Citation

Marangell LB, Bauer MS, Dennehy EB, Wisniewski SR, Allen MH, Miklowitz DJ, Oquendo MA, Frank E, Perlis RH, Martinez JM, Fagiolini A, Otto MW, Chessick CA, Zboyan HA, Miyahara S, Sachs G, Thase ME. Bipolar Disord. 2006; 8(5 Pt 2): 566-575.

Affiliation

Mood Disorders Center, Menninger Department of Psychiatry, Baylor College of Medicine and the Department of Veterans Affairs, VISN 16 MIRECC, Houston, TX.

Copyright

(Copyright © 2006, John Wiley and Sons)

DOI

10.1111/j.1399-5618.2006.00369.x

PMID

17042830

Abstract

Objectives: Bipolar disorders are associated with high rates of suicide attempts (SAs) and completions. Several factors have been reported to be associated with suicide in persons with bipolar disorder, but most studies to date have been retrospective and have not utilized multivariate statistics to account for the redundant prediction among variables submitted for analysis. Methods: This study examined the association between baseline clinical and demographic variables and subsequent SAs and completions through 2 years of follow-up of participants in the Systematic Treatment Enhancement Program for Bipolar Disorder using a pattern-mixture model. Results: Of the sample with complete data (n = 1,556), 57 patients (3.66%) experienced an SA or completion (CS). Several variables predicted suicidality (SA + CS) in this data set when considered alone, but after controlling for redundant prediction from other baseline characteristics, only history of suicide [odds ratio (OR) = 4.52, p < 0.0001] and percent days depressed in the past year (OR = 1.16, p = 0.036) were significantly associated with SAs and completions. A secondary analysis included a greater number of variables but a smaller sample size (n = 1,014). In the secondary analyses, only prior SAs predicted prospective suicidality (OR = 3.87, p = 0.0029). Conclusions: These results indicate that patients with bipolar disorder who present with a history of SAs are over four times as likely to have a subsequent SA or completion. Further studies are needed to evaluate and prevent future attempts in this high-risk cohort.

Language: en