Abstract
BACKGROUND: The increase in violence against health professionals in the COVID-19 pandemic makes it necessary to identify the predictors of violence, in order to prevent these events from happening. OBJECTIVE: Evaluating the prevalence and analyzing the variables involved in the occurrence of violence against health professionals during the COVID-19 pandemic in Brazil. METHOD: This is a cross-sectional study conducted online involving Brazilian health professionals during the COVID-19 pandemic. The data were collected through a structured questionnaire (Google Online Form) sent to health professionals on social networks and analyzed through logistic regression by using sociodemographic variables. The set of grouped variables was assigned to the final model when p <0.05. A network was built using the Mixed Graph Models (MGM) approach. A centrality measurement chart was constructed to determine which nodes have the greatest influence, strength and connectivity between the nodes around them. RESULTS: The predictors of violence in the adjusted regression model were the following: being a nursing technician / assistant; having been working for less than 20 years; working for over 37 hours a week; having suffered violence before the pandemic; having been contaminated with COVID-19; working in direct contact with patients infected by the virus; and having family members who have suffered violence. The network created with professionals who suffered violence demonstrated that the aggressions occurred mainly in the workplace, with an indication of psycho-verbal violence. In cases in which the aggressors were close people, aggressions were non-verbal and happened both in public and private places. The assaults practiced by strangers occurred in public places. CONCLUSIONS: Violence against health professionals occurs implicitly and explicitly, with consequences that can affect both their psychosocial well-being and the assistance given to their patients and families.