Gender Patterns in Borderline Personality Disorder
and Abstract
Gender differences in patients with borderline personality disorder are potentially relevant because they may guide clinicians in assessment and treatment. To date, a number of clinical features in borderline personality disorder have been examined for gender differences. As for prevalence, earlier research concluded that a higher proportion of women than men suffer from borderline personality disorder, although more recent research has determined no differences in prevalence by gender. In addition, there may not be gender differences in borderline personality disorder with respect to specific types of self-harm behavior, such as self-cutting or levels of psychological distress at clinical presentation. However, current evidence indicates that there are notable gender differences in borderline personality disorder with regard to personality traits, Axis I and II comorbidity, and treatment utilization. With regard to personality traits, men with borderline personality disorder are more likely to demonstrate an explosive temperament and higher levels of novelty seeking than women with borderline personality disorder. As for Axis I comorbidity, men with borderline personality disorder are more likely to evidence substance use disorders whereas women with borderline personality disorder are more likely to evidence eating, mood, anxiety, and posttraumatic stress disorders. With regard to Axis II comorbidity, men with borderline personality disorder are more likely than women to evidence antisocial personality disorder. Finally, in terms of treatment utilization, men with borderline personality disorder are more likely to have treatment histories relating to substance abuse whereas women are more likely to have treatment histories characterized by more pharmacotherapy and psychotherapy.
This ongoing column is dedicated to the challenging clinical interface between psychiatry and primary care—two fields that are inexorably linked.
Introduction
Are there any meaningful clinical differences between men and women who suffer from borderline personality disorder (BPD)? In this edition of The Interface, we discuss the available evidence regarding gender differences in BPD. Specifically, we will review the data for possible gender differences in BPD with regard to prevalence, personality traits, comorbid Axis I disorders, comorbid Axis II disorders, specific self-harm behaviors, level of distress at clinical presentation, and treatment histories. Some of the preceding areas demonstrate distinct gender differences whereas others do not.
Is Borderline Personality More Common in Women?
Since 1980, the year of the initial standardization of the criteria for personality disorders, all versions of the Diagnostic and Statistical Manual of Mental Disorders have indicated that BPD is unequivocally more common in women than men. For example, according to the most recent edition, the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR),1 there is a 3:1 female to male gender ratio. Yet, a recent and well-executed study by Grant et al2 (i.e., the National Epidemiologic Survey on Alcohol and Related Conditions) found that BPD is equally prevalent among men and women, which is our current impression.
What might explain the prior conclusion of a divergence in gender proportions? First, in studies dating back 25 years, a number of investigators have confirmed that clinicians have a subtle female gender bias with regard to the diagnosis of BPD.3–5 Thus, clinician bias in diagnosis may have partially contributed to the misperception of a female predominance. Not unexpectedly, other researchers have reported no evidence of gender bias in BPD diagnosis by clinicians.6,7 In support of this latter impression, in examining individual criterion for BPD, Davis8 found that no particular one was consistently rated by clinicians as more common in women versus men.
Another possibility for the appearance of gender disproportion is sampling bias.9,10 To clarify this point, the traditional settings for prevalence studies (i.e., psychiatric settings) may not reflect the true gender distribution of BPD. For example, if women with BPD undertake more self-directed self-harm behavior and, therefore, wind up in mental health settings more often than men, then the conclusion of studies in these settings would be that more women than men suffer from BPD. Likewise, if more men than women suffer from substance abuse and antisocial features, and wind up in related treatment programs or jail, then again, studies in mental health settings will invariably under-report the percentages of men with BPD. We will continue to illustrate the issue of sampling bias throughout this paper.
In addition to the possibilities of clinician bias and sampling bias, uneven expression in BPD according to gender could also be explained by cultural underpinnings,10 sex differences in parenting,11 and/or gender differences in normal behavior.11 Of course, these latter explanations assume that there is a genuine gender difference in prevalence in BPD.
To summarize, from a clinical perspective, and based upon the findings of the most recent study by Grant et al,2 we strongly suspect that the male/female distribution of BPD is fairly equal. However, we also believe that the genders tend to exhibit slightly different behaviors and presentations, which could easily culminate in different clinical dispositions and result in a sampling bias, as noted previously.
Do the Genders Evidence Differences with Regard to Personality Traits?
Using Cloninger’s psychobiological model, Barnow et al12 examined according to gender the general psychological characteristics of a large inpatient sample of individuals with BPD (n=202), and compared these with several different types of control groups. The researchers found that men with BPD were characterized by explosive temperaments as well as high levels of novelty seeking and harm avoidance whereas women with BPD were characterized by high levels of harm avoidance but not novelty seeking. Thus, explosive elements with novelty seeking appear to differentiate men from women with BPD.
Do the Genders Evidence Any Differences in Axis I Comorbidity?
