(Un)Broken: Lateral violence among hospital nurses, user violence, burnout, and general health: A structural equation modeling analysis.

Introduction: Workplace violence is a social problem yet to be solved. Although it is present in virtually all work environments, its prevalence in healthcare settings stands out, being perceived as something inherent to the job. Most studies in this context have focused on user violence against professionals. However, it has been observed that violence among colleagues in these types of jobs is a risk factor for the health of workers and has rarely been studied as a whole. Among the main consequences of exposure to violence reported in the literature, burnout syndrome, depression, anxiety, or somatic problems have been among the most studied. On the one hand, some authors claim that being exposed to workplace violence can increase the associated physical and psychological pathology and lead to a picture congruent with burnout. On the other hand, it has been hypothesized that violence is associated with burnout, which can trigger physical and psychological symptoms. Taking into account this background, the aim of this study is to explore workplace violence in health personnel, symptomatology, and burnout syndrome through mediation models that allow us to know the interrelationships between the variables. Methods: A cross-sectional design with a double descriptive-associative strategy was used. The sample was composed of 950 nursing professionals from public hospitals. The scales of physical and non-physical violence from users to professionals HABS-U, personal, social, and occupational violence among co-workers using the Health Aggressive Behavior Scale - Co-workers and Superiors (HABS-CS) scale, the burnout scale Maslach Burnout Inventory - General Survey (MBI-GS) which evaluates professional exhaustion, efficacy and cynicism, and the factors referring to depression, anxiety, somatization, and dysfunction of the GHQ-28 scale were applied. In order to calculate the models, workplace violence was used as a predictor of symptomatology, using the burnout variables as mediators. Regression coefficients with and without mediation model, direct and standardized estimates were obtained. For statistical power, Bootstrap analysis was used to calculate direct mediation effects. Results: After controlling the mediation effects of burnout and cynicism, physical and non-physical user violence toward healthcare personnel were significant predictors of the GHQ-28 scores. These same results were obtained when assessing the relationship between social, occupational, and personal violence among co-workers and GHQ-28 scores. Conclusion: Our results contribute to increase the evidence about the effects of violence on the health of professionals and to advance in the characterization of the possible consequent psychological damage. Regardless of the type of violence experienced, exposure to violence can lead to anxious, depressive or somatization symptoms, among others. Violence is also a predictor of burnout syndrome, which in turn accentuates the rest of the consequences studied. Despite the limitations of the proposed model, these results serve to highlight the complexity of the situation experienced by healthcare professionals. Moreover, it serves as a basis for proposing intervention/prevention programs to raise awareness and protect professionals from these risks. To this end, self-care tools should be proposed with which professionals take care of their own health through the management of violent situations and/or the improvement of occupational health.