Khumbudzo Mavhunga, Duah Evans, Grobler Estelle, Maluleke Kuhlula
School of Health Systems and Public Health Faculty of Health University of Pretoria Pretoria South Africa.
Department of Library Services University of Pretoria Pretoria South Africa.
Public Health Chall. 2025 Sep 13;4(3):e70116. doi: 10.1002/puh2.70116. eCollection 2025 Sep.
Mathematical modelling plays an important role in public health by enabling the prediction of disease outbreaks, assessment of transmission dynamics and evaluation of intervention strategies. Although widely applied in high-resource settings, its use in underserved contexts remains underexplored. This review aimed to examine and synthesize current evidence on the application of mathematical modelling for predicting and controlling infectious diseases in underserved settings.
A comprehensive and reproducible search was conducted using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and population, intervention, comparison and outcome (PICO) frameworks across databases, including PubMed, Scopus, Medline, ScienceDirect and EBSCOhost. Keywords and Medical Subject Headings (MeSH) terms related to mathematical modelling and infectious disease control were applied. Two reviewers independently screened titles, abstracts and full texts, with a third resolving discrepancies. Thematic analysis and meta-analysis were used for synthesis.
Out of 838 studies screened, 27 (3.2%) met inclusion criteria. Deterministic models were most used, followed by stochastic and agent-based models. Diseases modelled included COVID-19, malaria, tuberculosis (TB), Ebola, Zika, chikungunya, dengue, diphtheria, respiratory infections, visceral leishmaniasis (VL) and Mpox. Modelling predicted the impact of interventions on transmission, with pooled effect size (Ro) of 1.32 ( = 1.3, < 0.0001). However, challenges, such as data underreporting, gaps and inconsistencies, were common, potentially affecting model accuracy and real-world applicability.
Mathematical modelling has demonstrated value in supporting infectious disease control in underserved settings. However, the predominance of deterministic models limits adaptability across diverse contexts. Poor data quality further constrains reliability. Future work should focus on expanding modelling approaches, strengthening data infrastructure and addressing a broader range of diseases. These findings can guide public health policy by supporting data-driven decision-making, improving resource allocation and integrating modelling into outbreak preparedness and response strategies in underserved settings.
数学建模通过预测疾病爆发、评估传播动态以及评价干预策略，在公共卫生领域发挥着重要作用。尽管数学建模在资源丰富的环境中得到广泛应用，但其在服务不足地区的应用仍有待深入探索。本综述旨在审视和综合当前关于数学建模在服务不足地区预测和控制传染病方面应用的证据。
使用系统评价与Meta分析的首选报告项目（PRISMA）以及人群、干预措施、对照和结局（PICO）框架，在包括PubMed、Scopus、Medline、ScienceDirect和EBSCOhost在内的多个数据库中进行全面且可重复的检索。应用了与数学建模和传染病控制相关的关键词及医学主题词（MeSH）。两名评审员独立筛选标题、摘要和全文，由第三名评审员解决分歧。采用主题分析和Meta分析进行综合。
在筛选的838项研究中，27项（3.2%）符合纳入标准。使用最多的是确定性模型，其次是随机模型和基于主体的模型。所建模的疾病包括新冠病毒病、疟疾、结核病、埃博拉病毒病、寨卡病毒病、基孔肯雅热、登革热、白喉、呼吸道感染、内脏利什曼病和猴痘。建模预测了干预措施对传播的影响，合并效应量（Ro）为1.32（ = 1.3， < 0.0001）。然而，诸如数据漏报、差距和不一致等挑战很常见，可能会影响模型的准确性和实际适用性。
数学建模已证明在支持服务不足地区的传染病控制方面具有价值。然而，确定性模型占主导地位限制了其在不同环境中的适应性。数据质量差进一步限制了可靠性。未来的工作应侧重于扩展建模方法、加强数据基础设施以及应对更广泛的疾病。这些发现可通过支持数据驱动的决策、改善资源分配以及将建模纳入服务不足地区的疫情防范和应对策略来指导公共卫生政策。
