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

项目总结 严重急性呼吸综合征冠状病毒2型(SARS-CoV-2)介导的冠状病毒病(新冠肺炎)是 这是一个进化上史无前例的自然实验，它会对宿主免疫系统造成重大变化。 已经发现了几个新冠肺炎的高危人群。老年人、男性、有色人种以及那些 有某些潜在健康状况(如糖尿病、肥胖症等)严重感染的风险更高 来自新冠肺炎的疾病。虽然现在完全了解新冠肺炎对整体健康和健康的影响还为时过早 健康，已经有几个关于严重后遗症的报告，似乎与疾病有关 严肃性。显然，迫切需要制定对不利的短期和长期结果的预测测试， 尤其是对于新冠肺炎的高危人群。我们假设互补的多维空间 在症状出现时间附近收集的信息可以用来预测新的发作或恶化 虚弱、器官功能障碍和新冠肺炎发病后一年内死亡。单个参数提供了 有限的信息，不能充分描述复杂的生物反应 有症状的新冠肺炎可以预测结果。因为它们是为其他疾病而设计的，所以不太可能 现有的临床工具，如呼吸、心血管和其他器官功能评估评分，将 准确评估这种新疾病的长期预后。我们在生物标记物方面的丰富经验 发展表明，结合分子和临床数据可以提高长期预测的准确性 结果。我们选择在反映美国人口结构的人群中测试我们的假设 新冠肺炎不良后果的风险增加。我们将招收患者，广泛反映美国 人口统计数据，来自该国最大的大都市之一的住院平民人口和 有代表性的国家退伍军人的人口。我们预计，在这方面表现良好的预测测试 住院患者组将：帮助指导分诊和治疗决定，从而减少发病率和 提高死亡率，提高患者的生活质量，并提高医疗保健成本效益。更准确 预后信息还将帮助临床医生在可能徒劳的情况下制定护理讨论的目标 并在这一决策过程中帮助患者和家属。最后，它将提供一种符合逻辑的手段 分配供不应求的资源，如呼吸机或治疗药物，但供应有限。

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.