Purpose: To determine how physicians monitor their patients after initial curative-intent treatment for breast carcinoma. Methods: A custom-designed survey instrument with four idealized patient vignettes (TNM stages 0 to III) was e-mailed to the 3,245 members of ASCO who had identified themselves as having breast cancer as a major focus of their practice. Respondents were asked how they use 12 specific follow-up modalities during post-treatment years 1 to 5 for each vignette. Mean, median, standard deviation, and range of the intensity of use for each modality were calculated for the four vignettes. Results: Of the 3,245 ASCO members surveyed, 1,012 (31%) responded. Of these, 915 (90%) were evaluable and were included in our analysis. Office visit, mammogram, complete blood count, and liver function tests were the most commonly recommended surveillance modalities. There was marked variation in surveillance intensity. For example, office visit was recommended 4.1 ± 2.2 times (mean ± SD) in year 1 after curative treatment of a patient with stage III breast cancer. Similar variation was observed for all modalities. Conclusions: The intensity of post-treatment surveillance performed by ASCO members caring for patients with breast cancer varies markedly despite evidence from well-designed, adequately powered randomized controlled trials. Many modalities not recommended by ASCO guidelines are used routinely, which constitutes evidence of overuse. The lack of consensus is likely due to multiple factors and constitutes an appealing target for interventions to rationalize surveillance.