Method: A documentary study that used the records of the incidents published in the Reports of Adverse Events (AE) in Brazil. The following variables were selected: number of incidents by type, type of health service, hospital unit, and degree of harm. To find whether there was a significant difference across the Brazilian regions by notifications related to general incidents, AE, and deaths, the analysis of variance and the Tukey tests were used.Results: A total of 109,082 incidents were reported, of which 75,088 were AE, with 649 deaths. In relation to the types of incidents reported, there was a higher frequency in the categories other (30.04%) and failures during health care (26.72%). A total of 93.90% of the incidents occurred in hospitals, with 54,950 cases registered in hospitalization units and 30,141 cases in intensive care units. Statistically significant differences across the Brazilian regions were observed in the number of incidents (P = 0.004), AE (P = 0.004), and deaths (P = 0.024).Conclusions: A significant underreporting of incidents was found in Brazil, demonstrating only the tip of a giant iceberg. More than half of the incidents were reported as AE and were registered in hospitals, reiterating the importance of establishing public health policies at national, state, and municipal levels, with adequate supervision of the health service regarding the implementation of the Patient Safety Nuclei and the preparation of new protocols based on the most prevalent incidents.