Skip to main content

Verified by Psychology Today


There May Be 3 Types of Borderline Personality Disorder

Recognizing different presentations could improve outcomes.

Key points

  • Borderline personality disorder (BPD) is a relatively common problem, often difficult to treat.
  • Because there are hundreds of possible symptom combinations, a system for classifying BPD is desirable.
  • New research using "latent class analysis" identifies 3 distinct types of BPD.
  • This information is useful for clinical treatment, self-assessment, and research and development.

Borderline personality disorder (BPD) is one of several personality disorders described in the Diagnostic and Statistical Manual (DSM-5) by the American Psychiatric Association. BPD belongs to a group called “Cluster B” personality disorders, alongside antisocial, narcissistic, and histrionic personality disorders.

These disorders are grouped together because they share common features of emotional dysregulation, alterations in sense of self, and a proclivity for erratic and dramatic cognition, behavior, and long-standing patterns of interpersonal relationship problems associated with insecure, often disorganized, attachment style. They are classified as “disorders” if and only if they are associated with chronic, clinically-significant distress or dysfunction.

The term “borderline” relating to personality has its origins in psychoanalysis (Stern, 1938), not as a diagnosis per se but as a description of personality organization—in the borderland between higher-functioning neurosis and more profound psychosis. People with BPD tend to have difficulty reflecting on their own thoughts and behaviors. This difficulty with metacognition renders self-regulation more difficult and limits the therapeutic effectiveness of insight (Martin & Del-Monte, 2023).

While BPD is at times romanticized, even valorized, in the media because of the excitement that goes along with the heady, often exhilarating, ups, the downside of the disorder is too often tragedy and isolation. Folks with BPD are at significantly increased risk for suicide, significant physical and mental health difficulties, relationship and professional problems, and a slew of other adverse outcomes.

What Is Borderline Personality Disorder?

The diagnosis of BPD is made when an individual has at least five of nine characteristics1 present chronically and consistently in work, personal life, or with oneself—from emotional instability in reaction to ordinary events and chronic feelings of emptiness to frantic efforts to avoid real or imagined abandonment and transient paranoid ideation.

BPD appears to be caused by a combination of genetic and environmental factors, often but not exclusively overlapping with post-traumatic stress symptoms related to developmental adversity. BPD often co-occurs with other problems including major depression, anxiety disorders, PTSD, substance and alcohol use disorders, and others.

Because there are hundreds of possible combinations of the symptoms that would meet the criteria for BPD diagnosis, identifying clinically-significant subtypes would represent a huge advance, offering great potential clinical benefit in personalizing care, furthering what's known as "precision medicine."

Latent Class Analysis of Borderline PD

While there have been efforts to investigate BPD subtypes, there has been no consensus. Antoine and colleagues (2023), in the journal Psychiatry Research, report that prior work has looked at symptom-based subtypes, distinguished BPD from other disorders including ADHD and Bipolar Disorder, or grouped BPD according to severity, but not yet identified BPD types.

To that end, over the course of three studies, they used data from over 500 people with BPD in treatment in various specialty clinical settings (for example, a hospital-based mental health clinic), and applied latent class analysis (LCA) to determine whether statistically-significant subgroups were present and, if so, what distinguished them from one another. Measures used in the study included the BPD section of the International Personality Disorder Examination, the Childhood Trauma Questionnaire, the Beck Depression Inventory, and the Symptom Checklist 90.

They tested several models to see how many groups, or classes, of symptom clusters best fit the data in this sample of 504 patients. The modeling found that a three-factor solution had the greatest validity: Non-Labile (more stable), Dissociative/Paranoid, and Interpersonal Unstable classes.

  • Non-Labile (10.5 percent). People in this group showed less emotional dysregulation and were less likely to dissociate when distressed, with a tendency to impulsivity that did not quite reach statistical significance. This group had lower associations with childhood adversity, including sexual abuse.
  • Dissociative/Paranoid (55.4 percent). Those in this group had substantially higher levels of dissociative and paranoid symptoms but had a more stable sense of self and were less likely to try and do anything to avoid abandonment. This class was associated with an increased risk of comorbidity, that is, having additional diagnoses over the course of a lifetime. There were higher rates of childhood abuse in this group compared with the Non-Labile class, including a higher correlation with childhood sexual abuse.
  • Interpersonally Unstable (34.1 percent). This group had the highest level of frantic efforts to avoid abandonment, with increased symptoms of anger and aggression and relationship instability. There were higher rates of childhood abuse in this group compared with the Non-Labile class, including a higher correlation with childhood sexual abuse, albeit not as strong as among the Dissociative/Paranoid group.


