OBJECTIVES: To measure and compare preferences for attributes of colorectal cancer (CRC) screening tests using a stated preference survey of the general population and physicians in Canada and the United States (US). METHODS: A stated preference survey was administered online with 11 choice tasks between two hypothetical CRC screening tests. Each test was described by nine attributes: process, pain, preparation, frequency, follow-up, complication risk, sensitivity, specificity, and out-of-pocket cost. Each scenario included a follow-up opt-out question to choose no screening. 1087 US and 501 Canadian respondents participated and 100 physicians responded in both countries. Physicians were asked to indicate their patients' preferences. Responses were modeled using bivariate regression with main effects and interactions with the optout term. Willingness-to-pay was calculated for common CRC screening tests. RESULTS: Physicians expected patients to choose the option of 'no screening' more frequently than patients themselves (55% vs 29% respectively, p<0.001). For all groups the most important attribute was sensitivity, but physicians' perception of patients' preferences were significantly different from actual patient preferences. Other key attributes were those related to test performance or the testing process. Fecal DNA, colonoscopy, and virtual colonoscopy were the most preferred tests by all groups, but respondents were willing-to-pay more than physicians predicted. CONCLUSION: Physicians' perception of patients' preferences are significantly different from those of the general population, although both preferred tests with high sensitivity. The significant difference in the frequency of choosing no screening between physicians and their patients may have serious implications for CRC screening uptake since physicians generally exert a strong influence on decisions about health care treatment, and especially screening programs. Among general population and physicians, Canadian and US respondents' preferences were similar.