Suicide Risk in Mood Disorders

Zoltán Rihmer


Curr Opin Psychiatry. 2007;20(1):17-22. 

In This Article

Suicide Risk Factors in Patients with Mood Disorders

As suicidal behaviour manifests in patients with mood disorder in a state-dependent fashion,[8,9,10,11**] a majority (but not all) of suicide risk factors are timely related to the given mood episodes.

Risk Factors Related to Current or Past Mood Episodes

The most powerful clinically explorable suicide risk factors in mood disorders are mainly related to the major mood episodes. They are also called 'proximal' risk factors and are listed in Table 2 . Suicidal ideation, a major precursor of attempted and completed suicide[11**,14,17*,20,21*] that shows high consistency across MDEs[22**] and, of course, recent suicide attempt are the most alarming signs of the short-term suicide risk.

Patients with minor depression and pure dysthymic disorder (dysthymia without 'comorbid' major depression) are relatively underrepresented among suicide victims and attempters,[4,7,23,24] which is consistent with the universal finding that severity of symptoms of depression (particularly in the presence of hopelessness and guilt) is a significant suicide risk factor in patients with depression.[11**,12*,14,17*,25,26**,27] The suicide risk in hospitalized patients with mood disorder is very high[16**,17*] and peaks immediately after hospital admission and discharge, particularly in the case of short hospital treatment.[28*]

With regard to the role of the unipolar-bipolar nature of mood disorders in suicide prediction, a recent review[29] of 10 published studies including more than 3000 patients has concluded that bipolar patients, in general, and bipolar II patients, in particular, are overrepresented among both committed and attempted suicides. Another recent study[11**] investigating 90 bipolar I and 101 bipolar II patients also found a (nonsignificant) trend for higher rate of prior suicide attempts at index episode in bipolar II (25%) than in bipolar I patients (16%). On the contrary, however, another large-scale, 40-44-year follow-up study[16**] has found that a higher rate of 186 unipolar (14%) than the 220 bipolar (I + II) patients (8%) committed suicide.

A relatively newly recognized important proximate suicide risk factor in MDE might be the mixed state of depression (three or more simultaneously cooccurring intradepressive hypomanic symptoms),[30**,31,32**] as the frequency of past suicide attempts and suicidal ideation has been reported to be much higher among mixed than nonmixed unipolar and bipolar patients with major depression.[31,33,34,35] In our most recent study,[36**] we used the opposite strategy, and we analysed the frequency and clinical characteristics of mixed states of depression among the consecutively investigated 100 nonviolent suicide attempters. Current mixed depression was present in 63% of the total sample and in 71% among the 89 suicide attempters with depression. Irritability, distractibility and psychomotor agitation were present in more than 90% of patients with mixed depression. The rate of mixed depression was significantly higher among the 29 patients with bipolar (I + II) depression than in 60 suicide attempters with nonbipolar depression (90 and 62%, respectively). These figures are almost two times higher than the same rates (59 and 27%, respectively) that were previously reported among 241 bipolar II and 164 nonsuicidal outpatients with unipolar major depression.[37] These findings indicate that suicide attempters with depression come primarily from mixed bipolar and unipolar depressions and support the role of mixed depression in suicidal behaviour. These results can also explain, at least in part, why bipolar II depression (which is most often mixed in nature) carries the highest suicide risk among all subtypes of major mood disorders.[29]

Investigating 247 adolescent outpatients with current MDE (100 of them have had bipolar I or bipolar II disorder), a close link between suicidality and bipolar mixed depression has also been observed but only for girls: out of the 82 patients with mixed depression, girls had nearly four times the risk of having made a past suicide attempt compared with patients having nonmixed depression.[38]

Another recent study[30**] of ours has demonstrated that outpatient 'unipolar' mixed major depression and agitated depression are greatly overlapping conditions (90% of patients with agitated depression also met the criteria for mixed state of depression), and a significantly higher rate of patients with agitated than nonagitated depression (62 versus 43%) showed current suicidal ideation and positive family history of bipolar II disorder (24 versus 12%). Of symptoms of mixed depression, we found a significant association between suicidal ideation, psychomotor activation and racing thought. The results support the view that agitated unipolar depression should also be classified as a bipolar mixed state of depression and are in good agreement with prior studies,[5,34,39] showing that agitation is a suicide risk factor in patients with depression.

