Background & Aims: Patients with Barrett’s esophagus (BE)- have a risk of esophageal adenocarcinoma of approximately 0.5% per year. Patients may have difficulty understanding this risk. This study assessed the perceived risk of cancer in patients with BE, and correlated their risk estimates with their health care use behaviors. Methods: We performed a survey of patients with BE participating in an endoscopic surveillance program at 2 sites: a university teaching hospital and a Veterans’ Administration hospital. A questionnaire also elicited their demographics as well as their sources of health information. Health care behaviors, including physician visits and endoscopic surveillance behaviors, were assessed. Patients were classified as either overestimators or nonoverestimators of risk. Characteristics of overestimators, as well as health care use patterns, were assessed. Results: One hundred eighteen patients met entry criteria, and 92 (78% ) completed all the questionnaires. Sixty-eight percent of patients overestimated their 1-year risk of cancer, with a mean estimated 1-year cancer risk being 13.6% . The lifetime risk also was overestimated by 38% of patients. Patients who overestimated risk were more likely to be Veterans’ Administration medical center patients, have more symptomatic reflux, and were more likely to use the Internet to get health care information. There was no significant difference in physician visits between overestimators and nonestimators (1.2 visits per year vs 1.0, P = .20), nor in endoscopy use (5.7 endoscopies per 5-year period vs 5.0, P = .42). Conclusions: The majority of patients with prevalent BE participating in an endoscopic surveillance program overestimated their chances of developing adenocarcinoma of the esophagus. Efforts to improve education of such patients with BE are warranted. Background & Aims: Patients with Barrett’s esophagus (BE) - have a risk of esophageal adenocarcinoma of approximately 0.5% per year. Patients may have difficulty knowing this risk. This study assessed the perceived risk of cancer in patients with BE, and correlated their risk estimates with their health care use behaviors. Methods: We performed a survey of patients with BE participating in an endoscopic surveillance program at 2 sites: a university teaching hospital and a Veterans’ Administration hospital. A questionnaire also elicited their demographics as well as their sources of health information. Health care behaviors, including physician visits and endoscopic surveillance behaviors, were assessed. Patients were classified as either overestimators or nonoverestimators of risk. Characteristics of overestimators, as well as health care use patterns, were assessed. Results: One hundred eighteen patients met entry criteria, and 92 (78%) completed all the quesires. Sixty-ei The lifetime risk also was overestimated by 38% of patients. Patients who overestimated risk were more likely to be Veterans’ Administration medical center patients, have more likely to use the Internet to get health care information. There was no significant difference in physician visits between overestimators and nonestimators (1.2 visits per year vs 1.0, P = .20), nor in endoscopy use (5.7 endoscopies per 5-year period vs 5.0, P = .42). Conclusions: The majority of patients with prevalent BE participating in an endoscopic surveillance program overestimated their chances of developing adenocarcinoma of the esophagus. Efforts to improve education of such patients with BE are warranted.