Background: The goal of this study was to investigate the stability and correlates of adolescent suicidal risk across adolescence. Methods: Suicidal risk was studied longitudinally in N = 593 subjects in 1994, 1997, and 2001 at mean ages of 13, 16, and 20 years. Three partly overlapping suicidal risk groups were compared to three control groups matched by age and gender regarding behavioural and emotional problems, number and impact of life events, coping capacities, self-related cognitions, and the size and efficiency of the social network. Psychiatric diagnoses were obtained in young adulthood. Results: The group stability of suicidal risk was 43% from time 1 to time 2 and 18% from time 1 to time 3. Intra-individual stability of suicidal risks across adolescence was significantly higher than expected. At all three times of assessment, there was a clear excess of abnormal psychosocial findings in the suicidal risk groups as compared to the matched controls. Substance abuse disorders, depressive disorders, phobias and any psychiatric disorders in young adulthood were predicted by suicidal risk in preadolescence irrespective of behavioural and emotional disorders. Conclusions: Suicidal risk individuals are remarkably stable across adolescence and show a very consistent pattern of associated behavioural and emotional problems and psychosocial variables across time from preadolescence to young adulthood. They also display an increased risk for some mental disorders in young adulthood.