To investigate predisaster mental illness as a risk factor of poor postdisaster mental health outcomes, veterans with (n = 249) and without (n = 250) preexisting mental illness residing in the Gulf Coast during Hurricane Katrina were surveyed after Katrina and screened for posttraumatic stress disorder (PTSD), depression, generalized anxiety disorder, and panic. Logistic regression examined the association between preexisting mental disorders and positive screens after the hurricane, adjusting for demographics and exposure to hurricane-related stressors. The odds of screening positive for any new mental disorder were 6.8 times greater for those with preexisting mental illness compared with those without preexisting mental illness. Among those with preexisting PTSD, the odds of screening positive for any new mental illness were 11.9 times greater; among those with schizophrenia, 9.1 times greater; and among those with affective disorders, 4.4 times greater. Persons with preexisting mental illnesses, particularly PTSD, should be considered a high-risk group for poor outcomes after a disaster.