| Sumario: | This article focuses on the challenges and ethical considerations of using artificial intelligence (AI) to mine electronic health records (EHR) for social and behavioral determinants of health (SBDOH) to support clinical trial recruitment. It highlights that SBDOH documentation in EHRs is often incomplete, inconsistent, and scattered across structured and unstructured fields, limiting AI’s effectiveness and potentially introducing bias in recruitment pools. The article discusses concerns about privacy, patient consent, and the risk of stigmatization, emphasizing the need for transparency, patient choice, and empirical research on patient perspectives regarding AI use of sensitive SBDOH data. It concludes that addressing these issues requires ongoing public engagement, improved documentation practices, and ethical governance to ensure trust and fairness in AI-enabled research recruitment.
|
|---|