Determinants and Inequalities of Interpersonal Violence: A 10-Year Analysis of Emergency Department Visits Before and During COVID-19

Abstract

Introduction

Violence is a complex phenomenon with significant physical and psychological consequences and multifactorial determinants, requiring a multidisciplinary approach. Epidemiological data is often fragmented and underestimated. According to the World Health Organization, there are around 475,000 homicides and hundreds of millions of non-fatal violent incidents occur each year, resulting in substantial long-term consequences.

Objective

This study aimed to quantify violence events in Liguria and to describe their relationship with sociodemographic characteristics of people involved. Routinely collected healthcare data can inform surveillance and prevention strategies. Temporal trends were analyzed with a focus on the effect of the pandemic on the phenomenon.

Methods

A retrospective study was conducted, collecting data on all emergency department (ED) visits for violence and self-harm recorded in a northern Italian region between January 2016 and September 2025 from the regional healthcare data warehouse. Descriptive analyses were performed to characterize temporal and sociodemographic patterns. An Interrupted Time Series based on a Poisson model was used to assess changes associated with the lockdown measures implemented during the pandemic.

Results

During the study period, 5,541,712 ED visits were recorded, of which 54,321 (0.98%) were related to violence. Rates varied across provinces. Foreign nationals accounted for approximately one-third of visits, with an increasing trend over time, suggesting possible inequalities in risk exposure. Visits were more frequent among males, during evening and nighttime hours, on weekends, and in the summer months, consistent with contexts of increased social interaction.

The ITSA model demonstrated no significant pre-intervention trends both for Italian (IRR 0.999; p = 0.605) and foreign residents (IRR 0.997; p = 0.191), indicating stable incidence prior to the lockdown. In a similar way, there was no evidence of an immediate level change at the onset of COVID pandemic restrictions, while a marked and statistically significant reduction in incidence was observed during the lockdown period in both groups, with an approximate 59% and 57% decrease in admissions respectively for Italians and foreign residents. The post-COVID trend highlighted differences among the two population groups, as Italians’ trajectory of violence-related ED admissions did not vary (IRR 1.002; p = 0.488), while a significant change in the post-intervention trend was observed for foreign residents (IRR 1.007; p = 0.027), indicating a modest but statistically significant increase in incidence over time following the lockdown in this group.

Conclusions

This study provides a reliable, population-based picture of violence-related ED visits in a Northern Italian region over the last decade. It identifies population groups at greater risk -particularly young adult males and foreign nationals- as well as temporal and geographical patterns, suggesting clear targets for preventive and supportive interventions. The differential impact of the pandemic on interpersonal violence, and the diverging post-lockdown trends between Italian and foreign residents, highlight the role of social contexts and structural inequalities in shaping violence-related outcomes. These findings underscore the epidemiological relevance of violence as a public health issue and call for evidence-based policies, targeted prevention strategies, and the strengthening of surveillance systems within healthcare and social services.