Abstract 303: Evaluating The Utility Of A Predictive Analytic For Reducing Intensive Care Unit Readmissions

Bryce Benson, Sooin Lee,Ashwin Belle

Circulation(2022)

引用 0|浏览0
暂无评分
摘要
Background: The reported incidence of readmission (bounceback) to the ICU ranges from 4% to 10% of the total number of ICU admissions. Studies have shown that patients readmitted to the ICU have significantly higher mortality rates, longer lengths of stay, and higher costs than patients who survive the ICU and are not readmitted. Being able to tell whether patients have high or low risks of returning to the ICU could help physicians optimize the timing of ICU discharge. Hypothesis: A single lead continuous ECG based real-time analytic can help reduce ICU readmission rates. Methods: This retrospective analysis considered any admission back to the ICU following transfer to a lower-level care during the same hospitalization as an ICU Readmission. This study included 65 consecutive Michigan Medicine ICU patients who were transferred to a lower level of care and had ECG waveforms collected in real time. The Analytic for Hemodynamic Instability – Predictive Indicator (AHI-PI) processed the ECG to predict the likelihood of a future episode of hemodynamic instability with High, Moderate, and Low Risk indicators, which are updated every 2 minutes. Results: Of the 65 consecutive patients transferred from ICU to a lower-level of care, 14 patients (22%, 14/65) had AHI-PI High Risk indication prior to transfer. The remaining 51 patients (78%, 51/65) had AHI-PI Moderate or Low Risk indication prior to transfer. Of the AHI-PI High Risk indicated subgroup, 29% of patients had ICU Readmissions. In the other subgroup, only 10% did. This means that physicians with access to AHI-PI Risk indicators could potentially reduce ICU Readmission rates by 66% (from 29% to 10%). Further, the subgroup with no AHI-PI High Risk indication had an average of 73 hours (median 41 hours) of AHI-PI Moderate or Low Risk indication prior to transfer. Conclusion: Results indicate that patients with AHI-PI High Risk indication prior to transfer had higher risks of ICU Readmission, compared to those without. Prolonged AHI-PI Moderate or Low Risk indication prior to transfer can potentially help physicians decide to transfer patients earlier than otherwise. This can ultimately help reduce lengths of stay and improve bed and resource utilization in the ICU. Footnote: AHI is an FDA-cleared device for sale in the US.
更多
查看译文
关键词
intensive care unit,predictive analytic
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要