Diagnosing and Treating Borderline Personality Disorder

Mark Zimmerman, MD, professor of psychiatry and human behavior, Brown University, Providence, Rhode Island, explores key takeaways from his live 2021 Psych Congress session where he discussed the diagnoses and treatment of borderline personality disorder (BPD).

In part 2, Dr Zimmerman gives the key takeaways from his session titled "Not Too Much or Too Little: Facing the Challenges of Pharmacotherapy in Borderline Personality Disorder" that explored the pharmacology of BPD. 

Read the transcript:

Hello, everyone. I am Mark Zimmerman, a psychiatrist in the Department of Psychiatry and Human Behavior at Brown University, Director of the Partial Hospital Program and the Outpatient Program at Rhode Island Hospital. I've just discussed the treatment of borderline personality disorder, focusing predominantly on psychopharmacology.
Before talking more about treatment, I think it's necessary to first discuss diagnosis, because if you don't recognize and diagnose borderline personality disorder, it certainly makes it more difficult to treat it. Before you treat, you need to diagnose.

Let's first talk about prevalence. The largest study using semi-structured interviews to assess borderline personality disorder in clinical settings has found that in outpatient settings, about 10% of individuals are diagnosed with borderline personality disorder. In partial hospital settings, it's closer to 20%, around 18%. In inpatient settings, it's also about 20%.

Despite its relatively high frequency, it's frequently underdiagnosed, as several studies have found. The question is why. The probable answer to that is related to the high comorbidity associated with borderline personality disorder. On average, patients with borderline personality disorder have 2 to 3 additional diagnoses, in outpatient settings and in partial hospital settings.

We recently published a study in which we found 70-80% of individuals with borderline personality disorder had 3 or more diagnoses, the most frequent of which was major depressive disorder and also post traumatic stress disorder.

When individuals present for treatment, they typically don't present for features of borderline personality disorder. Patients don't come into your practice and say, "I'm here because I'm feeling empty," or "I'm here because I don't have a good sense of myself."

Instead, they're presenting for complaints related to depression and anxiety. That contributes to the underrecognition and underdiagnosis.

Another reason for underdiagnosis is what I call the hidden borderline. By that, I mean that there are individuals who present for treatment without evidence of self injury or suicidality but nonetheless still meet the criteria for borderline personality disorder.

In fact, a patient that I saw a few months ago came in saying that her therapist told her that she does not have borderline personality disorder because she's never attempted suicide and she's never self-injured.

It's important for clinicians to realize that the prevalence of that diagnostic criterion for borderline personality disorder, self injury or suicidality, is only present in about 50-60%. This has been shown in several studies.

Just because a person isn't cutting themselves or burning themselves or engaging in other self-harm behavior does not mean they may not have borderline personality disorder.

The most common screening instrument for borderline personality disorder is the McLean Screening Instrument, the MSI, for Borderline Personality Disorder, MSI BPD. We'll talk a little bit more about that in just a second.

When a patient comes in for treatment, let's say for depression, it's typical to screen for a number of different disorders by asking about the necessary feature, that is, if you will, the gate criterion to making that diagnosis.

Whether that be the presence of panic attacks to diagnose panic disorder or the presence of drinking in someone who may have an alcohol use disorder or gambling behavior in someone who may have a gambling disorder, for many of the disorders that we diagnose, there is a necessary feature that you need to establish as being present in order to make the diagnosis.

However, borderline personality disorder is similar to some other disorders diagnosed based on a polythetic set of criteria, meaning there's a list of features. A minimum number need to be present, but none of them need to be present. You need 5 of 9 criteria to make the diagnosis, but no particular criterion needs to be present. As a result, there is no gate criterion.

There have been a number of studies that have looked at the psychometric properties of the borderline personality disorder criteria, looking at to what degree the criteria set hangs together.

Our clinical research group looked at this from a different perspective. We were interested in whether or not there was a criterion that was sufficiently frequent that it could be used as a screening criterion. In other words, the sensitivity of this criterion was so high that most individuals with the disorder would have this criterion.

In addition, to be a clinically useful screening criterion, you would want the criterion to have high negative predictive value, meaning if the criterion was absent, then you can be reassured that the person does not have the disorder.

What we found when analyzing the data in over 3,500 patients who were administered a semi-structured interview, the criterion of affective instability had a sensitivity of 93% and a negative predictive value of 99%.

