BACKGROUND, The observed increase in sojourn time for preclinical breast carcinoma raises the question of whether women age greater than or equal to 65 years can be screened less frequently than younger women. METHODS. A cost-utility analysis using a computer model that simulates the demography, epidemiology, and natural history of breast carcinoma to estimate expected life-years gained, extra incidence, extra life-years with disease, and costs incurred by different breast carcinoma screening programs in the general population was conducted. RESULTS. The estimated ratio of favorable/unfavorable effects was lower for longer screening intervals compared with shorter screening intervals. The cost-effectiveness ratio was much less favorable in shorter screening intervals. CONCLUSIONS, The results of the current analysis showed that although a longer sojourn time for preclinical breast carcinoma should not necessarily be accompanied by a longer screening interval, a shorter screening interval was not very efficient. (C) 1999 American Cancer Society.