OBJECTIVE: To investigate the prevalence and predictors of changes in diet, physical activity, and dietary supplement use among cancer patients. Design/subjects Telephone interviews of a population-based sample of 126 breast, 114 prostate, and 116 colorectal cancer patients from the state of Washington. ANALYSIS: Logistic regression was used to estimate the odds ratio as a measure of the association of participant characteristics with lifestyle changes in the 12 months before the interview. RESULTS: Overall, 66.3% of patients reported making lifestyle changes: 40.4% made one or more dietary changes, 20.8% added new physical activity, and 48.0% started taking new dietary supplements. Compared with men, women were 2.2 times more likely to take new dietary supplements (P <.01). Compared with patients aged 35 to 59, those aged 60 to 69 and 70 or older were statistically significantly less likely to make dietary changes (odds ratio = 0.39 and 0.54, respectively) or to take new supplements (odds ratio = 0.42 and 0.69, respectively). Compared with patients who received only one medical treatment, those receiving three or more treatments were more likely to make dietary changes (odds ratio = 2.6) or to start new physical activity (odds ratio = 3.0). Patients diagnosed 12 to 24 months before the interview were as likely to report making lifestyle changes as those diagnosed within one year of the interview. Having a stronger desire for personal control or internal locus of control predicted use of new dietary supplements (P for trend <.05 for both). Applications/conclusions Cancer survivors are likely to be making lifestyle changes and represent a group that could benefit from counseling on diet and physical activity.