Abstract: As health care becomes increasingly personalized to the needs and values of individual patients, informational interventions that aim to inform and debias consumer decision-making are likely to become important tools. In a randomized controlled experiment, we explore the effects of providing participants with published fact boxes on the benefits and harms of common cancer screening procedures. Female participants were surveyed about breast cancer screening by mammography, while male participants were surveyed about prostate cancer screening by prostate-specific antigen (PSA) testing. For these screening procedures, we expect consumers to have overly optimistic prior beliefs about the benefits and harms. We find that participants update their beliefs only modestly and change their stated preferences to seek screening even more modestly. Participants who scored higher on a numeracy test updated their beliefs and preferences about screening more in response to the fact boxes than did patients who scored lower on the numeracy test. More-numerate subjects also seem to become more anxious in response to the risk information.