| Abstrakt: |
Researchers from the University of Pennsylvania conducted a study to evaluate the predictive performance of an electronic health record-based 6-month mortality risk model for triggering palliative care consultation among different patient groups. The study found that there were differences in accuracy, false positive rate, and false negative rate among patients based on factors such as race, ethnicity, age, insurance status, and socioeconomic status. The researchers concluded that evaluating predictive performance is crucial in addressing algorithmic inequities in healthcare and emphasized the need for oversight, monitoring, and accountability in healthcare algorithms. The study was funded by the NIH National Library of Medicine and has been peer-reviewed. [Extracted from the article] |
