Unlike other forms of self-injury, suicidal self-injury has special meaning, particularly in the context of borderline personality disorder. How is suicidal self-injury differentiated from non-suicidal self-injury in these patients, and how can their behavior be properly assessed and treated?
Borderline personality disorder (BPD) is characterized by unstable relationships, self-image and affect, as well as impulsivity, that begin by early adulthood. Patients with BPD make efforts to avoid abandonment. They often exhibit recurrent suicidal and/or self-injuring behavior, feelings of emptiness, intense anger, and/or disassociation or paranoia. Suicidal and non-suicidal self-injury are extremely common in BPD. Zanarini et al. (1990) found that over 70% of patients with BPD had self-injured or made suicide attempts, as compared to only 17.5% of patients with other personality disorders. Nevertheless, clinicians consistently misunderstand and mistreat this aspect of BPD.
There has been considerable controversy surrounding the diagnosis of BPD, ranging from a sense that the term itself is misleading and frightening, to the fact that the diagnosis is often made in an inconsistent manner (Davis et al., 1993), to a lack of clarity about whether the diagnosis should be Axis I or Axis II (Coid, 1993; Kjellander et al., 1998). Furthermore, these patients are often excluded from clinical trials due to perceived risk.
More important, however, is the fact that suicidal self-injurious behavior is usually understood within the context of major depressive disorder, while the phenomenology of this behavior within BPD is quite different. In addition, self-injurious non-suicidal behavior is often understood by clinicians to be synonymous with suicidal behavior, but again, it may be distinguished separately, particularly within the context of BPD. It is possible that, although self-injury and suicidal behavior are distinct, they may serve similar functions. This phenomenon has important implications for treatment recommendations.
Suicidality in BPD Versus Major Depression
In traditional conceptualizations developed from suicidality seen as an aspect of major depression, suicidal behavior is usually understood to be a response to a deep sense of despair and desire for death, which, if unsuccessful, typically results in a persistence of depression. Vegetative signs are prominent, and the suicidal feelings subside when the major depression is successfully treated with antidepressants, psychotherapy or their combination. In contrast, suicidality in the context of BPD seems to be more episodic and transient in nature, and patients often report feeling better afterward.
Risk factors for suicidal behavior in BPD show some differences, as well as similarities, with individuals who are suicidal in the context of major depression. Brodsky et al. (1995) noted that dissociation, particularly in patients with BPD, is correlated with self-mutilation. Studies of comorbidity have produced unclear results. Pope et al. (1983) found that a large number of patients with BPD also display a major affective disorder, and Kelly et al. (2000) found that patients with BPD alone and/or patients with BPD plus major depression are more likely to have attempted suicide than patients with major depression alone. In contrast, Hampton (1997) stated that the completion of suicide in patients with BPD is often unrelated to a comorbid mood disorder (Mehlum et al., 1994) and to degree of suicidal ideation (Sabo et al., 1995).
Suicidal behavior is usually defined as a self-destructive behavior with the intent to die. Thus, there must be both an act and intent to die for a behavior to be considered suicidal. Non-suicidal self-harm generally implies self-destructive behavior with no intent to die and is often seen as being precipitated by distress, often interpersonal in nature, or as an expression of frustration and anger with oneself. It usually involves feelings of distraction and absorption in the act, anger, numbing, tension reduction, and relief, followed by both a sense of affect regulation and self-deprecation. Confusion in the field regarding the definition of the term parasuicide can lead to a misunderstanding of the differences in function and danger of suicidal and non-suicidal self-injury. Parasuicide, or false suicide, groups together all forms of self-harm that do not result in death--both suicide attempts and non-suicidal self-injury. Many people who engage in non-suicidal self-harm are at risk for suicidal behavior.
We propose that non-suicidal self-injury in BPD uniquely resides on a spectrum phenomenologically with suicidality. Perhaps the most distinguishing factor, as pointed out by Linehan (1993), is that self-injury may help patients to regulate their emotions--an area with which they have tremendous difficulty. The act itself tends to restore a sense of emotional equilibrium and reduces an internal state of turmoil and tension. One striking aspect is the fact that physical pain is sometimes absent or, conversely, may be experienced and welcomed, as validation of psychological pain and/or a means to reverse a sense of deadness. Patients often report feeling less upset following an episode. In other words, while the self-injury is borne out of a sense of distress, it has served its function and the patient's emotional state is improved. Biological findings pointing to relationships among impulsivity and suicidality support the notion that suicidality and self-mutilation, particularly within the context of BPD, may occur on a continuum (Oquendo and Mann, 2000; Stanley and Brodsky, in press).
It is crucial to recognize, however, that even if patients with BPD self-mutilate and attempt suicide for similar reasons, death may be the accidental and unfortunate result. Because patients with BPD try to kill themselves so often, clinicians often underestimate their intent to die. In fact, individuals with BPD who self-injure are twice as likely to commit suicide than others (Cowdry et al., 1985), and 9% of the 10% of outpatients who are diagnosed with BPD eventually commit suicide (Paris et al., 1987). Stanley et al. (2001) found that suicide attempters with cluster B personality disorders who self-mutilate die just as frequently but are often unaware of the lethality of their attempts, compared to patients with cluster B personality disorders who do not self-mutilate.
