Background: The Norwegian Labour Inspection Authority records fatal occupational injuries in mainland bases activities, i.e. all sectors except offshore, aviation, shipping, hunting and fishing; the Registry's information on these injuries has been considered complete. The present study aimed at testing this assumption.
Material and methods: In 2000 - 03, the Labour Inspection Authority recorded 183 fatal occupational injuries; 171 of the deceased were residents in Norway. Each of these deaths were compared with fatal occupational injuries in the Norwegian Cause of Death Registry. A capture-recapture model was used to estimate the real number of fatal occupational injuries.
Results: In 2000 - 03, 214 fatal occupational injuries were recorded among residents in Norway employed in all sectors (except offshore, aviation, shipping, hunting and fishing) in at least one of the two registries (98 cases were reported in both registries). The Norwegian Labour Inspection Authority mainly lacked information about (in comparison with the Death Registry) deaths in the military (1 of 9), in the health and social services (3 of 7), road traffic accidents (36 of 52) and deaths in Northern Norway/Spitsbergen (17 of 28). One third of all recorded cases (77 of 214) were caused by transport accidents. Risk by industry (deaths per 100 million work hours) was highest for primary industries (7.0) and for <<transport and communication>> (4.1). The real number of fatal injuries was estimated to 246, or 44 % more than the 171 deaths registered by the Labour Inspection Authority.
Interpretation: Fatal occupational injuries are much more frequent than reported in the official registries. Underreporting, particularly of road traffic accidents/transport accidents, may lead to misinterpretation of risks and time trends and of need for preventive action.