Several studies have examined gender differences in BPD in relationship to comorbid Axis I diagnoses (Table 1).2,13–16 According to the findings in Table 1, substance abuse is consistently more common in men with BPD, with one study also indicating a greater prevalence of intermittent explosive disorder. In contrast, eating disorders are consistently more common in women with BPD as well as a greater likelihood of mood, anxiety, and/or posttraumatic stress disorders. Given these specific differences in Axis I comorbidity, it is understandable how women might wind up in mental health settings for treatment, thereby contributing to a potential sampling bias, whereas men might be more likely to wind up in prison settings and remain under-represented in mental health settings.
Table 1
Gender differences in the prevalence of comorbid Axis I disorders in individuals with borderline personality disorder
| FIRST AUTHOR (YEAR) | AXIS I DISORDERS WITH A GREATER PREVALENCE IN MEN |
AXIS I DISORDERS WITH A GREATER PREVALENCE IN WOMEN |
|---|---|---|
| Tadic (2009)16 | Any substance use disorder | Any mood disorder Any anxiety disorder Any eating disorder |
| Grant (2008)2* | Any substance use disorder | Any mood disorder Any anxiety disorder Posttraumatic stress disorder |
| Johnson (2003)15 | Any substance disorder | Any eating disorder Posttraumatic stress disorder |
| Zlotnick (2002)14 | Any substance abuse disorder Intermittent explosive disorder |
Any eating disorder |
| Zanarini (1998)17 | Any substance abuse/dependence | Any eating disorder Posttraumatic stress disorder |
Do the Genders Evidence Any Differences in Axis II Comorbidity?
Zanarini et al17 examined Axis II comorbidity in patients with BPD according to gender in 1998. According to their findings, the rates of avoidant and dependent personality disorders were similar for both genders. However, men with BPD were significantly more likely to have comorbid paranoid, passive-aggressive, narcissistic, sadistic, and antisocial personality disorders. In support of these data, in the previously noted study by Tadic et al,16 researchers also found a higher frequency of antisocial personality disorder in men compared with women (57% vs. 26%). Therefore, men with BPD appear to be characterized by antisocial overtones. Again, given the overload of antisocial features in the psychological styling of men with BPD, disposition in the correctional system would be more likely.
Do the Genders Evidence Any Differences in Self-Harm Behavior?
We have already partially addressed this query in our analysis of Axis I disorders, in which we found that men with BPD were more prone to substance use difficulties and women to eating disorders. However, in a British study,18 investigators examined the behavior of skin-cutting according to gender. In this study, there were no differences in gender, indicating that at least one type of self-harm behavior emerges in equal proportions in the sexes.
Do the Genders Evidence Different Levels of Distress at Clinical Presentation?
In a study of gender differences in BPD, Zlotnick et al19 examined and detected among 130 outpatients no male/female differences in the degree of overall impairment at clinical presentation. Although there were gender differences with regard to impulse patterns (again, men evidenced more substance abuse, antisocial features, and intermittent explosive disorder, and women evidenced more eating disorders), both genders presented for treatment with equal levels of emotional distress.
Do the Genders Evidence Any Differences with Regard to Treatment-Utilization Patterns?
According to the findings of Goodman et al,20 there appear to be distinct gender differences with regard to treatment utilization. While the lifetime levels of mental healthcare utilization are high for both genders, men with BPD are more likely than women to utilize drug/alcohol rehabilitation services but less likely to utilize pharmacotherapy and psychotherapy services. Again, this suggests a risk of sampling bias with regard to prevalence studies.
Conclusions
While there may not be gender differences in BPD with regard to prevalence, some specific self-harm behaviors (e.g., self-cutting), and presenting levels of overall psychological distress, there appear to be notable gender differences with regard to personality traits, Axis I and II comorbidity, and treatment utilization histories. With regard to these differences, men with BPD are more likely to demonstrate explosive temperaments coupled with high levels of novelty seeking. Men with BPD are also more likely to evidence substance abuse whereas women with BPD are more likely to evidence eating, mood, anxiety, and posttraumatic stress disorders. With regard to Axis II comorbidity, men with BPD are more likely than women to have antisocial personality characteristics. Finally, men with BPD are more likely to have treatment histories for substance abuse whereas women with BPD are likely to have utilized more pharmacotherapy and psychotherapy services. Thus, given the similarities in men and women with BPD, there are also clear gender differences in BPD as well. These differences, reported in a number of different studies, may explain why women with BPD are more likely to be over-represented in mental health services and men with BPD are more likely to be over-represented in substance-abuse treatment programs and/or jails. An awareness of these clinical differences is particularly important in evaluating patients in psychiatric or primary care settings, as men and women with BPD appear to have slightly different clinical presentations and treatment histories.
References