Infectious Disease
Spatial Clustering and Multilevel Determinants of Vaccine Misinformation in Angola, Mozambique, São Tomé & Príncipe, and Guinea: Evidence from Demographic and Health Surveys Ahmed Fentaw Ahmed* Ahmed Fentaw Ahmed University of Brasilia
Abstract:
Background and aim: Vaccine misinformation, expressed through fear, limited awareness, and sociocultural barriers, remains a major obstacle to immunization in sub-Saharan Africa. Subnational spatial patterns and multilevel determinants of misinformation are not well understood.
Objective: This study aimed to identify spatial distribution and hotspots of vaccine misinformation and to examine individual- and community-level determinants in Angola, Mozambique, São Tomé & Príncipe, and Guinea.
Methods and Analysis: Secondary analyses of the latest Demographic and Health Surveys from Angola (2023–24), Mozambique (2022–23), Guinea (2018), and São Tomé & Príncipe (2008–09) were conducted. Data management and multilevel logistic regression were performed in STATA 17, spatial clustering and hotspots were assessed using ArcGIS (Global Moran’s I, LISA), and SaTScan detected purely spatial clusters. Multilevel binary logistic regression identified individual- and community-level determinants. Assumptions included independence within clusters, correct specification of fixed and random effects, and absence of multicollinearity. Statistical significance was set at p < 0.05, and associations were reported as adjusted odds ratios (AORs) with 95% confidence intervals (CIs).
Results: Vaccine misinformation exhibited significant spatial clustering in Angola (Moran’s I = 0.612, p < 0.001). Higher maternal education markedly reduced misinformation (secondary: AOR = 0.55, 95% CI: 0.39–0.78; higher: AOR = 0.04, 95% CI: 0.003–0.55). Attending ≥4 antenatal care visits was protective (AOR = 0.61, 95% CI: 0.47–0.77), whereas higher birth order increased the odds (2–3 children: AOR = 1.81, 95% CI: 1.30–2.51). Spatial heterogeneity was pronounced across provinces, with Cunene (AOR = 6.23, 95% CI: 1.87–20.80), Mexico (AOR = 5.20, 95% CI: 1.56–17.34), and Huambo (AOR = 3.57, 95% CI: 1.12–11.33) identified as hotspots.
Conclusion:
Vaccine misinformation is spatially clustered and shaped by a combination of individual and contextual factors. Targeted interventions in high-risk provinces, along with enhanced maternal education and antenatal care engagement, are essential to mitigate misinformation and improve equitable vaccine coverage. Leveraging spatially informed strategies can optimize resource allocation and maximize public health impact across sub-Saharan Africa.
Keywords: Vaccine Hesitancy; Spatial Epidemiology; Health Geography; Health Communication; Multilevel Modeling; DHS; Sub-Saharan Africa