The research, while preliminary, is notable for people with BPD, those close to them, and clinicians. There are implications for specific treatment choices, including psychotherapeutic approaches, medication, and adjunctive approaches such as meditation, lifestyle/context modification, and self-regulatory practices.

Those in the Non-Labile group have different environmental influences, with lower rates of childhood trauma. They may benefit from a focus on reducing impulsive actions through building executive function. Awareness of this smaller, less stereotypical subset is important as the treatment approach may be very different.

The Dissociative/Paranoid group has the highest rates of childhood trauma, including sexual abuse. There may be a higher association with complex PTSD, requiring clinical focus. Becoming and remaining aware of potential threats without being overly mistrustful or needing to emotionally disconnect is key for them to make effective choices about relationships.

With relatively low abandonment-avoidant (classic BPD "frantic efforts to avoid abandonment") behaviors, like excessive texting, acting "needy," and being unable to tolerate rejection or being apart, people in this class may have learned instead to shut down when threatened by separation, a form of "learned helplessness" that predisposes to certain forms of trauma and depression. This tuning-out dissociative reaction reduces perceived distress but may increase vulnerability to perpetrators by rendering the individual numb to others' efforts to manipulate and abuse.

Restoring awareness, emotional and cognitive, reduces revictimization and retraumatization by allowing us to recognize "red flags" in others, rather than getting attached to someone unlikely to bring health, happiness, and healing.

The Interpersonally Unstable group was similar to the Dissociative/Paranoid group in some ways, with higher rates of trauma and abuse, but the opposite regarding frantic efforts to avoid abandonment and a classic pattern.2 They display splitting, or either/or thinking, a tendency to flip from seeing others as all good or all bad (“idealization and devaluation”), and greater hostility and aggression, leading to relationship problems.

Future work will help establish whether the findings are more broadly applicable and whether different treatment approaches and adjunctive work, such as various forms of meditation or environmental modification, are more or less effective as a function of BPD type. In the meantime, the research is relevant and intriguing as it stands, and may allow clinicians to think more clearly about BPD and identify treatment focus, and help BPD patients live most fully by embracing challenges and opportunities with compassion for self and others.

Facebook image: J Walters/Shutterstock

LinkedIn image: CarlosBarquero/Shutterstock


1. Borderline PD Diagnostic Criteria: 1) fears of abandonment, 2) unstable relationships, 3) unstable self-image and identity, 4) transient dissociative, paranoid or psychotic thinking only when distressed, 5) anger and mood-regulation problems, 6) persistent depression-like feelings of sadness and inner emptiness, 7) a proclivity for self-harm or suicidal thinking and behavior, *) mood instability or dysregulation (emotional “lability”, and 9) a tendency toward reckless or impulsive behavior in areas like spending, sexuality, driving, and other areas.

2. This group is the most similar in many ways to classic clinical ideas of BPD, and most matches media depictions. Loving hard, getting stuck in “bad relationships”, raging and suffering the consequences, often with limited capacity not only to self-regulate and make better choices, but also limited ability to reflect on one’s own contributions without feeling blames, rather than being able to step back and use feedback for growth. Trauma-focused therapy is likely to be helpful, along with grounding in mindfulness and cultivation of self-compassion.

Stern, A. (1938) Psychoanalytic investigation of and therapy in the borderline group
of neuroses. Psychoanalysis Quarterly, 7, 467–489.

Sylvia Martin, Jonathan Del-Monte, Metacognition and insight dynamics exploration in borderline personality disorder: Exploring the underlying dynamics, Journal of Psychiatric Research, Volume 160, 2023, Pages 225-231, ISSN 0022-3956, ttps://

Silvia M. Antoine, Beverley K. Fredborg, David Streiner, Tim Guimond, Katherine L. Dixon-Gordon, Alexander L. Chapman, Janice Kuo, Paul Links, Shelley McMain, Subgroups of Borderline Personality Disorder: A Latent Class Analysis, Psychiatry Research, 2023, 115131,ISSN 0165-1781,

Note: An ExperiMentations Blog Post ("Our Blog Post") is not intended to be a substitute for professional advice. We will not be liable for any loss or damage caused by your reliance on information obtained through Our Blog Post. Please seek the advice of professionals, as appropriate, regarding the evaluation of any specific information, opinion, advice, or other content. We are not responsible and will not be held liable for third party comments on Our Blog Post. Any user comment on Our Blog Post that in our sole discretion restricts or inhibits any other user from using or enjoying Our Blog Post is prohibited and may be reported to Sussex Publishers/Psychology Today. Grant H. Brenner. All rights reserved.

More from Grant Hilary Brenner MD, DFAPA
More from Psychology Today