When we went back and reviewed the literature, this was consistent with other studies looking at the psychometrics of the criteria, which found that affective instability had very high sensitivity and negative predictive value and can be used as a screening criterion.

When you do your review of systems, whether or not someone is coming in for the treatment of depression or panic disorder or post-traumatic stress disorder, assessing affective instability, persistent affective instability, not just in the context of recent symptomatology, would be a good way of screening for the disorder.

I mentioned the MSI BPD which is a self-administered questionnaire that can screen for the disorder. Rarely is it used in clinical practice, though there have been a number of studies that have examined the performance of the scale.

More important than that, the scale has been used as an indicator of BPD in 2 recent studies that have examined whether or not ECT is effective in depressed individuals who have BPD.

In both of these studies, the diagnosis of borderline personality disorder was made with this screening questionnaire. Both studies found that the response to ECT in depressed patients was similar in those who did and did not screen positive.

Both studies concluded that ECT should be considered and certainly should not be ruled out in depressed individuals who have borderline personality disorder. That may well be the case, but it is important to recognize that screening should not be confused with diagnosis. Let me elaborate on that.

Neither of these studies discussed another important statistic to calculate when examining the performance of a screening scale. That is positive predictive value. Positive predictive value indicates how many individuals who screen positive would receive the diagnosis when evaluated with the more thorough diagnostic procedure.

A recent review of the performance of the McLean Screening Instrument found that the sensitivity of the scale is about 80%. The specificity is about 65%.

When the positive predictive value was computed for the scale, in the most recent study of ECT response, it was found that about two-thirds of individuals who would screen positive on the scale would not be diagnosed with BPD if they were administered a semi-structured interview.

That's because the psychometric properties of the scale and the relatively low prevalence of borderline personality disorder result in most individuals, the majority of individuals who screen positive and get referred for ECT, not having the diagnosis.

This highlights the importance of making sure you understand the difference between screening and diagnostic procedures. You should not use a screening procedure to make a diagnosis.

Whether screening for BPD by assessing affective instability or screening by using a self report scale, you should not make the diagnosis based on that screening procedure. It just indicates whether you should go further in your evaluation to determine the presence or absence of the disorder.

[10:07] To summarize this first part, 4 takeaway points.

First, borderline personality disorder is common. If you're not diagnosing it in 10-20% of your patients and, of course, it depends on the setting, you're likely missing the diagnosis.

Second, related to this, underdiagnosis is a problem. It's not surprising. It's not surprising, because individuals come to you complaining of depression, complaining of panic, complaining of recent trauma and how they've reacted to that.

They're not coming to you saying, "My problem is I go from loving to hating someone and that's why I'm here," or "My problem is I have an identity disturbance," or "I feel chronically empty all the time." Therefore, not surprising why the diagnosis may get missed.

Third, if someone is not suicidal and is not self harmed, that doesn't mean they don't have borderline personality disorder. Only about 50-60% of patients with BPD display or report that criterion.

Last, as you screen for other pathology, after you establish the principal diagnosis, you should screen for BPD. That should be part of your review of systems. When someone presents for depression, you determine whether or not they also have chronic anxiety and are a chronic warrior that may be consistent with generalized anxiety disorder.

Whether they've had panic attacks or whether they have a history of trauma and may have post traumatic stress disorder in the same way you can assess affective instability.

Not just limited to the depressive episode, but persistent affective instability as they screen for borderline personality disorder.
When you screen for other disorders, you need to follow it up and inquire about the other features in order to establish the diagnosis. When you screen for BPD, you need to assess the other features of the disorder in order to determine and make the diagnosis.

Thank you for your attention and I hope you now move on to part 2 of this presentation on the psychopharmacology of borderline personality disorder.

Mark Zimmerman, MD, is a Professor of Psychiatry and Human Behavior at Brown University and director of the Partial Hospital Program and Outpatient Practice at Rhode Island Hospital. Dr Zimmerman is principal investigator of the Rhode Island Methods to Improve Diagnostic Assessment and Services (MIDAS) project. The MIDAS project has been ongoing for more than 25 years. The goal of the MIDAS project has been to integrate research methodology into routine clinical practice in order to improve clinical practice. 

Dr Zimmerman is the author of more than 450 articles published in peer-reviewed journals, and serves on the editorial board of 10 journals. He is the associate editor of the Journal of Personality Disorders. He has developed several measures of psychiatric disorders for use in clinical practice. He is the author of the Interview Guide to Diagnose DSM-5 Psychiatric Disorders and the Mental Status Examination.