Korsah Maame Akua, Johnston Stuart T, Tiedje Kathryn, Day Karen, Walker Camelia R, Flegg Jennifer A
School of Mathematics and Statistics, University of Melbourne, Melbourne, Australia.
School of BioSciences and Bio21 Institute, University of Melbourne, Melbourne, Australia.
medRxiv. 2025 Aug 26:2025.08.21.25334195. doi: 10.1101/2025.08.21.25334195.
Globally rising cases of malaria have prompted concentrated efforts to control malaria transmission, utilising various mathematical models to support the Roll Back Malaria agenda. Many existing models with their specific modifications exhibit rigidity, limiting their application to inform malaria control interventions. This study addresses this limitation by employing a reduction technique on a comprehensive malaria control model to derive a simplified system that preserves the essential dynamics of the original system. We validate the accuracy of the reduced model by comparing the two models via Bayesian MCMC. Based on a simulation study, parameter identifiability analysis and sensitivity analysis, we compare the two models and show that the reduced system exhibits similar transmission characteristics as the full model. Our results demonstrate that the reduced model effectively captures the essential behaviour of the comprehensive model, while providing flexibility and computational efficiency, making it a valuable tool for evaluating and implementing malaria control strategies.
全球疟疾病例不断增加，促使人们集中力量控制疟疾传播，利用各种数学模型来支持“遏制疟疾”议程。许多现有模型经过特定修改后仍表现出局限性，限制了它们在为疟疾控制干预措施提供信息方面的应用。本研究通过对一个全面的疟疾控制模型采用约化技术来解决这一局限性，以导出一个保留原始系统基本动态的简化系统。我们通过贝叶斯马尔可夫链蒙特卡罗方法比较这两个模型，验证了约化模型的准确性。基于模拟研究、参数可识别性分析和敏感性分析，我们对这两个模型进行了比较，结果表明约化系统与完整模型具有相似的传播特征。我们的结果表明，约化模型有效地捕捉了综合模型的基本行为，同时提供了灵活性和计算效率，使其成为评估和实施疟疾控制策略的宝贵工具。