The role of 'mood instability' (i.e., 'bipolarity') in the suicidal behaviour was also supported by two other recent studies,[40**,41] showing that a history of rapid mood switching and panic attacks was associated with an increased likelihood of history of self-reported suicidal thought or action,[40**] and variability in suicidal ideation was a significantly better predictor of previous suicide attempts than duration and intensity of ideation.[41] The persistent, frequent and marked instabilities of mood, thinking and behaviour are the most characteristic features of cyclothymia, which is the attenuated manifestation of major bipolar mood disorders. Cyclothymia might be a predisposing factor for suicidal behaviour, as two recently published studies[42,43**] found that in patients with MDE, cyclothymic personality was significantly related to lifetime and current suicidal behaviour (ideation and attempts) both in an adult sample[42] and in a paediatric sample.[43**]

The recognition of the important role of 'pseudo-unipolar' mixed states of depression in suicidal behaviour[1*,30**,31,36**] has clear implications for suicide prevention. The correct identification of the 'covert' bipolar nature of the given episode of depression, as reflected in mixed/agitated clinical picture, is crucial for selecting the most appropriate treatment. A growing body of evidence exists that antidepressant monotherapy, unprotected by mood stabilizers or atypical antipsychotics in patients with bipolar and bipolar spectrum (including 'unipolar' mixed states of depression and unipolar depression with cyclothymic personality and with bipolar family history) can not only produce (hypo)manic switches and rapid cycling but also worsen the preexisting mixed state of depression or generate mixed conditions de novo, resulting in treatment resistance, destabilization of the mood disorder, worsening of the depression and, ultimately, suicidal behaviour in few patients.[1*,30**,31] In other words, the rarely observed suicidality-antidepressant link appears mediated by agitated, excited mixed states of depression, and recent data suggest that concomitant use of mood stabilizers, atypical antipsychotics or benzodiazepines in such cases could prevent the newly developing suicidality in this high-risk population.[1*,30**,31]

The psychotic/nonpsychotic nature of major mood episodes does not seem fundamental regarding suicidal behaviour in unipolar depression.[44] Psychotic symptoms, however, were significantly associated with completed suicide in adults[16**] and with suicidal ideation and plans in paediatric[45] bipolar I and bipolar II patients.

In agreement with the previous findings, recent studies[6,12*,14,21*,27,39,46,47,48] also show that comorbid anxiety/anxiety disorders and comorbid substance use disorders as well as concomitant serious medical illnesses increase the risk of all forms of suicidal behaviour in unipolar depression and bipolar disorder.

As untreated MDE is the most frequent diagnosis in suicide victims and attempters,[1*,3,4,16**,23,29] suicidal behaviour in bipolar patients is not exclusively restricted to episodes of depression as mixed (major) affective episode (meeting the full syndromal criteria for mania and major depression at the same time) also increases the risk of attempted and completed suicide.[9,11**,12*] In addition, a recent study[49] demonstrated that the distinction between mixed (dysphoric or depressive) mania and pure (euphoric) mania is also crucial for predicting suicide risk: more than 40% of patients with dysphoric mania had current suicidal ideation or attempt, whereas for patients with pure mania it was less than 10%.

Risk Factors Unrelated to Current or Past Mood Episodes

The fact that adverse childhood experiences and current psychosocial stressors have a predisposing and triggering role in suicidal behaviour is well known. Although clinicians cannot influence what has happened in the past history of their patients, collecting information about family history and early development as well as current psychosocial circumstances is also important in predicting suicide risk. The well recognized (familial and psychosocial) suicide risk factors in patients with mood disorder that are not directly related to current or past mood episodes (e.g., distal suicide risk factors) are listed in Table 3 .

The voluminous literature on this subject consistently shows that family history of committed suicide (and mood disorder) in first-degree relatives,[12*,27,47,50] adverse childhood experiences (parental loss and emotional, physical and sexual abuse),[12*,21*,47,50,51*,52] early onset of mood disorder,[12*,16**,27,47,48] comorbid (mostly cluster B) personality disorder and aggressive/impulsive personality features, as well as cigarette smoking,[11**,14,23,24,26**,51*,53,54*,55*] adverse life situations (unemployment, isolation and acute psychosocial stressors)[3,6,21*,26**,55*] and lack of emotional, social and medical/psychiatric support[3,9,16**,24,29] are significantly associated with completed and attempted suicides, both in the general population and in patients with mood disorders.

In contrast to the fact that men are markedly overrepresented among unselected suicide victims compared with women among suicide attempters,[3,5,19,21*,23,28*] gender is not a significant predictor of committed[16**,17*] and attempted[14,26**,27,47,52,53] suicides in patients with unipolar or bipolar mood disorder.


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.
Post as: