Related Experiment Video
Updated: Jan 13, 2026

Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
Spatial zero-inflated poisson analysis of homicide mortality in Mexican municipalities, 2020
Emerson Augusto Baptista1, Claudio Alberto Dávila Cervantes2
1El Colegio de México, Center for Demographic, Urban and Environmental Studies, Mexico City, 14110, Mexico.
Objectives:
Homicide remains one of the most pressing public health challenges in Mexico, with mortality rates rising dramatically since 2006 in the context of escalating organized crime and militarized security strategies. Understanding its spatial distribution and structural determinants is essential for designing prevention strategies that address inequality, urbanization, and marginalization.
Study Design:
We employed a Spatial Zero-Inflated Poisson (S-ZIP) model to analyze homicide mortality across Mexico's municipalities.
Methods:
We analyzed homicide mortality across 2469 Mexican municipalities in 2020 using data from the National Institute of Statistics and Geography (ICD-10 codes X85-Y09, Y87.1) and census-derived socioeconomic indicators. To account for excess zeros, overdispersion, and spatial dependence, we applied a Spatial Zero-Inflated Poisson (S-ZIP) model. The explanatory variables included the municipal marginalization index, the proportion of the urban population, and the Gini coefficient of income inequality. Spatial dependence was modeled through Besag-York-Mollié random effects.
Results:
Results from the Spatial Zero-Inflated Poisson model showed significant associations between the explanatory variables and homicide mortality, with urbanization and income inequality increasing the risk and reducing the probability of zero homicides. While all covariates were statistically significant, the effect of marginalization partly reflected unmodeled spatial patterns. The model further estimated that nearly one-quarter of Mexican municipalities had a structural zero probability of homicides. Overall, the findings highlight substantial territorial heterogeneity in homicide mortality across the country.
Conclusions:
Homicide mortality in Mexico is shaped by structural inequalities and spatial clustering. By framing lethal violence as a public health issue, our findings support geographically targeted policies that reduce inequality, address risks linked to urbanization and strengthen institutional and community resilience.
Related Concept Videos
Statistical Methods for Analyzing Epidemiological Data
Parametric Survival Analysis: Weibull and Exponential Methods
Weibull Distribution
The Weibull distribution is a flexible model used in parametric survival analysis. It can handle both increasing and decreasing hazard rates, depending on its shape parameter...
Bias in Epidemiological Studies
Causality in Epidemiology
Poisson Probability Distribution
The...
Hypothesis Test for Test of Independence
H0: The two variables (factors)...

