North American outcome studies of schizophrenia conducted within the past quarter century are reviewed if their minimum average followup is 10 years and they meet at least some modern design criteria. Ten such investigations are described and summarized. Taken as a whole, they demonstrate that schizophrenia can be a chronic disease whose outcome on the average is worse than that of other major mental illnesses. It is associated with an increased risk for suicide, physical illness, and mortality. The schizophrenic process, however, is not relentlessly progressive, as originally described, but appears to plateau after 5-10 years of manifest illness. Overall, outcome is heterogeneous, but much of the variance can be linked to sample characteristics, including expressions of psychopathology (broad vs. narrow diagnostic criteria, subtypes, and comorbidity), dimensions of chronicity (length of manifest illness, treatment resistance, age of onset, and institutionalization), and other predictor variables (gender, marital status, socioeconomic status, physical setting, and premorbid health). Long-term followup studies have yet to demonstrate clearly any effect of treatment on the natural history of schizophrenia. Finally, these studies support a broad definition of schizophrenia.