Suicide and deliberate self-harm in personality disorders

PURPOSE OF REVIEW: This article reviews literature published over the period January 2004-May 2005 on suicidal behaviour and self-harm in personality disorders. RECENT FINDINGS: Studies have confirmed that personality disorders and their co-morbidity with other psychiatric conditions are risk factors for both fatal and nonfatal suicidal behaviours, and self-mutilation. Negative life events, childhood sexual abuse, difficulties in social functioning, deficits in future-directed thinking and time perception, as well as familial and neurocognitive factors may be related to increased suicide risk in individuals with borderline and other personality disorders. Findings seem to confirm that suicidality and self-injurious behaviour are efficient DSM-IV diagnostic criteria for borderline personality disorder. Out of several psychosocial and pharmacological interventions for treating suicidality in personality disorders, only one randomized, controlled study has recently been published. Medico-legal concerns related to the clinical management of chronically suicidal patients, including hospitalization and alternative treatment approaches, are also discussed. SUMMARY: Although recent studies have contributed to the theoretical knowledge and clinical practice, there are unsettled questions that should be addressed in the future. More randomized, controlled trials evaluating the efficacy of interventions in suicidal individuals with personality disorders should be conducted. As the majority of studies conducted to date have concentrated on borderline personality disorder and antisocial personality disorder, the prevalence and risk factors for suicidal behaviours and self-mutilation in other personality disorders require further clarification. The introduction of unified nomenclature related to suicidal behaviours and self-mutilation would facilitate comparability of results across studies.     

Suicidal ideation and suicide attempts among patients with panic disorder: a survey of two outpatient clinics.

OBJECTIVE: Recent epidemiologic studies found that 20% of subjects with the diagnosis of panic disorder had attempted suicide. This study sought to determine the prevalence of suicidal ideation and suicide attempts among patients with panic disorder and whether the presence of comorbid borderline personality disorder influenced the prevalence of suicidal thoughts and behavior. METHOD: At two outpatient clinics, experienced clinicians conducted retrospective reviews of data from the intake interviews and charts of 59 patients with panic disorder and comorbid borderline personality disorder and 234 patients with panic disorder with or without axis II disorders other than borderline personality disorder. RESULTS: Suicide attempts were reported by 2% of the patients with panic disorder, compared to 25% of the patients with both panic disorder and borderline personality disorder. In addition, 2% of the patients with panic disorder, compared to 27% of the patients with panic disorder and borderline personality disorder, reported suicidal ideation that was judged to be of clinical significance. CONCLUSIONS: The rate of suicidal ideation and suicide attempts for psychiatric outpatients with panic disorder was discrepant with the findings of the earlier studies. The increased suicide risk in this group of patients was associated with borderline personality disorder, increased substance abuse, and affective instability. While 61% of the panic disorder patients and 78% of the patients with both panic disorder and borderline personality disorder reported thinking about death, this must be distinguished from actual suicidal ideation and clinical risk of suicide.     

Bipolar disorder with comorbid cluster B personality disorder features: impact on suicidality

BACKGROUND: Because of their overlapping phenomenology and mutually chronic, persistent nature, distinctions between bipolar disorder and cluster B personality disorders remain a source of unresolved clinical controversy. The extent to which comorbid personality disorders impact course and outcome for bipolar patients also has received little systematic study. METHOD: One hundred DSM-IV bipolar I (N = 73) or II (N = 27) patients consecutively underwent diagnostic evaluations with structured clinical interviews for DSM-IV Axis I and cluster B Axis II disorders, along with assessments of histories of childhood trauma or abuse. Cluster B diagnostic comorbidity was examined relative to lifetime substance abuse, suicide attempt histories, and other clinical features. RESULTS: Thirty percent of subjects met DSM-IV criteria for a cluster B personality disorder (17% borderline, 6% antisocial, 5% histrionic, 8% narcissistic). Cluster B diagnoses were significantly linked with histories of childhood emotional abuse (p = .009), physical abuse (p = .014), and emotional neglect (p = .022), but not sexual abuse or physical neglect. Cluster B comorbidity was associated with significantly more lifetime suicide attempts and current depression. Lifetime suicide attempts were significantly associated with cluster B comorbidity (OR = 3.195, 95% CI = 1.124 to 9.088), controlling for current depression severity, lifetime substance abuse, and past sexual or emotional abuse. CONCLUSIONS: Cluster B personality disorders are prevalent comorbid conditions identifiable in a substantial number of individuals with bipolar disorder, making an independent contribution to increased lifetime suicide risk.     

