Becker's Hospital Review

37 CIO / HEALTH IT How Kaiser, HCA, Geisinger + more are using AI to save lives By Naomi Diaz H ospitals are using artificial intelligence in emergency rooms, inpatient wards and intensive care units to help identify and treat patients at highest risk for sepsis and an impending car- diac arrest or stroke, e Wall Street Journal reported April 10. Here are five AI projects the health systems have created and deployed to help patients, according to the Journal: 1. AI and sepsis response: Durham, N.C.-based Duke University Hos- pital created Sepsis Watch, a deep learning model that provides an ear- ly warning system within Duke Health for patients who are at risk of sepsis. e model can predict sepsis quickly and accurately with data from the hospital's patient records. 2. Advanced alert: Oakland, Calif.-based Kaiser Permanente devel- oped a predictive model called Advance Alert Monitor that scans pa- tient data continuously, assigning scores that predict the risk of trans- fer to the ICU or death. 3. Sepsis prediction: Nashville, Tenn.-based-HCA Healthcare devel- oped a predictive algorithm, SPOT — Sepsis Prediction and Optimi- zation of erapy — to continuously monitor vital signs, lab results, nursing reports and other data. e tool fires an alert directly to nurses the moment that signals converge that indicate impending sepsis. e algorithm was able to detect sepsis six hours earlier and more accu- rately than clinicians. 4. Next-gen analytics. HCA also developed the Next-Gen Analytics for Treatment and Efficiency, or NATE, which uses machine learning to detect other critical or life-threatening conditions such as shock in trauma patients, complications aer surgery and early signs of deteri- oration in all patients. 5. Cancer screening: Danville, Pa.-based Geisinger partnered with Medial EarlySign to identify patients overdue for a colorectal-cancer screening. ey also used a machine-learning algorithm to flag those at higher risk and were able to schedule screenings for 68.1 percent of the patients flagged. n How email warnings can stop snooping hospital employees By Naomi Diaz E mail warnings were 95 percent effective in pre- venting healthcare employees from inappro- priately accessing patient health information, according to an April 13 study published in JAMA. Researchers conducted a nonrandomized controlled trial at a large academic medical center to under- stand the effectiveness of email warnings on reduc- ing repeated unauthorized access to patient health information. Here's what they found: • In the intervention group, who had email warn- ings, 4 out of 219 employees' repeated offenses occurred between 20 and 70 days after the initial unauthorized access. • In the control group, who didn't have access to email warnings, 90 out of the 225 employees accessed protected health information for a sec- ond time, between 20 and 70 days after their initial unauthorized access. • Email warnings were 95 percent effective for reducing repeated offenses. • The nonrandomized controlled trial found that when left unchecked, hospital employees repeatedly committed unauthorized access to patient health information. The study's findings are limited by lack of data on the prevalence of using email warnings to contain unau- thorized access among hospitals. Email warnings may also not be effective for different health systems as the landscape of cybersecurity con- tinues to evolve. n Epic CEO Judy Faulkner net worth hits personal high By Naomi Diaz E pic CEO Judy Faulkner's net worth has exceeded $7 billion, a new personal high, according to an April Forbes report. Ms. Faulkner, 78, of Madison, Wis., who founded the EHR system in her basement in 1979, appeared on Forbes' 10 Richest Self-Made Women in the World list published April 5, with a net worth that increased by $1 billion over the past year. As of April 6, Ms. Faulkner's real-time net worth was $7.3 billion. Ms. Faulkner's wealth has seen a mostly upward trend since 2013 when her net worth was $2 billion. In 2020 her net worth was listed at $2.5 billion; and in April 2021 it increased to $6 billion. The Verona, Wis.-based EHR giant supports the medical records of more than 253 million U.S. patients. n

• Cover