Community-Level Social Norms and Childhood Immunization: A Multilevel Contextual Analysis

Abstract Background Immunization coverage of children is still low in many low and middle income countries including Ethiopia. The impact of community level social norms, shared beliefs, attitudes and behaviours about vaccination within a community remain poorly understood, while some individual level factors such as maternal education and wealth have been extensively studied. Understanding norms in the community that affect immunization behaviour is crucial for designing effective context-specific interventions. The aim of the study is to investigate the relationship between community level social norms and immunization status of children age 12–23 months in Ethiopia, thru multi-level contextual analysis framework. Methods We used the data from EDHS 2024–2025, a nationally representative cross sectional survey of children aged 12–23 months. Community-level social norms were measured using a cluster level aggregation of the individual-level indicators, which include: (1) community-level maternal education (proportion of mothers with secondary education or higher), (2) community-level female media exposure (proportion of women with regular media exposure), (3) community-level antenatal care utilization (proportion of women with ANC ≥ 4 visits), (4) community-level institutional delivery (proportion of women with institutional delivery), and (5) community-level women's decision-making autonomy (proportion of women who participate in household decision making). A multilevel mixed-effects logistic regression was used to assess the relationship between the social norms at the community level and child immunization, while controlling for individual-level and household-level covariates. Results The intra-class correlation coefficient (ICC) was 0.28, meaning that 28% of the variability in immunization status was due to variation between communities. In the model adjusted for all variables, community level maternal education had significant association with full immunization (AOR = 1.62; 95% CI: 1.23–2.15). Full immunization was also significantly associated with the utilization of ANC at the community level (AOR = 1.83; 95% CI: 1.39–2.41). The odds of full immunization were higher among women who delivered in a community (AOR = 1.45; 95% CI: 1.12–1.88). The community level media exposure was significantly associated with the full immunization status (AOR = 1.38, 95% CI: 1.08–1.76). The community level of women's autonomy in decision making was positively linked to full immunization (AOR = 1.29; 95% CI: 1.02–1.63). Other factors at the individual level such as maternal education, wealth, and ANC attendance were also independently associated with immunization status. Conclusion In Ethiopia, aggregated individual behaviors and attitudes related to social norms at the community level were related to childhood immunization status. The results indicate that immunization interventions should include barriers at the individual level and the community level based on the normative factors. Community-based strategies that utilize positive social norms, involve community leaders and address community-level misconceptions might help to increase childhood immunization rates.