A screening tool that predicts short-term mortality of older adults could potentially be used to improve care for high-risk patients visiting emergency departments and other health care settings, including nursing homes.

A recent study published in JAMA has highlighted the effectiveness of the Geriatric End-of-Life Screening Tool (GEST) in predicting 6-month mortality in older patients visiting emergency departments.

Researchers found that GEST is a more effective tool for identifying older adults at high risk of short-term mortality compared to traditional serious illness criteria. They suggested that GEST’s ability to use readily available electronic health record (EHR) data makes it a practical option for resource-constrained EDs, potentially improving the identification and care of high-risk patients.

For skilled nursing operators, this means that implementing GEST could significantly enhance the accuracy of mortality risk predictions, allowing for better-targeted end-of-life care interventions.

“The findings of this study suggest that both serious illness criteria and GEST identified older ED patients at risk for 6-month mortality, but GEST offered more useful screening characteristics,” researchers wrote. “Future trials of serious illness interventions for high mortality risk in older adults may consider transitioning from diagnosis code criteria to GEST, an automatable EHR-based algorithm.”

The research included a cohort of 82,371 ED encounters by 40,505 patients aged 65 and older at Beth Israel Deaconess Medical Center in Boston, Massachusetts, from 2017 to 2021. The primary outcome measured was the 6-month mortality rate following an ED visit.

GEST, a logistic regression algorithm using data from EHRs, was compared against common serious illness criteria such as stroke, liver disease, cancer, lung disease, and advanced age. The results showed that GEST had a robust area under the receiver operating characteristic curve (AUROC) of 0.79, indicating high accuracy in predicting 6-month mortality. The study found that GEST outperformed the serious illness criteria across various thresholds of mortality risk.

The data revealed that 53.4% of the encounters involved patients with serious illnesses, with GEST showing a sensitivity of 77.4% and a specificity of 50.5%. Adjusting the GEST cutoffs demonstrated the tool’s flexibility in balancing sensitivity and specificity according to clinical needs. Notably, GEST reclassified 45.1% of patients with serious illnesses as low risk, with an observed mortality rate of 8.1%, and identified 2.6% of patients without serious illnesses as high risk, with an observed mortality rate of 34.3%.