Comorbid personality disorders in subjects with panic disorder: do personality disorders increase clinical severity?

Personality disorders are common in subjects with panic disorder. Personality disorders have been shown to affect the course of panic disorder. The purpose of this study was to examine which personality disorders affect clinical severity in subjects with panic disorder. This study included 122 adults (71 women, 41 men) who met Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition ( DSM-IV ) criteria for panic disorder (with or without agoraphobia). Clinical assessment was conducted by using the Structured Clinical Interview for DSM-IV Axis I Disorders, the Structured Clinical Interview for DSM-IV Axis II Personality Disorders, and the Panic and Agoraphobia Scale, Global Assessment Functioning Scale, Beck Depression Inventory, and State-Trait Anxiety Inventory. Patients who had a history of sexual abuse were assessed with Sexual Abuse Severity Scale. Logistic regressions were used to identify predictors of suicide attempts, suicidal ideation, sexual abuse, and early onset of disorder. The rates of comorbid Axes I and II psychiatric disorders were 80.3% and 33.9%, respectively, in patients with panic disorder. Patients with panic disorder with comorbid personality disorders had more severe anxiety, depression, and agoraphobia symptoms, had earlier ages at onset, and had lower levels of functioning. The rates of suicidal ideation and suicide attempts were 34.8% and 9.8%, respectively, in subjects with panic disorder. The rate of patients with panic disorder and a history of childhood sexual abuse was 12.5%. The predictor of sexual abuse was borderline personality disorder. The predictors of suicide attempt were comorbid paranoid and borderline personality disorders, and the predictors of suicidal ideation were comorbid major depression and avoidant personality disorder in subjects with panic disorder. In conclusion, this study documents that comorbid personality disorders increase the clinical severity of panic disorder. Borderline personality disorder may be the predictor of a history of sexual abuse and early onset in patients with panic disorder. Paranoid and borderline personality disorders may be associated with a high frequency of suicide attempts in patients with panic disorder.     

Long-term follow-up of narcissistic/borderline patients

In a long-term follow-up study of hospitalized border-line patients with narcissistic traits that either fell short of or fulfilled DSM criteria for narcissistic personality disorder (NPD), outcome was similar to outcome in the borderline group as a whole. An exception was encountered in the subgroup of narcissistic borderlines who also showed marked antisocial traits: poor outcome was the rule in this subgroup. In general, borderlines with NPD tended to be male and to be more at risk for suicide than non-NPD borderlines.     

Characteristics of suicide attempts of patients with major depressive episode and borderline personality disorder: a comparative study

OBJECTIVE: Suicidal behavior is highly prevalent in borderline personality disorder and major depressive episode, although the characteristics of suicide attempts in the two disorders are believed to differ. Comorbidity of borderline personality disorder and major depressive episode may obscure characteristics of suicide attempts that are uniquely related to the psychopathology of each disorder. We compared suicidal behavior in patients with borderline personality disorder, major depressive episode, and borderline personality disorder plus major depressive episode to determine whether characteristics of suicide attempts differed between groups and if aspects of core psychopathology predicted specific attempt characteristics. METHOD: Eighty-one inpatients with borderline personality disorder, including 49 patients with borderline personality disorder plus major depressive episode, were compared to 77 inpatients with major depressive episode alone on measures of depressed mood, hopelessness, impulsive aggression, and suicidal behavior, including lifetime number of attempts, degree of lethal intent, objective planning, medical damage, and degree of violence of suicide methods. RESULTS: No significant differences were found in the characteristics of suicide attempts between patients with borderline personality disorder and those with major depressive episode. However, patients with both disorders had the greatest number of suicide attempts and the highest level of objective planning. An increase in either impulsive aggression or hopelessness or a diagnosis of borderline personality disorder predicted a greater number of attempts. Hopelessness predicted lethal intent in all three groups and predicted objective planning in the group with both disorders. Medical damage resulting from the most serious lifetime suicide attempt was predicted by number of attempts. CONCLUSIONS: Comorbidity of borderline personality disorder with major depressive episode increases the number and seriousness of suicide attempts. Hopelessness and impulsive aggression independently increase the risk of suicidal behavior in patients with borderline personality disorder and in patients with major depressive episode.     

