Figure 3. Graph illustrates that the number of CT studies performed for pulmonary embolism per inpatient increased significantly between 1992 and 2001 (P = .006). Figures 1-3 demonstrate the timing of changes that occur when a new technology replaces an old one; in this case, a downturn in the use of pulmonary angiography and ventilation-perfusion scintigraphy almost exactly coincides with a steep increase in CT pulmonary angiography usage. (Fig 1 modified and Figs 1-3 reprinted, with permission, from reference 12.)