An International Spatial Analysis of the Welfare Spending’s Influence on Measles Immunization
Background: Welfare policy may reflect political-economic contexts that have important implications for peoples’ lives, including their health. Previous research has explored how welfare policies influence health, but no research has explored how welfare spending influences health interventions. Furthermore, little research on welfare policy or published in nursing journals has used spatial analysis. Purpose: The objective of this study is to develop an analytical approach to maximise exploration of welfare spending’s influence on the relationship between measles vaccination rates and measles infection rates over time and geographic location. Methods: The objective of this study is addressed with four manuscripts. A scoping review provides an analysis of literature that explores welfare spending in relation to immunizations. A theoretical model manuscript combines the Levels of Prevention model and the Ecological model for Health Promotion to outline relationships between the concepts of interest. The methodology manuscript outlines an analytical approach for statistical methods that lead to implementation of spatial regression. It includes the process of building generalized linear mixed models and implementing Bayesian analysis. Using this analytical approach, global and local Moran’s I tests indicate that spatial relationships are present among the variables of interest. Therefore, a conditional autoregressive model is also tested to account for spatial random effects. In the fourth manuscript, these results outline the findings from the model testing. Results: The final model finds that both the first dose of measles vaccine (B = -0.835, 95% Cr. I. = -0.975, -0.699), public social protection (B = -0.936, 95% Cr. I. = -1.132, -0.744), and their interaction (B = -0.239, 95% Cr. I. -0.319, -0.156) have a negative relationship with measles rates. Spatial random effects are not included in the final model because they do not improve the model fit. Significance of findings: These results suggest that national welfare spending may influence the relationship between measles infection rates and measles immunizations. Furthermore, the analytical approach manuscript makes spatial regression more accessible to health researchers. These findings, and the analytical approach used to reach them, have potential to build on the nursing and health literature while increasing the understanding of policy’s influence on health.
Bayesian modelling, immunization, measles, nursing, spatial analysis, welfare spending
Doctor of Philosophy (Ph.D.)