The medical literature was searched to find reports on the mortality of 63 medical disorders and all psychiatric disorders. Out of the approximately 4,000 mortality studies located, 227 reports on medical disorders and 249 reports on psychiatric disorders or treatment settings were abstracted. Amongst the 63 medical disorders, a significantly increased risk of suicide was seen for HIV/AIDS, malignant neoplasms as a group, head and neck cancers, multiple sclerosis, peptic ulcer, renal disease, traumatic spinal cord injury and systemic lupus erythematosus. Most of these disorders are associated with mood disorder, substance abuse or both. Inconclusive evidence for increased risk was observed for limb amputation, heart valve replacement and surgery, disorders of the intestine (Crohn's disease, ileostomy, ulcerative colitis); hormone replacement therapy, alcoholic liver disease, neurobifromatosis, systemic sclerosis and Parkinson's disease. Pregnancy and the first post-natal year had a significantly decreased risk. There was no evidence of either increased or decreased risk for any of the other medical disorders studied. Of the 44 psychiatric disorders for which mortality studies were located, 36 had a significantly raised risk of suicide, five had a raised risk which failed to reach statistical significance, one had neither a raised or reduced risk and for two, no suicides were recorded. Each of the eight different psychiatric treatment settings had a significantly increased risk of suicide. The results put into context the relatively small risk of suicide associated with a few medical disorders, compared with the much higher risk of suicide associated with almost all mental disorders. These findings strengthen the evidence that mental disorders are the most important disease element in the causation of suicide. Effective treatment of the underlying mental disorder could theoretically prevent many premature deaths by suicide.