BMJ  2004;328 (14 February), doi:10.1136/bmj.328.7436.0-e

Filler

POEM^*

Risk of suicide in bipolar disorder is least with lithium

Question Which drug used to treat bipolar disorder is most effective for reducing the risk of suicide?

Synopsis The use of lithium in the treatment of bipolar disorder has decreased as the use of anticonvulsants has steadily increased. Consistent evidence shows that lithium is effective for reducing the risk of suicide, but little is known about other agents. In this retrospective cohort study, the authors wanted to compare the risk of a suicide attempt and death during lithium treatment with that during treatment with divalproex (Epival, Depakote) and carbamazepine (Tegretol). Data were obtained from two large managed care organisations in California on 20 638 members aged 14 years or older with at least one outpatient diagnosis of bipolar disorder and at least one filled prescription for lithium, divalproex, or carbamazepine. The mean follow up period was approximately three years per individual. Suicide attempts were identified by using emergency department visit and hospital discharge diagnoses. Suicide deaths were identified from health plan mortality records and death certificate reports. Because of the potential for confounding bias in analyses of large databases like this, the authors adjusted for age, sex, health plan, year of diagnosis, comorbid medical and psychiatric conditions, and concomitant psychotropic drug use. However, because of the retrospective study design, we can never be certain that confounding bias did not occur. The risk of suicide was 2.7 times higher (95% confidence interval 1.1 to 6.3) during treatment with divalproex than with lithium. Rates for non-fatal attempts were also higher during treatment with divalproex. Although the power of the analysis to evaluate carbamazepine was low, patients taking carbamazepine were more likely to be hospitalised for suicide attempts.

Bottom line The risk of suicide attempts and death in patients with bipolar disorder seems to be lower during treatment with lithium than during treatment with divalproex and carbamazepine. More reliable evidence is needed from prospective randomised trials that compare these drugs head to head and with others.

Level of evidence 2b (see www.cebm.net/levels_of_evidence.asp). Individual cohort study or low quality randomised controlled trials (< 80% follow up).

---

Goodwin FK, Fireman B, Simon GE, et al. Suicide risk in bipolar disorder during treatment with lithium and divalproex. JAMA 2003;290: 1467-73[Abstract/Free Full Text].

©infoPOEMs 1992-2003 www.infoPOEMs.com/informationmastery.cfm

^* Patient-Oriented Evidence that Matters. See editorial ( BMJ 2002;325: 983[Free Full Text])

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

Rapid Responses:

Read all Rapid Responses

A Deficiency of Lithium ?

Dr. Herbert H. Nehrlich
bmj.com, 13 Feb 2004 [Full text]