Treatment of Suicidal Behavior and Self-Injury
While non-suicidal self-harm can result in death, it is more likely not to and, in fact, only occasionally leads to serious injury such as nerve damage. Yet, patients are often hospitalized on a psychiatric unit in the same way that they would be for a frank suicide attempt. In addition, while the intent is most often to alter the internal condition, as opposed to an external condition, clinicians and those in relationships with self-injurers experience this behavior as manipulative and controlling. It has been noted that self-injury can elicit quite strong countertransference reactions from therapists.
Although there is clearly a biological component to this disorder, the results of pharmacologic interventions have been inconclusive. Different classes and types of medications are often used for different aspects of the behavior (e.g., sadness and affective instability, psychosis and impulsivity) (Hollander et al., 2001).
One class of psychological intervention has been cognitive-behavioral therapy (CBT), of which there are a few models, e.g., Beck and Freeman (1990), cognitive-analytic therapy (CAT) developed by Wildgoose et al. (2001), and an increasingly well-known form of CBT called dialectical behavior therapy (DBT), developed by Linehan (1993) specifically for BPD. Dialectical behavior therapy is characterized by a dialectic between acceptance and change, a focus on skill acquisition and skill generalization, and a consultation-team meeting. In the psychoanalytic arena, there is controversy as to whether a confrontative, interpretative approach (e.g., Kernberg, 1975) or a supportive, empathic approach (e.g., Adler, 1985) is more effective.
This paper addresses contemporary conceptual and treatment issues that come into play in understanding suicidal and self-injuring behavior in the context of BPD. Diagnostic issues and the phenomenology of self-injurious behavior are important to consider. Treatment approaches include pharmacologic interventions, psychotherapy and their combination.
Adler G (1985), Borderline Psychopathology and Its Treatment. New York: Aronson.
Beck AT, Freeman A (1990), Cognitive Therapy of Personality Disorders. New York: The Guilford Press.
Brodsky BS, Cloitre M, Dulit RA (1995), Relationship of dissociation to self-mutilation and childhood abuse in borderline personality disorder. Am J Psychiatry 152(12):1788-1792 [see comment].
Coid JW (1993), An affective syndrome in psychopaths with borderline personality disorder? Br J Psychiatry 162:641-650.
Cowdry RW, Pickar D, Davies R (1985), Symptoms and EEG findings in the borderline syndrome. Int J Psychiatry Med 15(3):201-211.
Davis RT, Blashfield RK, McElroy RA Jr (1993), Weighting criteria in the diagnosis of a personality disorder: a demonstration. J Abnorm Psychol 102(2):319-322.
Hampton MC (1997), Dialectical behavior therapy in the treatment of persons with borderline personality disorder. Arch Psychiatr Nurs 11(2):96-101.
Hollander E, Allen A, Lopez RP et al. (2001), A preliminary double-blind, placebo-controlled trial of divalproex sodium in borderline personality disorder. J Clin Psychiatry 62(3):199-203.
Kelly TM, Soloff PH, Lynch KG et al. (2000), Recent life events, social adjustment, and suicide attempts in patients with major depression and borderline personality disorder. J Personal Disord 14(4):316-326.
Kernberg OF (1975), Borderline Conditions and Pathological Narcissism. New York: Aronson.
Kjellander C, Bongar B, King A (1998), Suicidality in borderline personality disorder. Crisis 19(3):125-135.
Linehan MM (1993), Cognitive-Behavioral Treatment for Borderline Personality Disorder: The Dialectics of Effective Treatment. New York: The Guilford Press.
Mehlum L, Friis S, Vaglum P, Karterud S (1994), A longitudinal pattern of suicidal behavior in borderline disorder: a prospective follow-up study. Acta Psychiatr Scand 90(2):124-130.
Oquendo MA, Mann JJ (2000), The biology of impulsivity and suicidality. Psychiatr Clin North Am 23(1):11-25.
Paris J, Brown R, Nowlis D (1987), Long-term follow-up of borderline patients in a general hospital. Compr Psychiatry 28(6):530-535.
Pope HG Jr, Jonas JM, Hudson JI et al. (1983), The validity of DSM-III borderline personality disorder. A phenomologic, family history, treatment response, and long-term follow-up study. Arch Gen Psychiatry 40(1):23-30.
Sabo AN, Gunderson JG, Najavits LM et al. (1995), Changes in self-destructiveness of borderline patients in psychotherapy. A prospective follow-up. J Nerv Ment Dis 183(6):370-376.
Stanley B, Brodsky B (in press), Suicidal and self-injurious behavior in borderline personality disorder: the self-regulation model. In: Borderline Personality Disorder Perspectives: From Professional to Family Member, Hoffman P, ed. Washington, D.C.: American Psychiatric Press Inc.
Stanley B, Gameroff MJ, Michalsen V, Mann JJ (2001), Are suicide attempters who self-mutilate a unique population? Am J Psychiatry 158(3):427-432.
Wildgoose A, Clarke S, Waller G (2001), Treating personality fragmentation and dissociation in borderline personality disorder: a pilot study of the impact of cognitive analytic therapy. Br J Med Psychol 74(pt 1):47-55.
Zanarini MC, Gunderson JG, Frankenburg FR, Chauncey DL (1990), Discriminating borderline personality from other axis II disorders. Am J Psychiatry 147(2):161-167.