Suppr超能文献

用于监测 COVID-19 阶段的最优控制图选择:以美国每日死亡人数为例的研究

The optimal control chart selection for monitoring COVID-19 phases: a case study of daily deaths in the USA.

机构信息

School of Mathematics and Statistics, Xi'an Jiaotong University, Xi'an, Shaanxi 710049, China.

Department of Statistics, University of WAH, Taxila 47040, Pakistan.

出版信息

Int J Qual Health Care. 2023 Aug 11;35(3). doi: 10.1093/intqhc/mzad058.

Abstract

Epidemiologists frequently adopt statistical process control tools, like control charts, to detect changes in the incidence or prevalence of a specific disease in real time, thereby protecting against outbreaks and emergent health concerns. Control charts have proven essential in instantly identifying fluctuations in infection rates, spotting emerging patterns, and enabling timely reaction measures in the context of COVID-19 monitoring. This study aims to review and select an optimal control chart in epidemiology to monitor variations in COVID-19 deaths and understand pandemic mortality patterns. An essential aspect of the present study is selecting an appropriate monitoring technique for distinct deaths in the USA in seven phases, including pre-growth, growth, and post-growth phases. Stage-1 evaluated control chart applications in epidemiology departments of 12 countries between 2000 and 2022. The study assessed various control charts and identified the optimal one based on maximum shift detection using sample data. This study considered at Shewhart ($\bar X$, $R$, $C$) control charts and exponentially weighted moving average (EWMA) control chart with smoothing parameters λ = 0.25, 0.5, 0.75, and 1 were all investigated in this study. In Stage-2, we applied the EWMA control chart for monitoring because of its outstanding shift detection capabilities and compatibility with the present data. Daily deaths have been monitored from March 2020 to February 2023. Control charts in epidemiology show growing use, with the USA leading at 42% applications among top countries. During the application on COVID-19 deaths, the EWMA chart accurately depicted mortality dynamics from March 2020 to February 2022, indicating six distinct stages of death. The third and fifth waves were extremely catastrophic, resulting in a considerable loss of life. Significantly, a persistent sixth wave appeared from March 2022 to February 2023. The EWMA map effectively determined the peaks associated with each wave by thoroughly examining the time and amount of deaths, providing vital insights into the pandemic's progression. The severity of each wave was measured by the average number of deaths $W5(1899),\gt,W3(1881),\gt,W4(1393),\gt,W1(1036),\gt,W2(853),\gt,(W6(473)$. The USA entered a seventh phase (6th wave) from March 2022 to February 2023, marked by fewer deaths. While reassuring, it remains crucial to maintain vaccinations and pandemic control measures. Control charts enable early detection of daily COVID-19 deaths, providing a systematic strategy for government and medical staff. Incorporating the EWMA chart for monitoring immunizations, cases, and deaths is recommended.

摘要

流行病学家经常采用统计过程控制工具,如控制图,实时检测特定疾病发病率或患病率的变化,从而防止疫情爆发和突发健康问题。控制图在 COVID-19 监测中已经证明对于即时识别感染率波动、发现新出现的模式以及及时采取反应措施至关重要。本研究旨在回顾并选择一种用于监测 COVID-19 死亡人数变化和了解大流行死亡率模式的最佳流行病学控制图。本研究的一个重要方面是选择一种合适的监测技术,用于监测美国七个阶段的不同死亡人数,包括增长前、增长和增长后阶段。第一阶段评估了 2000 年至 2022 年间 12 个国家流行病学部门的控制图应用情况。该研究评估了各种控制图,并根据使用样本数据检测到的最大偏移量确定了最佳控制图。本研究考虑了 Shewhart($\bar X$,$R$,$C$)控制图和指数加权移动平均(EWMA)控制图,其中$\lambda$=0.25、0.5、0.75 和 1 的平滑参数。在第二阶段,我们应用了 EWMA 控制图进行监测,因为它具有出色的偏移检测能力并且与现有数据兼容。自 2020 年 3 月至 2023 年 2 月,每天都对死亡人数进行了监测。控制图在流行病学中的应用越来越多,美国在应用最多的国家中处于领先地位,占 42%。在 COVID-19 死亡应用中,EWMA 图准确地描绘了 2020 年 3 月至 2022 年 2 月的死亡率动态,表明有六个不同的死亡阶段。第三波和第五波极为灾难性,导致大量生命损失。值得注意的是,从 2022 年 3 月到 2023 年 2 月,出现了持续的第六波。EWMA 图通过彻底检查死亡的时间和数量,有效地确定了与每一波相关的峰值,为了解大流行的进展提供了重要的见解。通过平均死亡人数$W5(1899),\gt,W3(1881),\gt,W4(1393),\gt,W1(1036),\gt,W2(853),\gt,(W6(473)$来衡量每一波的严重程度。从 2022 年 3 月到 2023 年 2 月,美国进入了第七阶段(第六波),死亡人数较少。虽然令人安心,但仍需要保持疫苗接种和大流行控制措施。控制图可以帮助早期发现每日 COVID-19 死亡人数,为政府和医务人员提供系统的策略。建议采用 EWMA 图来监测免疫接种、病例和死亡人数。

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验