Objective
To estimate the association of social determinants, the urban environment, and public policies with the occurrence and evolution over time of homicides in Santiago de Cali, Colombia (2015–2024), to inform decision-making in public health and territorial planning.
Methods
A mixed-methods approach was used. First, an ecological analysis of 2,264 neighborhood units using multivariable logistic regression evaluated the association between homicides and variables such as the multidimensional poverty index (MPI), unemployment, education, liquor-licensed establishments, and presence of state institutions between 2019 and 2024. Second, an interrupted time series (ITS) model and an autoregressive integrated moving average with exogenous variables (ARIMAX) time-series analysis were applied to analyze the effect of four municipal policies on the monthly homicide rate between 2010 and 2024.
Results
The MPI (odds ratio [OR]:1.45; p < 0.001), unemployment (OR: 1.30; p = 0.005), and density of liquor-licensed establishments (OR: 1.18; p = 0.027) were associated with a higher likelihood of homicides. Educational attainment had a protective effect (OR: 0.78; p = 0.006). A ban on male pillion riders on motorcycles reduced the homicide rate by 19% (incidence rate ratio [IRR]: 0.81; p = 0.005), whereas Casas de Justicia community justice centers led to a sustained −1.2% monthly decline (p = 0.004).
Conclusions
Homicidal violence in Cali is conditioned by structural inequalities and responds to specific municipal interventions. Policies can have immediate or sustained impacts. These findings allow for targeted resource allocation to critically affected territories, strengthening of community justice, and planning of intersectoral interventions on health, public safety, and development.