We tend to cringe when we hear "So, how much time do I have, doc?" Yet prognostic discussions are a core skill of being a compassionate physician, preparing patients and families to live with serious illnesses, and enabling informed medical and personal decisions. The first challenge of prognostic communication is the inherent and unavoidable uncertainty, that the exact trajectory of health (or illness) is unknowable. Thus, clinicians are inaccurate prognosticators, overestimating by up to a factor of 5. 1 We have difficulty making accurate short-term time-based prognostic estimates for common diseases, such as congestive heart failure, and can be off by as much as 1 to 2 years. 2 Compounding this uncertainty, we worry about upsetting our patients with too much or unwanted information. On one hand, patients say that they want to know their prognosis and report that it is one of their highest priorities. 3 On the other hand, patients also say that they are unsure about how much they want to know and give mixed messages about how much they want to talk about it. 4 As one patient said, "I ask the question, and then I don't want to know the answer. But the question is out there, and then I am devastated." Uncertain about the information and not wanting to cause emotional harm, we hesitate to talk with patients about their futures. We watch colleagues hesitate with statements such as "Well, I don't have a crystal ball" or "You know I can't tell you that," or "Only God knows," or "We are very bad at predicting this sort of thing." But we also worry that our patients do not have the prognostic information that they need, however imprecise. And perhaps worse, we worry that patients perceive that we cannot handle the tough discussions they need from us.We propose that to succeed in prognostic communication, what clinicians most need is not precision about the time ahead or the fortitude to discuss it, but rather a softened approach. To do so, we recommend communicating a prognosis by pairing our hopes and worries: 5 "I am hoping that you have a long time to live with your heart disease and I am also worried that the time may be short, as short as a few years." We have found this approach to be popular with colleagues-they like it and incorporate it into practice quickly.It works well for several reasons. First, by expressing hopes and worries, clinicians incorporate "I" statements, such as "I am hoping" or "I am worried." These statements share the feelings, beliefs, or values of the clinician rather than an objective prediction of the future. Originating in Carl Roger's nondirective approach to therapy and the parenting literature of the 1960s, 6 "I" statements acknowledge that the viewpoint expressed is personal. They contrast with "you" statements, which focus on the person being spoken to: "You