Social Development and Health Promotion Research Center, Kermanshah University of Medical Sciences, Kermanshah, Iran.
Mechanical Engineering, Poznan University of Technology, Poznan, Poland.
Disaster Med Public Health Prep. 2022 Feb;16(1):187-193. doi: 10.1017/dmp.2020.312. Epub 2020 Sep 3.
The UK is one of the epicenters of coronavirus disease (COVID-19) in the world. As of April 14, there have been 93 873 confirmed patients of COVID-19 in the UK and 12 107 deaths with confirmed infection. On April 14, it was reported that COVID-19 was the cause of more than half of the deaths in London.
The present paper addresses the modeling and forecasting of the outbreak of COVID-19 in the UK. This modeling must be accomplished through a 2-part time series model to study the number of confirmed cases and deaths. The period we aimed at a forecast was 46 days from April 15 to May 30, 2020. All the computations and simulations were conducted on Matlab R2015b, and the average curves and confidence intervals were calculated based on 100 simulations of the fitted models.
According to the obtained model, we expect that the cumulative number of confirmed cases will reach 282 000 with an 80% confidence interval (242 000 to 316 500) on May 30, from 93 873 on April 14. In addition, it is expected that, over this period, the number of daily new confirmed cases will fall to the interval 1330 to 6450 with the probability of 0.80 by the point estimation around 3100. Regarding death, our model establishes that the real case fatality rate of the pandemic in the UK approaches 11% (80% confidence interval: 8%-15%). Accordingly, we forecast that the total death in the UK will rise to 35 000 (28 000-50 000 with the probability of 80%).
The drawback of this study is the shortage of observations. Also, to conduct a more exact study, it is possible to take the number of the tests into account as an explanatory variable besides time.
英国是世界上冠状病毒病（COVID-19）的中心之一。截至 4 月 14 日，英国已确诊 COVID-19 患者 93873 例，感染确诊死亡 12107 例。4 月 14 日，有报道称 COVID-19 是伦敦半数以上死亡的原因。
本文旨在对英国 COVID-19 的爆发进行建模和预测。这种建模必须通过一个两部分的时间序列模型来研究确诊病例和死亡人数。我们的预测期为 2020 年 4 月 15 日至 5 月 30 日的 46 天。所有的计算和模拟都是在 Matlab R2015b 上进行的，平均曲线和置信区间是基于拟合模型的 100 次模拟计算得出的。
根据得到的模型，我们预计到 5 月 30 日，累计确诊病例数将达到 282000 例，置信区间为 242000 至 316500 例，而 4 月 14 日的确诊病例数为 93873 例。此外，预计在此期间，每日新增确诊病例数将下降至 1330 至 6450 例，点估计值为 3100 左右，概率为 0.80。至于死亡，我们的模型确定英国大流行的实际病死率接近 11%（80%置信区间：8%-15%）。因此，我们预测英国的总死亡人数将上升到 35000 人（80%置信区间：28000-50000 人）。
本研究的缺点是观测数据不足。此外，为了进行更准确的研究，可以将检测数量作为时间之外的一个解释变量。
