Lithium treatment and suicide risk in major affective disorders: update and new findings

J Clin Psychiatry. 2003;64 Suppl 5:44-52.


Background: Evidence that therapeutic benefits of psychiatric treatments include reduction of suicide risk is remarkably limited and poorly studied. An exception is growing evidence for such suicidal risk reduction with long-term lithium maintenance. This report updates and extends analyses of lithium treatment and suicides and attempts.

Method: We pooled data from studies providing data on suicidal acts, patients at risk, and average exposure times with or without lithium maintenance therapy, and considered effects of lithium on selected subgroups.

Results: Data from 34 reported studies involved 42 groups with lithium maintenance averaging 3.36 years, and 25 groups without lithium followed for 5.88 years, representing 16,221 patients in a total experience of 64,233 person-years. Risks for all suicidal acts/100 person-years averaged 3.10 without lithium versus 0.210 during treatment (93% difference) versus approximately 0.315 for the general population. For attempts, corresponding rates were 4.65 versus 0.312 (93% difference), and for completed suicides, 0.942 versus 0.174 (82% difference). Subjects with bipolar versus various recurrent major affective disorders showed similar benefits (95% vs. 91% sparing of all suicidal acts). Risk reductions for unipolar depressive, bipolar II, and bipolar I cases ranked 100%, 82%, and 67%. Suicide risk without lithium tended to increase from 1970 to 2002, with no loss of effectiveness of lithium treatment.

Conclusion: The findings indicate major reductions of suicidal risks (attempts > suicides) with lithium maintenance therapy in unipolar >/= bipolar II >/= bipolar I disorder, to overall levels close to general population rates. These major benefits in syndromes mainly involving depression encourage evaluation of other treatments aimed at reducing mortality in the depressive and mixed phases of bipolar disorder and in unipolar major depression.

Publication types

  • Research Support, Non-U.S. Gov't
  • Review

MeSH terms

  • Bipolar Disorder / drug therapy
  • Bipolar Disorder / prevention & control
  • Bipolar Disorder / psychology
  • Depressive Disorder / prevention & control
  • Depressive Disorder / psychology
  • Humans
  • Lithium / therapeutic use*
  • Mood Disorders / drug therapy
  • Mood Disorders / prevention & control*
  • Mood Disorders / psychology
  • Risk Factors
  • Suicide / prevention & control*
  • Suicide / statistics & numerical data
  • Suicide, Attempted / prevention & control*
  • Suicide, Attempted / statistics & numerical data


  • Lithium