Correlates of Historical Suicide Attempt in Rapid-Cycling Bipolar Disorder: A Cross-Sectional Assessment

Objective: A rapid-cycling course in bipolar disorder has previously been identified as a risk factor for attempted suicide. This study investigated factors associated with suicide attempts in patients with rapid-cycling bipolar I or II disorder.

Method: Cross-sectional data at the initial assessment of patients who were enrolled into 4 clinical trials were used to study the factors associated with suicide attempt. An extensive clinical interview and the Mini-International Neuropsychiatric Interview were used to ascertain DSM-IV diagnoses of rapid-cycling bipolar disorder, substance use disorders, anxiety disorders, psychosis, and other clinical variables. Chi-square, t test, and logistic regression or Poisson regression were used to analyze the data where appropriate, with odds ratios (ORs) for relative risk estimate. The data were collected from September 1995 to June 2005.

Results: In a univariate analysis, 41% of 561 patients had at least 1 lifetime suicide attempt. Earlier age of depression onset, bipolar I subtype, female sex, unmarried status, and a history of drug use disorder, panic disorder, sexual abuse, and psychosis were associated with significantly higher rates of attempted suicide (all p < .05). After considering 31 potential confounding factors in the stepwise logistic regression model (n = 387), any Axis I comorbidity (OR = 2.68, p = .0219), female sex (OR = 2.11, p = .0005), psychosis during depression (OR = 1.84, p = .0167), bipolar I subtype (OR = 1.83, p = .0074), and history of drug abuse (OR = 1.62, p = .0317) were independent predictors for increased risk of attempted suicide. However, white race was associated with a lower risk for suicide attempt (OR = 0.47, p = .0160). Psychosis during depression (p = .0003), bipolar I subtype (p = .0302), and physical abuse (p = .0195) were associated with increased numbers of suicide attempts by 248%, 166%, and 162%, respectively; white race was associated with a 60% decrease in the number of suicide attempts (p = .0320).

Conclusion: In this highly comorbid group of patients with rapid-cycling bipolar disorder, 41% had at least 1 suicide attempt. Among the demographics, female sex was positively associated, but white race was negatively associated, with the risk for suicide attempt. Independent clinical variables for increased risk and/or number of attempted suicides were any Axis I comorbidity, psychosis during depression, bipolar I subtype, a history of drug abuse, and physical abuse.