Researchers highlight Epic Deterioration Index lack of transparency

The EPIC Deterioration Index (EDI), which helps identify when to move a patient in or out of intensive care, is only moderately successful in differentiating low-risk and high-risk patients, according to researchers. Physicians Vishal Khetpal, MD, and Nishant Shah, MD concluded that a research study involving 392 Covid-19 patients found that the Deterioration Index was moderately successful in discriminating between low-risk patients and those at high risk of ICU transfer, ventilation, or death. Rolled out at speed during the COVID-19 pandemic, t he EDI is a propr itary machine learning-based algorithm developed by p rivate electronic health record company Epic Systems that helps identify when to move a patient in or out of intensive care. Epic is said to hold over 250 million patients' electronic records in the US , and h as been taken up by hundreds of hospitals across the US. However, despite general concerns about medical system accuracy and bias, Epic chose to limit expert a cce ss to raw data and equations and failed to have the system independently validated or peer-reviewed before launch, the reseachers said. The finding raised questions about Epic System's governance and commitment to transparency and accountability, as well as about the EDI's accuracy and reliability. System 🤖 EPIC Deterioration Index Developer: Epic Systems Corporation Country: USA Sector: Healthcare Purpose: Predict patient outcomes Technology: Machine learning Issue: Accountability; Accuracy/reliability; Fairness; Transparency

Date it happened

2021-05-01

Organisation involved

Parkview Health; University of Michigan; Multiple

Product, system or model

Epic Deterioration Indec (EDI)

This is a record of what was reported, not a finding that anyone broke the law. If it names your organisation and you believe it is wrong, the corrections process is free and open to everyone.