Abstract
BACKGROUND: Malaria continues to pose a major public health challenge in sub-Saharan Africa (SSA), where more than 95% of global cases and deaths occur. Despite WHO Global Technical Strategy for Malaria (GTS) targeting a 90% reduction in malaria mortality by 2030, progress is hindered by persistent transmission conditions, fragile health systems, and emerging resistance to drugs and insecticides. Epidemiological models are increasingly applied to guide malaria control, yet their diversity and complexity present both opportunities and limitations for policy use. The aim of the proposed systematic review is to explore the epidemiological models that have been applied to malaria transmission in SSA, discuss how these models have been applied in informing malaria control measures as well as long-term elimination planning, and address the methodological strengths, limitations, challenges in implementing them, and their policy and strategic implications. METHODS: This study presents a systematic review of malaria modeling efforts in SSA, with a focus on the strengths, weaknesses, and practical applications of different approaches. Following PRISMA 2020 guidelines and a PROSPERO-registered protocol, we searched five databases (PubMed, Scopus, LILACS, Web of Science, and African Medicus Index) for studies published up to December 31, 2024. Eligible articles were screened by three independent reviewers using predefined PECO criteria, and data were extracted on study context, model type, interventions, populations, and outcomes. The quality of the methodology used in the modelling studies that were included was determined using the ISPOR-SMDM good research practices framework. This framework assesses major areas of model structure, assumptions, transparency, validation and reporting. Since most of the included studies represented mechanistic epidemiologic transmission models, rather than clinical prediction, studies, a formal risk-of-bias instrument, like PROBAST, was not utilized. Risk of bias was actually not measured, the modelling quality and reporting practices were instead appraised with the help of the ISPOR-SMDM assessment. RESULTS: Following systematic screening, a total of 102 studies met the inclusion criteria. The most prevalent models were transmission-focused models (52.9%, 54 articles), which involved disease dynamics. Intervention models contributed 21.6% (22 articles), optimal control models 9.8% (10 articles) and combined optimal control-cost-effectiveness models 15.7% (16 articles). Key gaps include limited incorporation of drug and insecticide resistance, migration dynamics, and climate variability. CONCLUSION: We conclude that future modeling for SSA must be better tailored to local transmission patterns, age-specific vulnerabilities, and programmatic needs, while promoting open access, transparent methods, and collaborative use cases. Strengthening the alignment between modeling outputs and policy priorities will be critical for achieving effective and sustainable malaria control in the region.