EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Automatic detection of sexual and reproductive consults of people with disabilities from Electronic Health Records

Mariela Rajngewerc, Sabrina López, Laura Alonso Alemany, Verónica Xhardez, Cecilia Palermo, María Victoria Tiseyra, Milagro Teruel

Oxford Open Digital Health · 13.11.2026

DOI: 10.1093/oodh/oqag026 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Oxford Open Digital Health, online erschienen am 13.11.2026. Verfasst von Mariela Rajngewerc und 6 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

People with disabilities (PWD) often face barriers to accessing appropriate sexual and reproductive health (SRH) care, partly due to the underrepresentation of their needs in clinical data systems. Addressing how health systems can integrate inclusive and representative data practices is therefore essential for achieving equitable access to emerging AI-supported SRH services for PWD. This study responds to this policy challenge by examining how targeted dataset augmentation and intersectional model evaluation can improve AI-enabled decision support tools. We present a case study from Argentina, where clinicians, disability community representatives and data scientists co-developed a disability and SRH taxonomy and used it to annotate 3 300 free-text SRH consultation records for people with disabilities. Machine learning models trained on this dataset achieved 80% accuracy in identifying PWD or SRH-related content and 60% accuracy for records addressing both simultaneously, showing that it is feasible to automatically identify these consults, but at the same time highlighting the increased complexity of intersectional cases. Fairness assessments showed higher error rates for certain minority subgroups, like women in disability-related records and men in SRH-related records, highlighting the need for targeted data augmentation and model tuning. Our results show promise for the intended application, but underscore the importance of inclusive data curation and fairness-focused evaluation in building clinical decision support tools for marginalized populations. We have shown that, even with domain-expert engagement across all stages of project governance, in-depth, situated fairness analyses are crucial to make identifying SRH consults for PWD inclusive and equitable for minority groups.

Zitieren

APA

Rajngewerc, M., López, S., Alemany, L. A., Xhardez, V., Palermo, C., Tiseyra, M. V., & Teruel, M. (2026). Automatic detection of sexual and reproductive consults of people with disabilities from Electronic Health Records. Oxford Open Digital Health. https://doi.org/10.1093/oodh/oqag026

Vancouver

Rajngewerc M, López S, Alemany LA, Xhardez V, Palermo C, Tiseyra MV, et al. Automatic detection of sexual and reproductive consults of people with disabilities from Electronic Health Records. Oxford Open Digital Health. 2026. doi:10.1093/oodh/oqag026

BibTeX
@article{Rajngewerc2026,
  author = {Rajngewerc, Mariela and López, Sabrina and Alemany, Laura Alonso and Xhardez, Verónica and Palermo, Cecilia and Tiseyra, María Victoria and Teruel, Milagro},
  title = {Automatic detection of sexual and reproductive consults of people with disabilities from Electronic Health Records},
  journal = {Oxford Open Digital Health},
  year = {2026},
  doi = {10.1093/oodh/oqag026}
}
Kostenlos starten →