Social risk factors for borderline personality disorder: a review and hypothesis.

A number of behaviours associated with borderline personality disorder (including attempted suicide, suicide, substance abuse, and antisocial behaviour) are on the increase among the young. The common factor in these disorders is impulsiveness. Evidence is reviewed suggesting that social disintegration reduces the threshold of impulsive behaviours. It is proposed that this is the mechanism through which social risk factors effect the prevalence and morbidity of borderline personality. A number of ways of testing this hypothesis are suggested.     

Personality Disorders in Suicide Attempters and Completers: Preliminary Findings

Personality disorders may play an important role in understanding suicide risk. The present study was designed to examine the frequency and type of personality disorder traits relevant to suicidal behavior. Four groups of subjects were compared: Suicide completers (n = 15), suicide attempters (n = 14), depressed natural deaths (n = 13), and nondepressed natural deaths (n = 15). The Structured Interview for DSM-III-R Personality Disorders – Revised (SIDP-R) was used to gather information needed to rate the presence and severity of 11 personality disorders. No differences were observed across groups on the categorical presence or absence of the different personality disorder diagnoses. However, dimensional ratings revealed that suicide completers displayed significant elevations in narcissistic, histrionic, and borderline personality traits. Suicide attempters displayed significant elevations on paranoid, avoidant, schizotypal, dependent, and borderline personality traits. At the level of specific personality traits, suicide completers displayed a sense of entitlement toward others, tended to be preoccupied with feelings of envy, and were likely to feel devastated when close relationships came to an end. The present findings suggest that personality traits are related to suicidal behavior, and that a focus on the dichotomous presence or absence of a diagnostic category fails to identify less severe forms of personality pathology.

     

Sex differences in clinical predictors of suicidal acts after major depression: a prospective study

OBJECTIVE: Whether sex differences exist in clinical risk factors associated with suicidal behavior is unknown. The authors postulated that among men with a major depressive episode, aggression, hostility, and history of substance misuse increase risk for future suicidal behavior, while depressive symptoms, childhood history of abuse, fewer reasons for living, and borderline personality disorder do so in depressed women. METHOD: Patients with DSM-III-R major depression or bipolar disorder seeking treatment for a major depressive episode (N=314) were followed for 2 years. Putative predictors were tested with Cox proportional hazards regression analysis. RESULTS: During follow-up, 16.6% of the patients attempted or committed suicide. Family history of suicidal acts, past drug use, cigarette smoking, borderline personality disorder, and early parental separation each more than tripled the risk of future suicidal acts in men. For women, the risk for future suicidal acts was sixfold greater for prior suicide attempters; each past attempt increased future risk threefold. Suicidal ideation, lethality of past attempts, hostility, subjective depressive symptoms, fewer reasons for living, comorbid borderline personality disorder, and cigarette smoking also increased the risk of future suicidal acts for women. CONCLUSIONS: These findings suggest that the importance of risk factors for suicidal acts differs in depressed men and women. This knowledge may improve suicide risk evaluation and guide future research on suicide assessment and prevention.     

