Statistical methods for the analysis of multi-antibody data to inform malaria control and elimination strategies

证据转化为实践从实践到政策/人口在全球减少疟疾负担方面已取得重大进展。根据世界卫生组织的《世界疟疾报告》，尽管疟疾预防和治疗服务中断带来了挑战，但2019年至2021年期间疟疾病例和死亡人数有所下降。疟疾负担的减少创造了低传播环境，在这些环境中，估计疟疾负担的传统方法，如寄生虫流行率和昆虫接种率，存在时间变化和低精度。为了克服这些限制，人们对利用免疫学数据进行疟疾监测越来越感兴趣。血清学通过量化先前接触疟疾寄生虫的情况，对传播强度提供了有价值的见解。然而，血清学数据分析和建模的标准化方法的缺乏给疟疾控制规划带来了挑战。目前，常用的血清学数据分析方法有混合模型、可逆催化模型和抗体获取模型。虽然与寄生虫流行率和昆虫学接种率等传统疟疾指标相比，它们提供了有价值的见解，但这些建模方法有不同的局限性，使它们容易对疟疾传播产生有偏差的估计。此外，分析血清流行病学数据的普遍做法涉及独立估计多种抗体反应对疟疾传播的影响，这对准确性和精密度有潜在影响。现有建模方法的局限性强调需要进一步改进和发展建模方法，以提高其准确性和鲁棒性。该博士项目旨在应对这些挑战，促进疟疾监测建模战略的发展，并有可能推广到积极采用血清监测的其他传染病。特别是，该项目将开发多元统计方法，可以克服独立治疗多种抗体反应的常见做法的局限性。

T3 - Evidence into Practice T4 Practice to Policy/PopulationSignificant progress has been made to reduce the burden of malaria globally. Despite the challenges caused by the disruption of malaria prevention and treatment services, malaria cases and deaths dropped between 2019 and 2021 according to the World Health Organization's World Malaria Report. The decrease in the burden of malaria as created low transmission settings where traditional methods for estimating malaria burden, such as parasite prevalence and entomological inoculation rates, suffer from temporal variation and low precision. To overcome these limitations, there is a growing interest in utilizing immunological data for malaria surveillance. Serology offers valuable insights into transmission intensity by quantifying prior exposure to malaria parasites. However, the lack of standardized approaches for analysing and modelling serological data poses challenges for malaria control programs.Currently, the commonly used approaches in analysing serological data are mixture models, reversible catalytic models and antibody acquisition models. While they provide valuable insights when compared to traditional malaria metrics like parasite prevalence and entomological inoculation rate, these modelling approaches have different limitations that make them prone to producing biased estimates of malaria transmission. Additionally, the prevailing practice in the analysis of sero-epidemiological data involves the estimation of the transmission of malaria from multiple antibody responses independently, with potential effects on the accuracy and precision.The limitations of the existing modelling approaches underscore the need for further refinement and development of modelling approaches to enhance their accuracy and robustness. This PhD project aims to address these challenges and contribute to the advancement of modelling strategies for the surveillance of malaria with the potential of extension to other infectious diseases in which serosurveillance is actively employed. In particular, this project will develop multivariate statistical methods that can overcome the limitations of the common practice of treating multiple antibody responses independently.