Abstract
This study aims to analyze the historical evolution of violence against women across all life stages in Brazil between 2013 and 2023. Using data from the Notifiable Diseases Information System and population projections from the Brazilian Institute of Geography and Statistics, an ecological time-series study was conducted. Trends were assessed through Joinpoint regression, while future projections utilized the Prophet machine learning model. The Random Forest model identified the most impactful variables for prediction. The study reveals alarming trends, with a projected 95% increase in violence rates over the next decade. Variables such as firearm use and known perpetrators were highlighted as significant contributors. The results underscore the urgency of targeted public policies and predictive approaches to effectively combat gender-based violence. Limitations include underreporting and data gaps, emphasizing the need for improvements in data collection.
Get full access to this article
View all access options for this article.
