Modeling of Vector-Borne Disease Across Governorates and Districts in Oman, 2020–2024

Introduction: Oman has transitioned from travel-related dengue cases to local outbreaks since 2018, with heterogeneous patterns across governorates and districts. Understanding how climate, population, and vector indicators jointly shape dengue risk at different administrative levels is essential for targeted control. Methods: This study compiled weekly data (2020–2024) on dengue cases, mosquito surveillance, climate, and population from national sources. Using Partial Least Squares Structural Equation Modelling (PLS-SEM) in SmartPLS v4, we modelled constructs for Weather, Population, Vector, and Vector-borne Disease (VBD). Measurement quality was assessed using various statistics and with 5000-sample bootstrapping. Multigroup Analysis (MGA) with permutation and Measurement Invariance of Composite Models (MICOM) tested invariance and compared structural paths across governorates (Muscat, North Al Batinah, Ad Dakhiliyah) and districts (Seeb, Sohar, Bahla). Results: Vector abundance mediated climate and population effects on dengue, with marked spatial heterogeneity. At the governorate level, the Vector → VBD path was strongest in Ad Dakhiliyah (β ≈ 0.436) and negligible in Muscat (β ≈ −0.021); indirect effects from Population and Weather to VBD were significantly higher in Ad Dakhiliyah than comparators. At the district level, Bahla showed stronger Vector → VBD and Weather → Vector relationships than Seeb and Sohar, while Seeb exhibited low explanatory power across paths. MICOM indicated partial measurement invariance, suggesting caution in cross-group comparisons. Conclusions: Dengue risk in Oman is primarily vector-driven but differs by setting. Inland/rural areas are more sensitive to climate–vector dynamics, requiring enhanced surveillance and climate-informed early warning. Urban centers may need models incorporating mobility and behavior. Findings support localized interventions and the integration of trap positivity and density into district-level prediction and control.