Suicide risk in mood disorders

PURPOSE OF REVIEW: The aim of this review is to highlight the traditional and newly recognized suicide risk factors in patients with mood disorders. RECENT FINDINGS: Current research findings clearly suggest that suicidal behaviour in patients with mood disorder is a 'state-dependent' phenomenon. Recently, there is, however, a growing body of evidence that besides the well accepted clinically explorable suicide risk factors in mood disorders (e.g., severe depression, prior suicide attempt, comorbid anxiety, substance use, personality disorders and so on), mixed state of depression could also be an important precursor of suicidal behaviour. This might be particularly true in unrecognized cases of bipolar depressives, when antidepressant monotherapy (unprotected by mood stabilizers or atypical antipsychotics) can worsen the clinical picture and rarely induce an aggressive or self-destructive behaviour. SUMMARY: In the majority of patients with mood disorders, suicidal behaviour is predictable and preventable, with a good chance. A careful and systematic exploration of suicide risk factors in patients with mood disorder helps clinicians to identify patients at high suicide risk. A successful, acute and long-term treatment of these patients substantially reduces the suicidal behaviour even in this high-risk population.     

Borderline personality disorder criteria associated with prospectively observed suicidal behavior

OBJECTIVE: The authors prospectively examined associations between each DSM-IV borderline personality disorder criterion and suicidal behaviors. METHOD: Borderline personality disorder diagnosis and criteria, major depressive disorder, substance use disorders, and history of childhood sexual abuse were assessed with semistructured interviews. Participants (N=621) were followed for 2 years with repeated structured evaluations that included assessments of suicidality. RESULTS: With the self-injury criterion excluded, the borderline personality disorder criteria of affective instability, identity disturbance, and impulsivity significantly predicted suicidal behaviors. Only affective instability and childhood sexual abuse were significantly associated with suicide attempts (i.e., behavior with some intent to die). CONCLUSIONS: Affective instability is the borderline personality disorder criterion (excluding self-injury) most strongly associated with suicidal behaviors. Since major depressive disorder did not significantly predict suicidal behaviors, the reactivity associated with affective instability (more so than negative mood states) appears to be a critical element in predicting suicidal behaviors.     

Comorbid Personality Disorders and Violent Behavior in Psychotic Patients

Schizophrenia without any comorbidity confers a modest, but statistically significant elevation of the risk for violence. That risk is considerably increased by comorbid antisocial personality disorder or psychopathy as well as by comorbid substance use disorders. These comorbidities are frequent. Conduct disorder and conduct disorder symptoms elevate the risk for aggressive behavior in patients with schizophrenia. Violence among adults with schizophrenia may follow at least two distinct pathways-one associated with premorbid conditions, including antisocial conduct, and another associated with the acute psychopathology of schizophrenia. Aggressive behavior in bipolar disorder occurs mainly during manic episodes, but it remains elevated in euthymic patients in comparison with controls. The risk of violent behavior is increased by comorbidity with borderline personality disorder, antisocial personality disorder, and substance use disorders. These comorbidities are frequent. Borderline personality disorder and bipolar disorder are related in their phenomenology and response to medication. These two disorders share a tendency to impulsiveness, and impulsive behavior, including impulsive aggression, is particularly expressed when they co-occur.     

P3 event-related potential amplitude and the risk for disinhibitory disorders in adolescent boys

BACKGROUND: The children of parents who abuse alcohol typically show reduced amplitude of the P3 event-related potential wave. We determined if this effect was present in a population-based sample of older adolescent boys, whether it was associated with paternal antisocial personality and drug use, and whether it appeared in youth with childhood externalizing and substance use disorders. METHODS: A statewide sample of 502 male youth, identified from Minnesota birth records as members of twin pairs, had their P3 amplitude measured, using a visual oddball paradigm when they were approximately 17 years old. Structured clinical interviews covering attention-deficit/hyperactivity disorder, conduct disorder, oppositional defiant disorder, antisocial personality disorder, and substance use disorders were administered in person to the youth and his parents at the time of the P3 assessment and again to the youth 3 years later. RESULTS: Reduced P3 was associated with disorders and paternal risk for disorders, reflecting a behavioral disinhibition spectrum that included attention-deficit/hyperactivity disorder, oppositional defiant disorder, conduct disorder, antisocial personality disorder, alcoholism, nicotine dependence, and illicit drug abuse and dependence. Reduced P3 at age 17 predicted the development of substance use disorders at age 20. Most effect sizes associated with these group differences exceeded 0.70, indicating medium to moderately large group differences. Maternal alcoholism and substance use during pregnancy were unrelated to P3 amplitude in offspring. CONCLUSION: Small amplitude P3 may indicate genetic risk for a dimension of disinhibiting psychiatric disorders, including childhood externalizing, adult antisocial personality disorder, and substance use disorders.     

