意大利COVID-19的观察和估计患病率：是否有可能从医学拭子数据中估计总病例数？
Observed and estimated prevalence of Covid-19 in Italy: Is it possible to estimate the total cases from medical swabs data?
摘要：
在意大利当前的新冠肺炎大流行期间，官方数据是根据纯粹的便利标准用医疗拭子收集的，至少在早期阶段，这使得对表现出明显症状的患者的检查具有特权。然而，有证据表明，无症状患者的比例非常高（例如，G. Aguilar等人，2020; Chugthai等人，2020; Li等人，2020; Mizumoto等人，2020 a，2020 b和Yelin等人，2020年）。在这种情况下，为了估计感染的实际人数（并估计致死率），有必要进行一次适当设计的抽样调查，通过这种调查，有可能计算纳入的概率，从而得出合理的概率推断。一些研究人员提出了基于各种方法的总流行率估计数，包括流行病学模型、时间序列和分析在早期面临流行病的国家收集的数据（Brogi等人，2020年）。在本文中，我们建议通过重新加权Istituto Questiore di Sanità发布的可用官方数据来估计意大利的Covid-19患病率，以获得更具代表性的意大利人口样本。重新加权是通常用于人工修改样本组成以获得与总体更相似的分布的程序（Valliant等人，2018年）。在本文中，我们将使用后分层的官方数据，以获得必要的权重，重新加权他们使用年龄和性别作为后分层变量，从而获得更可靠的估计患病率和致死率。
英文摘要：
During the current Covid-19 pandemic in Italy, official data are collected with medical swabs following a pure convenience criterion which, at least in an early phase, has privileged the exam of patients showing evident symptoms. However, there are evidences of a very high proportion of asymptomatic patients (e. g. Aguilar et al., 2020; Chugthai et al, 2020; Li, et al., 2020; Mizumoto et al., 2020a, 2020b and Yelin et al., 2020). In this situation, in order to estimate the real number of infected (and to estimate the lethality rate), it should be necessary to run a properly designed sample survey through which it would be possible to calculate the probability of inclusion and hence draw sound probabilistic inference. Some researchers proposed estimates of the total prevalence based on various approaches, including epidemiologic models, time series and the analysis of data collected in countries that faced the epidemic in earlier time (Brogi et al., 2020). In this paper, we propose to estimate the prevalence of Covid-19 in Italy by reweighting the available official data published by the Istituto Superiore di Sanità so as to obtain a more representative sample of the Italian population. Reweighting is a procedure commonly used to artificially modify the sample composition so as to obtain a distribution which is more similar to the population (Valliant et al., 2018). In this paper, we will use post-stratification of the official data, in order to derive the weights necessary for reweighting them using age and gender as post-stratification variables thus obtaining more reliable estimation of prevalence and lethality.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
