Unrecognized intracranial injury can produce brain damage, severe disability and even death. Reports of """"""""occult"""""""" injuries have generated concern, and physicians, fearful of missing such injuries, have adopted the practice of obtaining computed head tomography on virtually all blunt head injury victims. This expensive practice exposes many patients to radiation, while detecting injuries in a minority. Existing data suggest that a limited number of clinical criteria can reliably identify blunt head injury victims who have """"""""no risk"""""""" of intracranial injury, and hence no need for tomography, without misidentifying any injured patient. It is estimated that nation-wide implementation of such criteria could resultant in an annual reduction in radiographic charges of 130 million dollars, and a significant decrease in radiation exposure and radiation induced malignancies. This proposal seeks to reduce computed head tomography by developing reliable """"""""no risk"""""""" criteria. This goal will be accomplished by confirming the following two hypotheses: I- A limited number of criteria can reliably identify blunt head injury victims who have no risk of intracranial injury. II- Implementing these criteria will reduce the number of computed head tomographs without missing any significant intracranial injuries. These hypotheses will be tested in a prospective, two-part, multicenter study. The first phase of the study will examine the interrater variability of candidate criteria to identify those individual criteria that can be reliably measured in the clinical setting. The second phase of the study will document the presence or absence of each reliable criterion, as well as the presence or absence of intracranial injury in a cohort of head injury victims. Using recursive partitioning, all combinations of the reliable criteria will then be examined to identify a """"""""no risk"""""""" subset that predicts intracranial injuries with 100 percent sensitivity, excludes injury with 100 percent negative predictive value, and retains the highest specificity. Developing criteria that have high statistical confidence requires evaluations on 737 intracranial injury victims. This in turn necessitates evaluations on approximately 15,000 blunt head injury patients. Potential reductions in computed tomography will be determined by counting the number of head tomographs performed on """"""""no risk"""""""" patients. Reduction in charges and radiation exposure will be determined by respectively summing radiographic charges and life- time decreases in radiation morbidity and mortality for all """"""""no risk"""""""" cases.
