering Systems & Services, Faculty of Technology, Policy and Management, Delft University of Technology, Delft, The Netherlands. Die Arbeit nennt 60 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

While algorithmic fairness research in healthcare has predominantly focused on disparities in model performance, less attention has been given to the underlying data structures that may drive such disparities. High-level fairness metrics often obscure the deeper feature-level dynamics necessary for a critical and context-aware assessment of fairness. To address this gap, we propose and apply a diagnostic framework termed Feature-Level Bias Identification And Sensemaking (FL-BIAS). As a case study, we conducted a secondary analysis of a retrospective cross-sectional cohort study using electronic health record data from Dutch general practitioners, linked with sociodemographic data from Statistics Netherlands. The dataset included 112,872 patients, of whom 16.2% had a non-Western migration background. Hospitalization in the following year was modeled using Johns Hopkins Aggregated Diagnosis Groups (ADGs). We trained logistic regression and XGBoost models on different subgroup datasets to evaluate performance disparities using fairness metrics. To analyze feature-level contributions to predictions and errors, we applied Kernel Shapley Values and Cohort Shapley methods. Exploratory analysis revealed significant differences in socioeconomic status (SES) and ADG distributions between Dutch and non-Western groups, though Multiple Correspondence Analysis showed minimal structural variation. Mediation analysis indicated that the effect of migration background on hospitalization was largely mediated by SES, with potential unobserved confounding. While standard fairness metrics indicated modest bias in favor of non-Western patients, deeper feature-level analyses revealed subgroup-specific patterns of variable importance that suggest potentially less favorable underlying conditions. For instance, malignancy (ADG 32) had a stronger predictive impact among non-Western patients but contributed less to false negatives compared to Dutch patients, possibly reflecting structural disparities in cancer diagnosis and care. These findings highlight the need for contextual, multi-level evaluations of algorithmic bias. Fairness in healthcare algorithms must be approached as a socio-technical challenge, requiring multidisciplinary collaboration to uncover root causes in the processes that generate the data.

Autor:innen und Institutionen

  • Department of Engineering Systems & Services, Faculty of Technology, Policy and Management, Delft University of Technology, Delft, The Netherlands.
  • Department of Public Health and Primary Care, Leiden University Medical Centre, The Hague, The Netherlands.
  • Leiden Institute of Advanced Computer Science (LIACS), Faculty of Science, Leiden University, Leiden, The Netherlands.

ORCID-Profile: Cathleen S Parsons, Sepinoud Azimi, Marco R Spruit, Marcel R Haas

Zitieren

APA

Parsons, C. S., Girwar, S. M., Azimi, S., Spruit, M. R., & Haas, M. R. (2026). Refocusing algorithmic fairness on feature-level bias: A diagnostic approach using dutch EHR data. PLOS Digital Health, 5(10), e0001105. https://doi.org/10.1371/journal.pdig.0001105

Vancouver

Parsons CS, Girwar SM, Azimi S, Spruit MR, Haas MR. Refocusing algorithmic fairness on feature-level bias: A diagnostic approach using dutch EHR data. PLOS Digital Health. 2026;5(10):e0001105. doi:10.1371/journal.pdig.0001105

BibTeX
@article{Parsons2026,
  author = {Parsons, Cathleen S and Girwar, Shelley-Ann M and Azimi, Sepinoud and Spruit, Marco R and Haas, Marcel R},
  title = {Refocusing algorithmic fairness on feature-level bias: A diagnostic approach using dutch EHR data},
  journal = {PLOS Digital Health},
  year = {2026},
  volume = {5},
  number = {10},
  pages = {e0001105},
  doi = {10.1371/journal.pdig.0001105}
}
Kostenlos starten →