Association of comorbid posttraumatic stress disorder and major depression with greater risk for suicidal behavior

OBJECTIVE: Posttraumatic stress disorder (PTSD) increases the risk of suicidal behavior; a major depressive episode also increases the risk for suicidal behavior. The authors' goal was to examine the effect of comorbid PTSD and major depressive episode on suicidal behavior. METHOD: Inpatients with a diagnosis of major depressive episode (N=156) were assessed for PTSD, suicidal behavior, and clinical risk factors for suicidal acts. RESULTS: Patients with comorbid major depressive episode and PTSD were more likely to have attempted suicide, and women with both disorders were more likely to have attempted suicide than men with both disorders. Cluster B personality disorder and PTSD were independently related to history of suicide attempts. CONCLUSIONS: The greater rate of suicide attempts among patients with comorbid PTSD and major depressive episode was not due to differences in substance use, childhood abuse, or cluster B personality disorders.     

Risk factors for suicide and attempted suicide among young people

OBJECTIVE: Suicide rates in young people have increased during the past three decades, particularly among young males, and there is increasing public and policy concern about the issue of youth suicide in Australia and New Zealand. This paper summarises current knowledge about risk factors for suicide and suicide attempts in young people. METHOD: Evidence about risk factors for suicidal behaviour in young people was gathered by review of relevant English language articles and other papers, published since the mid-1980s. RESULTS: The international literature yields a generally consistent account of the risk factors and life processes that lead to youth suicide and suicide attempts. Risk factor domains which may contribute to suicidal behaviour include: social and educational disadvantage; childhood and family adversity; psychopathology; individual and personal vulnerabilities; exposure to stressful life events and circumstances; and social, cultural and contextual factors. Frequently, suicidal behaviours in young people appear to be a consequence of adverse life sequences in which multiple risk factors from these domains combine to increase risk of suicidal behaviour. CONCLUSIONS: Current research evidence suggests that the strongest risk factors for youth suicide are mental disorders (in particular, affective disorders, substance use disorders and antisocial behaviours) and a history of psychopathology, indicating that priorities for intervening to reduce youth suicidal behaviours lie with interventions focused upon the improved recognition, treatment and management of young people with mental disorders.     

The longitudinal pattern of suicidal behaviour in borderline personality disorder: a prospective follow-up study

The aim was to study the longitudinal course of suicidal behaviour and ideation in patients with borderline personality disorder (BPD) compared with patients with other diagnoses. Ninety-seven patients (41 BPD, 33 other personality disorders, 23 no personality disorder) consecutively admitted to a day unit were given a prospective personal interview follow-up with evaluations at admission, discharge and at follow-up after 2–5 years. Even when controlled for Axis I disorders, BPD patients showed significantly more often a lifetime history of suicide attempts. BPD patients with a history of suicide attempts were more suicidal at index admission, continued to be so over the follow-up period and differed systematically in an unfavourable direction from other BPD patients on the major outcome measures. BPD patients without suicidal behaviour had an outcome nearly as good as non-BPD patients, and only 41% of them retained the BPD diagnosis at follow-up. Suicidal behaviour and ideation are highly prevalent in BPD. These suicidal expressions are of an enduring nature and seem as a diagnostic criterion to enhance the predictive capacity of the BPD diagnosis.     

