Risk Factors for Suicide Identified in BRIDGE-II-MIX Study

Pam Harrison

August 31, 2015

AMSTERDAM — Use of current diagnostic criteria may lead clinicians to miss a substantial proportion of patients with a major depressive episode (MDE) who have a mixed depressive state, as determined on the basis of less rigid diagnostic criteria, results from the BRIDGE-II-MIX study indicate.

Moreover, a number of risk factors in patients who are identified as having a mixed depressive state substantially increase the likelihood of suicide, and factoring these risk factors into the treatment decision may help prevent suicidal death.

The study was presented here at the 28th European College of Neuropsychopharmacology (ECNP) Congress.

"The primary aim of the study was to see how many patients who had attempted suicide were actually affected by mixed depression, and the second aim was to see which factors might predict suicidal behavior in patients who had attempted suicide," Dina Popovic, MD, PhD, Barcelona Bipolar Program, Hospital Clinic, University of Barcelona, Spain, told Medscape Medical News. The investigators used current DSM-5 criteria for MDE.

"We found that suicide was most significantly associated with risky behavior, psychomotor agitation, and impulsivity, and two of these risk factors are not in the DSM-VI either," she said.

"We also found that standard DSM criteria identified approximately 12% of patients as showing mixed states, whereas our methods showed almost 40% of patients at risk, so this means that standard methods are missing a lot of patients at risk of suicide."

The BRIDGE-II-MIX study, funded by sanofi-aventis, was a multicenter, multinational, cross-sectional study involving 2811 patients identified as having MDE. Psychiatric symptoms and sociodemographic and clinical variables representing risk factors for bipolar disorder were collected.

A history of suicide attempts was documented in 628 patients, or approximately 22% of the overall cohort.

Patients who had a history of a suicide attempt were significantly more likely to be female (P = .028), know a first-degree relative with bipolar disorder (P < .0001), and have either psychotic or atypical features (P < .001, P = .009, respectively) than patients with an MDE who did not have a history of suicide.

Patients who had attempted suicide in the past were also approximately twice as likely to have had their antidepressant therapy switched because of hypomania or manic symptoms, to be resistant to treatment, or to have either mood lability or irritability as part of their symptomatology than patients with no history of suicide (all P < .0001).

Table. Antidepressant Response

End Point MDE-Suicide History MDE-No Suicide History Odds Ratio
(Hypo)manic switches (%) 24.8 14.4 1.97
Treatment resistance (%) 40.6 24.8 2.07
Mood lability (%) 41.2 26.2 1.98
Irritability (%) 35.2 23.2 1.80


Furthermore, three major symptoms were highly associated with a prior suicide attempt. Risky behavior, for example, increased the risk for suicide by more than twofold (odds ratio [OR], 2.11; P < .0001).

Psychomotor agitation was also associated with an increased risk for risk (OR, 1.42; P = .003); impulsivity increased the risk for suicide by 26% (OR, 1.26; P =.089).

Borderline personality disorder and substance abuse disorder were the two comorbid diagnoses most frequently associated with suicide attempts, investigators add.

"We know that mixed episodes are absolutely associated with bipolarity and also suicide, so it is important to ask patients directly about suicidal ideation, to ask, for example, if patients had engaged in reckless or impulsive behaviors," Dr Popovic said. "If they find these risk factors, management should change, because at that point, you know the patient doesn't have unipolar depression but likely has bipolar depression, and you need to use mood stabilizers or atypical antipsychotics for the treatment of bipolar disorder."


Asked to comment on the study, Erkki Isometsa, MD, PhD, professor of psychiatry, University of Helsinki, Finland, told Medscape Medical News that risk factors for suicide have been investigated extensively.

"There have also been recent meta-analyses on these factors, so we have information on risk factors," he said.

However, as Dr Isometsa pointed out, the value of the BRIDGE-II-MIX study is its size.

"Every time you investigate risk factors in a single large study, you can avoid some of the difficulties in meta-analyses, which are always dependent on the quality of the original studies, which may be small and heterogeneous," Dr Isometsa said. "So large-scale studies are valuable because they confirm findings at a different level of significance."

Dr Isometsa also noted that the BRIDGE-II-MIX study involved only clinicians who are experts in mood disorders.

"Large-scale studies are often epidemiological in nature, and this means that everyone who is interviewing patients can't be an expert," he said.

"Now we have a study where a lot of the interviewing clinicians are experts, and they are better at detecting certain clinical features than others who have only been trained to use some standardized interview."

"So I think the level of expertise and the size of the study are important factors to consider when interpreting its findings."

The study was sponsored by sanofi-aventis. Dr Popovic has disclosed no relevant financial relationships.

28th European College of Neuropsychopharmacology (ECNP) Congress. Abstract P.2.d.003. Presented August 30, 2015.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.