Departments of Neurology, All India Institute of Medical Sciences (AIIMS), New Delhi, India.
Department of Biostatistics and Bioinformatics, King Abdullah International Medical Research Center/King Saud bin Abdulaziz University for Health Sciences-MNGHA, Riyadh, Saudi Arabia.
PLoS One. 2020 May 12;15(5):e0233074. doi: 10.1371/journal.pone.0233074. eCollection 2020.
The outbreak of coronavirus disease, 2019 (COVID-19), which started from Wuhan, China, in late 2019, have spread worldwide. A total of 5,91,971 cases and 2,70,90 deaths were registered till 28th March, 2020. We aimed to predict the impact of duration of exposure to COVID-19 on the mortality rates increment.
In the present study, data on COVID-19 infected top seven countries viz., Germany, China, France, United Kingdom, Iran, Italy and Spain, and World as a whole, were used for modeling. The analytical procedure of generalized linear model followed by Gompertz link function was used to predict the impact lethal duration of exposure on the mortality rates.
Of the selected countries and World as whole, the projection based on 21st March, 2020 cases, suggest that a total (95% Cl) of 76 (65-151) days of exposure in Germany, mortality rate will increase by 5 times to 1%. In countries like France and United Kingdom, our projection suggests that additional exposure of 48 days and 7 days, respectively, will raise the mortality rates to10%. Regarding Iran, Italy and Spain, mortality rate will rise to 10% with an additional 3-10 days of exposure. World's mortality rates will continue increase by 1% in every three weeks. The predicted interval of lethal duration corresponding to each country has found to be consistent with the mortality rates observed on 28th March, 2020.
The prediction of lethal duration was found to have apparently effective in predicting mortality, and shows concordance with prevailing rates. In absence of any vaccine against COVID-19 infection, the present study adds information about the quantum of the severity and time elapsed to death will help the Government to take necessary and appropriate steps to control this pandemic.
2019 年 12 月，新型冠状病毒病（COVID-19）在中国武汉爆发，现已在全球范围内蔓延。截至 2020 年 3 月 28 日，全球共报告确诊病例 591971 例，死亡 27090 例。本研究旨在预测 COVID-19 暴露时间长短对病死率增加的影响。
本研究采用广义线性模型分析，利用德国、中国、法国、英国、伊朗、意大利和西班牙以及全球共 7 个 COVID-19 感染人数最多的国家的数据，以及 Gompertz 链接函数进行建模。
在所选择的国家和全球范围内，根据 2020 年 3 月 21 日的病例进行预测，假设德国的暴露时间（95%CI）总计为 76（65-151）天，死亡率将增加 5 倍至 1%。在法国和英国等国家，我们的预测表明，分别增加 48 天和 7 天的暴露时间，将使死亡率提高到 10%。对于伊朗、意大利和西班牙，死亡率将上升到 10%，额外暴露时间为 3-10 天。世界的死亡率将每三周增加 1%。每个国家的致命暴露时间预测间隔与 2020 年 3 月 28 日观察到的死亡率一致。
致命暴露时间的预测在预测死亡率方面显然是有效的，并且与当前的死亡率一致。在没有针对 COVID-19 感染的疫苗的情况下，本研究提供了有关严重程度和死亡时间的信息，这将有助于政府采取必要和适当的措施来控制这一流行疫情。
