Multi-Dimensional Outcome Prediction Algorithm for Hospitalized COVID-19 Patients

点击翻译按钮获取中文摘要

PROJECT SUMMARY Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-mediated coronavirus disease (COVID-19) is an evolutionarily unprecedented natural experiment that causes major changes to the host immune system. Several high risk COVID-19 populations have been identified. Older adults, males, persons of color, and those with certain underlying health conditions (e.g., diabetes mellitus, obesity, etc.) are at higher risk for severe disease from COVID-19. While it is too soon to fully understand the impact of COVID-19 on overall health and well-being, there are already several reports of significant sequelae, which appear to correlate with disease severity. There is a clear and urgent need to develop prediction tests for adverse short- and long-term outcomes, especially for high-risk COVID-19 populations. We hypothesize that complementary multi-dimensional information gathered near the time of symptom onset can be used to predict new onset or worsening frailty, organ dysfunction and death within one year after COVID-19 onset. A single parameter provides limited information and is incapable of adequately characterizing the complex biological responses in symptomatic COVID-19 to predict outcome. Since they were designed for other illnesses, it is unlikely that existing clinical tools, such as respiratory, cardiovascular, and other organ function assessment scores, will precisely assess the long-term prognosis of this novel disease. Our extensive experience in biomarker development suggests that integrating molecular and clinical data increases prediction accuracy of long-term outcomes. We have chosen to test our hypothesis in a population reflecting US-demographics that is at increased risk of adverse outcomes from COVID-19. We will enroll patients, broadly reflecting US demographics, from a hospitalized civilian population in one of the country’s largest metropolitan areas and a representative National Veteran’s population. We anticipate that a prediction test that performs well in this hospitalized patient group will: help guide triaging and treatment decisions and, therefore, reduce morbidity and mortality rates, enhance patient quality of life, and improve healthcare cost-effectiveness. More accurate prognostic information will also assist clinicians in framing goals of care discussions in situations of likely futility and assist patients and families in this decision-making process. Finally, it will provide a logical means for allocating resources in short supply, such as ventilators or therapeutics with limited availability.