Waseel Farhad, Streftaris George, Rudrusamy Bhuvendhraa, Dass Sarat C
School of Mathematical and Computer Sciences, Heriot-Watt University Malaysia, Putrajaya, Malaysia.
Faculty of Mathematics, Kabul University, Kabul, Afghanistan.
Infect Dis Model. 2024 Mar 12;9(2):527-556. doi: 10.1016/j.idm.2024.02.010. eCollection 2024 Jun.
The COVID-19 pandemic has significantly impacted global health, social, and economic situations since its emergence in December 2019. The primary focus of this study is to propose a distinct vaccination policy and assess its impact on controlling COVID-19 transmission in Malaysia using a Bayesian data-driven approach, concentrating on the year 2021. We employ a compartmental Susceptible-Exposed-Infected-Recovered-Vaccinated (SEIRV) model, incorporating a time-varying transmission rate and a data-driven method for its estimation through an Exploratory Data Analysis (EDA) approach. While no vaccine guarantees total immunity against the disease, and vaccine immunity wanes over time, it is critical to include and accurately estimate vaccine efficacy, as well as a constant vaccine immunity decay or wane factor, to better simulate the dynamics of vaccine-induced protection over time. Based on the distribution and effectiveness of vaccines, we integrated a data-driven estimation of vaccine efficacy, calculated at 75% for Malaysia, underscoring the model's realism and relevance to the specific context of the country. The Bayesian inference framework is used to assimilate various data sources and account for underlying uncertainties in model parameters. The model is fitted to real-world data from Malaysia to analyze disease spread trends and evaluate the effectiveness of our proposed vaccination policy. Our findings reveal that this distinct vaccination policy, which emphasizes an accelerated vaccination rate during the initial stages of the program, is highly effective in mitigating the spread of COVID-19 and substantially reducing the pandemic peak and new infections. The study found that vaccinating 57-66% of the population (as opposed to 76% in the real data) with a better vaccination policy such as proposed here is able to significantly reduce the number of new infections and ultimately reduce the costs associated with new infections. The study contributes to the development of a robust and informative representation of COVID-19 transmission and vaccination, offering valuable insights for policymakers on the potential benefits and limitations of different vaccination policies, particularly highlighting the importance of a well-planned and efficient vaccination rollout strategy. While the methodology used in this study is specifically applied to national data from Malaysia, its successful application to local regions within Malaysia, such as Selangor and Johor, indicates its adaptability and potential for broader application. This demonstrates the model's adaptability for policy assessment and improvement across various demographic and epidemiological landscapes, implying its usefulness for similar datasets from various geographical regions.
自2019年12月出现以来，新冠疫情对全球健康、社会和经济状况产生了重大影响。本研究的主要重点是提出一项独特的疫苗接种政策，并使用贝叶斯数据驱动方法评估其对控制马来西亚新冠病毒传播的影响，重点关注2021年。我们采用了一个分区的易感-暴露-感染-康复-接种（SEIRV）模型，纳入了随时间变化的传播率，并通过探索性数据分析（EDA）方法对其进行数据驱动的估计。虽然没有疫苗能保证对该疾病具有完全免疫力，且疫苗免疫力会随时间减弱，但纳入并准确估计疫苗效力以及恒定的疫苗免疫力衰减或减弱因素，对于更好地模拟疫苗诱导的保护随时间的动态变化至关重要。基于疫苗的分布和有效性，我们对疫苗效力进行了数据驱动的估计，计算得出马来西亚的疫苗效力为75%，突出了该模型对该国特定背景的现实意义和相关性。贝叶斯推理框架用于整合各种数据源，并考虑模型参数中的潜在不确定性。该模型与马来西亚的实际数据拟合，以分析疾病传播趋势并评估我们提出的疫苗接种政策的有效性。我们的研究结果表明，这项独特的疫苗接种政策，即在项目初始阶段强调加快接种速度，在减轻新冠病毒传播以及大幅降低疫情高峰和新感染病例方面非常有效。研究发现，采用此处提出的更好的疫苗接种政策，为57%-66%的人口接种疫苗（与实际数据中的76%相对），能够显著减少新感染病例数量，并最终降低与新感染相关的成本。该研究有助于构建一个关于新冠病毒传播和疫苗接种的强大且信息丰富的模型，为政策制定者提供有关不同疫苗接种政策的潜在益处和局限性的宝贵见解，尤其突出了精心规划和高效的疫苗接种推广策略的重要性。虽然本研究中使用的方法专门应用于马来西亚的国家数据，但其在马来西亚的地方区域（如雪兰莪州和柔佛州）的成功应用表明了其适应性和更广泛应用的潜力。这证明了该模型在各种人口统计学和流行病学背景下进行政策评估和改进的适应性，意味着它对来自不同地理区域的类似数据集有用。
