From Climate Shocks to Zoonotic Spillover Risk: A Systematic Review with Quantitative Evidence Synthesis of Climate-Mediated Pathways and One Health Adaptation in Africa

Zoonotic pathogens account for ~60% of known human infectious diseases and ~75% of emerging infectious diseases. The Quadripartite (FAO/WHO/WOAH/UNEP) identifies Africa as a hotspot due to high biodiversity, extensive human-wildlife-livestock contact, rapid land-use change, and constrained surveillance. Climate change disproportionately affects Africa. Mechanistic pathways linking climate shocks to spillover have been described piecemeal: (i) rainfall/ENSO flooding of dambo habitats driving Rift Valley fever (RVF) epizootics; (ii) drought concentration of wildlife/livestock/humans around water driving anthrax; (iii) rainfall-driven Mastomys natalensis dynamics driving Lassa fever seasonality; (iv) deforestation/forest fragmentation increasing human-wildlife contact driving Ebola virus disease and mpox spillover; (v) temperature-driven vector/pathogen range shifts; (vi) extreme storms/cyclones disrupting health systems and displacing populations. Prior reviews focus on single pathogen, single country, single climate driver, and English-language literature only, systematically under-representing Francophone (West/Central Africa), Lusophone (Mozambique, Angola), Arabic-context (North Africa, Sahel), and Swahili (Tanzania, East Africa) institutional evidence. Since 2020, One Health governance architecture has expanded rapidly: Quadripartite One Health Joint Plan of Action 2022-2026, World Bank Health Security Program for Western/Central Africa ($500M, 2023), Pandemic Fund-financed Strengthening Disease Surveillance and Response from a One Health Perspective in Southern Africa (2025, $35.8M). Evidence mapping of where primary research investment has/has not kept pace with this architecture is needed. Conventional meta-analysis pooling of effect sizes is not defensible due to heterogeneity in exposure operationalisation (absolute rainfall mm vs standardized anomaly vs ENSO phase), outcome definitions (case counts vs outbreak occurrence vs seroprevalence vs vector density), lag structures, and inconsistent effect-size reporting. A SWiM-structured vote-counting, harvest plot, evidence mapping, and GRADE appraisal is more transparent.