Mortality in recurrent mood disorders during periods on and off lithium. A complete population study in 362 patients.

Recent studies have suggested that long-term lithium treatment reduces the high mortality rates of recurrent mood disorders in patients selected for and compliant with treatment at specialized lithium clinics. Whether lithium also generally reduces mortality in this diagnostic category under less select treatment conditions is a question of vital public health interest. The impact of prophylactic lithium on mortality was studied in a complete population of 362 unselected patients with DSM-III-R diagnoses of mood disorders or schizoaffective disorder, hospitalized at least once between 1970 and 1977 and treated with lithium for a minimum of one year. The patients were followed until 1991 or until date of death. The final analyses included 3911 patient years with lithium and, because of temporary or permanent discontinuations, 1274 patient years without lithium prophylaxis. A total of 129 deaths were recorded, compared with the 60.7 deaths that would normally be expected in the general population, yielding a Standard Mortality Ratio (SMR) of 2.1, significantly different from 1.0 (p < 0.001, 95% confidence limits 1.8-2.5). The relative risk of death was 1.7 times higher (p < 0.01, 95% confidence limits 1.2-2.6) during periods off lithium than during periods on lithium. The relative risk of suicide was 4.8 times higher off lithium than on lithium (p < 0.02, 95% confidence limits 1.1-12.6). Suicide, pneumonia, pyelonephritis, and, unexpectedly, pulmonary embolism contributed to the excess mortality both on and off lithium.(ABSTRACT TRUNCATED AT 250 WORDS)