The September, 1985 Mexico earthquake provides potential data on actual injuries in a major earthquake of the type expected in the U.S. Such data could help modify current injury-prevention recommendations, which are currently too general. A retrospective study of victims of the Mexico earthquake will be conducted. This research, involving perspectives and methods from architecture, structural engineering, and epidemiology, will be done to identify patterns of injury and possible associations between specific types of injury and specific physical settings and/or occupant characteristics. This interdisciplinary investigation is expected to: (1) significantly increase the data base on human injury in earthquakes, and (2) serve as a basis for the formulation of building-specific planning, design, and occupant response guidelines.