This article updates a 2006 review of empirical data concerning whether clinicians can predict whether patients will die by suicide, or whether fatality can be prevented. Based on further empirical data, a negative conclusion remains justified. There is good evidence that treatment programs, using psychotherapy and medication, can reduce suicide attempts. But people who die by suicide are a distinct population from attempters, and those at high risk do not necessarily present for treatment. Research on suicide prevention has not shown that fatalities among patients can be predicted, or that clinical interventions can reduce the risk. The strongest evidence for prevention derives from reducing access to means. Population-based strategies are more effective than high-risk strategies focusing on patients with suicidal ideas or attempts.