Multilevel Predictors of Vulnerability and Resilience to Psychological Distress after Violent Victimization: Why Context Matters

Background/Purpose: Mexico recorded 35,588 homicides in 2019. Extortions, robbery, and other crimes also remain elevated. These potentially traumatic events underscore the daily threats and traumatic stress that many individuals in Mexico confront. They also suggest an increased risk of exposure to traumatic experiences and their psychological impacts. There is ample evidence that exposure to violence is associated with adverse mental and physical health outcomes and that effects could vary by gender and socio-economic status. Less understood, however, are the multilevel factors that could explain differences in psychological response to violent crime in addition to gender and SES, particularly within high crime contexts. The purpose of this study is to examine individual, social network, and neighborhood characteristics in predicting vulnerability and resilience of individuals directly exposed to crime and to compare these effects between those who experienced psychological distress post-victimization and those that did not.

Methods: We analyzed a subsample (n=46,345) from the 2019 National Survey on Victimization and Perception of Public Security in Mexico. We conducted descriptive analyses, ANOVA, and chi-square test to examine socio-demographic differences between groups. Hierarchical logistic regression (LR) was used to test 3 nested models and estimate the probability of experiencing psychological distress post-victimization using demographic, social, and neighborhood characteristics as predictors (e.g. gender; age; employment status; social network trust; neighborhood crime; community street lights; government-sponsored services, etc.). Bayesian Information Criterion was used to assess model fit.

Results: The sample is gender-balanced (51% males and almost 49% females) with a mean age of 39. The majority of the sample is employed but males have higher employment rates (84%) than females (55%). Results revealed a strong sense of social network trust but females scored lower (mean=11.4) than males (mean=12.3). Consistent with prior studies, LR results showed gender as a significant predictor of psychological distress, where females are two times more likely to experience distress after victimization (p<.001). Chronic stressors such as living in neighborhoods with frequent crimes, poor street lighting, and high juvenile delinquency rates increased the odds of psychological distress, even after accounting for the effects of poverty levels and the other variables in the model. Age, neighborhood gangs, private security, and government-sponsored services in the locality (e.g. local patrols; anti-narcotrafficking and anti-corruption programs) were not significant, however, the presence of unemployment programs (Odds Ratio (OR)=.70) (p<.05) and stronger trust in one’s social network (OR=.88) decreased the odds of psychological distress post-victimization (p<.001).

Conclusion/Implications: These results provide evidence of the potentially increased risk and vulnerability of some individuals for psychological distress after experiencing victimization, particularly those that confront chronic social and environmental stressors. However, results also show that improving the quality of one’s physical environment and social relationships have the potential for mitigating the negative impact of violence and foster resilience. These findings have implications for intervention research that incorporates multi-level contextual factors to reduce risk and promote resilience. We will discuss how these findings inform our development of a longitudinal study in Mexico City that examines multi-level predictors of resilience in violent contexts.