In this study, the effects of hospital, staff, and patient characteristics on the rates of use and abandonment of an outmoded medical technology, intermittent positive pressure breathing (IPPB) are analyzed. The study focuses specifically on the use of IPPB to treat inpatients with chronic obstructive pulmonary disease in a national sample of more than 500 community hospitals from 1980 to 1987. Cross-sectionally, hospitals with shorter case-mix-adjusted lengths of stay, private nonprofit or investor-owned hospitals, and hospitals located outside of the north central United States were more likely to abandon IPPB by 1980. Teaching status, location, ownership, volume, and source of payment all appeared to affect rates of IPPB use in 1980. The longitudinal analysis examines both the probability a hospital abandoned IPPB and declines in rates of IPPB use over the study period, conditioned on the availability of IPPB in 1980. The results show that changes in the characteristics of hospitals, patients, and physicians all help to explain variations in the abandonment of IPPB. These findings contrast with previous studies of technological change, which find hospital size to be the most important variable. Size is important in explaining the rate of use in 1980, but it has no effect on the rate of decline in use or abandonment after 1980. In general, the analysis demonstrates that a combination of factors, economic incentives as well as information, contribute to the abandonment of outmoded medical technologies. Given the surprisingly long time periods required for this process to occur, the analysis underscores the need to strengthen financial incentives that encourage appropriate medical decisions and to disseminate information about the efficacy of specific procedures more widely and effectively.