The aim of this paper was to identify the spatial pattern of IMR and to analyze the local and neighboring effects on such geographic differences before and after the introduction of decentralized health management in Colombia.
Between 1993 and 2005, we analyzed the spatial dependence and spatial heterogeneity of IMR by using the Moran’s I and LISA tests. Then, we related municipal determinants to IMR by using the spatial Durbin model to indentify which variables have locally contributed to spatial pattern of IMR between those years. We analyzed the effect of neighboring factors that contributed to the geographic concentration of this outcome across municipalities.
The results suggest that in 1993 there was a spatial dependency of IMR across municipalities, which had deepened in 2005, leading two clustering patterns of IMR. Centre municipalities exhibited a low IMR and peripheral municipalities were characterized by high rates. This map of IMR coincided with the spatial distribution of poverty. The spatial clustering was explained by the access of housing basic services and socioeconomic attributes in municipalities, but also they were largely explained by neighborhoods factors.
Geographic differences of IMR have not been addressed by the decentralized management. The socioeconomic situation of the neighboring municipalities should be taken into account in the design of decentralized policies; otherwise the decentralization policies would be more effective in reaching better-off municipalities than worst-off.
Learning Objectives: 1. Evaluate the local conditions to undertake decentralised health reforms, in a developing country context 2. Analyse the potential results of decentralisation reforms and health outcomes 3. Discuss the equity implications on the relationship between decentralisation and local institutional capacities