Risk of suicide in mood disorders

Suicide is commonly associated with mood disorders. Risk factors for suicide in mood disorders can be organized according to whether their effect is on the threshold or diathesis for suicidal acts or whether they serve mainly as triggers or precipitants of suicidal acts. Predisposition to suicidal behavior or diathesis is a key element that helps to differentiate patients who are at high risk versus those at lower risk. The objective severity of mood disorders does not identify depressed patients at high risk for suicide attempt. There is a lack of agreement over the suicide risk associated with characteristics of depression such as psychotic features, agitation, or anxiety, or mixed mood states as part of bipolar disorder. Risk factors affecting the diathesis for suicidal behavior include family history of suicide, low cerebrospinal fluid 5-hydroxyindolacetic acid, alcohol and/or substance abuse, cluster B personality disorder, high past impulsivity and aggression, chronic physical illness particularly involving the brain, marital isolation, parental loss before age 11, childhood history of physical and sexual abuse, hopelessness, and not living with a child under age 18. Most common precipitants of suicidal acts in mood disorders include interpersonal losses or conflicts, financial trouble, and job problems. Identification of high risk patients and effective treatment are required for suicide prevention to reduce morbidity and mortality in affective disorders.     

Personality disorder in the families of depressed, schizophrenic, and never-ill probands

In a blind family study of 176 probands with nonpsychotic major depression, psychotic major depression, schizophrenia, or no history of DSM-III disorders, only the relatives of depressed probands with mood-incongruent psychotic features had a risk for personality disorders higher than that for the relatives of never-ill probands. The authors did not find a high rate of borderline personality in relatives of depressed probands or of schizotypal personality disorder in relatives of probands with schizophrenia or any psychosis. However, depressed probands with normal dexamethasone test results had a significantly higher familial loading for the DSM-III cluster of histrionic, antisocial, borderline, and narcissistic personality disorders.     

Chronic suicidality among patients with borderline personality disorder

OBJECTIVE: This paper reviews research on chronic suicidality among patients with borderline personality disorder. METHODS: MEDLINE and PsycINFO databases were searched for all English-language articles published between 1984 and 2000 containing the keywords "borderline personality disorder" and "suicide" or "suicidality." A total of 170 articles located through this search and additional key articles published before 1990 were reviewed. The most relevant articles were selected of review. RESULTS AND CONCLUSIONS: One in ten patients with borderline personality disorder completes suicide, but this outcome is not readily preventable and does not necessarily occur during the course of treatment. In outpatient psychotherapy, chronic suicidal behavior by patients with borderline personality disorder can be best understood as a way of communicating distress. Hospitalization is of unproven value in preventing suicide by these patients and can sometimes have negative effects. Clinicians' fear of potential litigation resulting from a completed suicide should not be the basis for admission. With no evidence that full hospitalization prevents suicide completion by patients with borderline personality, suicidal risk is not a contraindication for day hospital treatment.     

Hostility and childhood sexual abuse as predictors of suicidal behaviour in Borderline Personality Disorder

Impulsivity is a multidimensional construct and has been previously associated with suicidal behaviour in borderline personality disorder (BPD). This study examined the associations between suicidal behaviour and impulsivity-related personality traits, as well as history of childhood sexual abuse, in 76 patients diagnosed with BPD using both the Structured Interview for Diagnostic and Statistical Manual of Mental Disorders III (DSM-III) Axis-II diagnoses and the self-personality questionnaire. Impulsivity-related traits were measured using the Barratt Impulsiveness Scale-11 (BIS-11), the Buss–Durkee Hostility Inventory (BDHI) and the Temperament and Character Inventory-Revised (TCI-R). We found that hostility and childhood sexual abuse, but not impulsivity or other temperament traits, significantly predicted the presence, number and severity of previous suicide attempts. Hostility traits and childhood sexual abuse showed an impact on suicide attempts in BPD. Our results support previous findings indicating that high levels of hostility and having suffered sexual abuse during childhood lead to an increased risk for suicidal behaviour in BPD.