The Epic Deterioration Index (EDI) can identify subsets of high- and low-risk patients with coronavirus disease 2019 (COVID-19) with good discrimination, according to study results published in the Annals of the American Thoracic Society.
The EDI is a prediction model that has been used in hospitals for medical decision-making during the COVID-19 pandemic. However, it has not been independently evaluated and other models have been shown to be biased against vulnerable populations; therefore, researchers independently evaluated the EDI in hospitalized patients with COVID-19 overall and in disproportionately affected subgroups.
The researchers studied patients with COVID-19 admitted to non-intensive care units (ICUs) at Michigan Medicine from March 9 to May 20, 2020. The EDI was calculated at 15-minute intervals to predict a composite outcome of ICU-level care, mechanical ventilation, or in-hospital death.
Of the 392 COVID-19 hospitalizations meeting inclusion criteria, 103 (26%) met the composite outcome at a median of 2.5 days after admission. Of all hospitalizations, 88 (22%) resulted in ICU-level care, 44 (11%) in mechanical ventilation, and 35 (8.9%) in death. Patients who experienced an adverse outcome were older, more likely to be White, and more likely to have a history of cardiac arrhythmias, chronic kidney disease, congestive heart failure, depression, diabetes, hypertension, metastatic cancer, and rheumatoid arthritis or other collagen vascular diseases (all P <.05). Disparate impact analyses demonstrated that EDI predictions did not differ by sex or race.
Patients who met or exceeded an EDI of 68.8 (14% of the study cohort) had a 74% probability (ie, positive predictive value) of experiencing the primary outcome during their hospitalization, with a sensitivity of 39%. Meanwhile, in the subset of 286 patients who had neither been discharged nor experienced the primary outcome at 48 hours, 55 (19%) experienced the composite outcome at some point during the remainder of their hospitalization. A total of 14 (13%) patients who never exceeded an EDI of 37.9 in the first 48 hours of hospitalization had a 90% probability (ie, negative predictive value) of not experiencing the outcome for the remainder of the hospitalization, with a sensitivity of 91%.
“Our study supports, in part, a role of the EDI to identify a small subset of high-risk patients who may benefit from additional resources and higher-level care and another limited subset of low-risk patients who may be cared for safely in lower-acuity settings,” concluded the study authors. “It also suggests opportunities to tailor and improve risk prediction for this condition beyond the EDI as data accumulate on COVID-19 patients.”
Singh K, Valley TS, Tang S, et al. Evaluating a widely implemented proprietary deterioration index model among hospitalized COVID-19 patients. Ann Am Thorac Soc. Published online December 24, 2020. doi:10.1513/AnnalsATS.202006-698OC