ABSTRACT
Prediction of COVID-19 spread plays a significant role in the epidemiology study and government battles against the epidemic. However, the existing studies on COVID-19 prediction are dominated by constant model parameters, unable to reflect the actual situation of COVID-19 spread. This paper presents a new method for dynamic prediction of COVID-19 spread by considering time-dependent model parameters. This method discretises the susceptible-exposed-infected-recovered-dead (SEIRD) epidemiological model in time domain to construct the nonlinear state-space equation for dynamic estimation of COVID-19 spread. A maximum likelihood estimation theory is established to online estimate time-dependent model parameters. Subsequently, an extended Kalman filter is developed to estimate dynamic COVID-19 spread based on the online estimated model parameters. The proposed method is applied to simulate and analyse the COVID-19 pandemics in China and the United States based on daily reported cases, demonstrating its efficacy in modelling and prediction of COVID-19 spread.
ABSTRACT
BACKGROUND: The coronavirus disease 2019 (COVID-19) pandemic has placed unprecedented burden on the delivery of intensive care services worldwide. RESEARCH QUESTION: What is the global point estimate of deaths and risk factors for patients who are admitted to ICUs with severe COVID-19? STUDY DESIGN AND METHODS: In this systematic review and meta-analysis Medline, Embase, and the Cochrane library were searched up to August 1, 2020. Pooled prevalence of participant characteristics, clinical features, and outcome data was calculated with the use of random effects models. Subgroup analyses were based on geographic distribution, study type, quality assessment, sample size, end date, and patient disposition. Studies that reported in-hospital mortality rate of adult patients (age >18 years) with confirmed COVID-19 admitted to an ICU met study eligibility criteria. Critical evaluation was performed with the Newcastle Ottawa Scale for nonrandomized studies. RESULTS: Forty-five studies with 16,561 patients from 17 countries across four continents were included. Patients with COVID-19 who were admitted to ICUs had a mean age of 62.6 years (95% CI, 60.4-64.7). Common comorbidities included hypertension (49.5%; 95% CI, 44.9-54.0) and diabetes mellitus (26.6%; 95% CI, 22.7-30.8). More than three-quarters of cases experienced the development of ARDS (76.1%; 95% CI, 65.7-85.2). Invasive mechanical ventilation was required in 67.7% (95% CI, 59.1-75.7) of case, vasopressor support in 65.9% (95% CI, 52.4-78.4) of cases, renal replacement therapy in 16.9% (95% CI, 12.1-22.2) of cases, and extracorporeal membrane oxygenation in 6.4% (95% CI, 4.1-9.1) of cases. The duration of ICU and hospital admission was 10.8 days (95% CI, 9.3-18.4) and 19.1 days (95% CI, 16.3-21.9), respectively, with in-hospital mortality rate of 28.1% (95% CI, 23.4-33.0; I2 = 96%). No significant subgroup effect was observed. INTERPRETATION: Critically ill patients with COVID-19 who are admitted to the ICU require substantial organ support and prolonged ICU and hospital level care. The pooled estimate of global death from severe COVID-19 is <1 in 3.