[Underreporting of non-fatal work-related injuries in Brazil].

OBJECTIVE The implementation of work-related injury prevention policies has been hindered by underreporting of incidents among formal workers, and substantial underreporting among informal workforce. This study aimed at estimating the underreporting of work-related injury in a median-sized city. METHODS A random survey was carried out among residences in the urban area of Brazil. Residents were interviewed about the occurrence of work-related injury to people aged more than nine years in the last 90 days. All incidents reported were double checked in the National Social Security Institute (INSS) records. RESULTS There were 9,626 residences visited. It was estimated 79.5% (CI 95%: 78.8%-80.3%) of underreporting of work-related injury. CONCLUSIONS Work-related injury reporting is poor in the study location and this may be occurring in other cities. Data suggest the need to build up information systems on Brazilian workers' health. It should incorporate methods, materials and human recourses necessary to recognize, store, analyze, and spread information to support injury prevention policies and promote workers' health programs.