Bipolar Disorder and “Borderline Personality Disorder”

What’s the difference?

(revised 11/2014)

First, my plain-er english conclusion

If you’re not a psychiatrist interested in this subject, this page is likely to be harder to read than most of the rest of this website. My apologies. This essay began several years ago as a presentation of my view on this subject, to save me time in explaining it. I was working on the inpatient unit of our hospital at the time, and this subject comes up every day there.

However, since that time I’ve used this page as a place to keep track of interesting references on the subject. The result is that there are too many article links to read very smoothly anymore. If you just skim along, at first, you may see my main point: trying to distinguish these two conditions is difficult because they share so many characteristics; and besides, the treatments to be considered are so similar, in the end it may be easier to give up trying to distinguish between bipolar disorder and borderline personality disorder. They may not be entirely separate conditions anyway, but even if they were, if for now the treatments are basically the same, why struggle so much with the naming?

In the material below, you’ll see the focus is on “how are they treated?” because presumably that is what diagnosis is supposed to be guiding. Although generally very similar, there are two important differences in treatment. In bipolar disorder, one uses antidepressants only with caution, whereas they are routinely used in the treatment of “borderline personality disorder.” Secondly, borderline treatment relies primarily on psychotherapy with medications coming second. Bipolar disorder is pretty clearly handled the other way around, with bipolar-specific psychotherapies coming second to medications (in general, no one advocates treating an average patient with bipolar disorder with psychotherapy alone).

Yet the psychotherapy for borderline is not going to make bipolar worse, rather quite the opposite in most cases. And the medications for bipolar disorder frequently make borderline better (including, in my experience, not just mood stabilizers, but also avoiding reliance on antidepressants; relying instead on mood stabilizers; in fact deliberately avoiding antidepressants at least for a time to see if that makes things turn out better). So I arrive at the conclusion that trying to distinguish these conditions creates more trouble than it solves. Have a look.

[Update 2014: the excellent research group led by Gin Malhi in Australia has looked at this issue and despite a detailed look Coulston still finds the whole issue controversial. Malhi]

Research on the overlap

[Apologies, 9/2009: I’ve not updated this material in several years. Many more articles than listed here have since appeared. But in general, my viewpoint has not altered despite a growing sense — I’ll admit — that there really is a valid distinction between these two conditions. I still think that for now, what really matters is “how are you going to treat this person”, and that has not changed much in the last 5 years. Psychotherapy is very important, particularly for “borderline” (I still hate that term, too). More therapies now have good evidence for their effectiveness.]

You may have heard about or even been told you are “borderline”. This used to be a very negative label, but now there’s good evidence that specific treatment is possible and can help a lot.Linehan et al, Blum So now the label is not as pessimistic and hopeless as it was even a few years ago.

A recent review from Canada found substantial overlap between borderline and bipolar disorder diagnoses, though still concluded that borderline deserved to be regarded as a valid diagnosis, separate from bipolar disorder.Magill This is an excellent review of this “overlap or one illness” issue, and I rather agree with the conclusion. Another borderline research group, led by Dr. Gunderson who’s been studying this phenomenon for decades, also warns against lumping the two conditions together.Gunderson He emphasizes the same differentiating factor I present in the next section: abandonment fear is prominent in borderline, not so in bipolar.

Updating references in 2014 I’m struck that in over 10 years, there as been little change in this “borderline vs. bipolar discussion.” You can still find both views.

  1. “they’re different, they need different treatments” ( this from the folks who think bipolar disorder is being overdiagnosed and that many of those patients actually have bipolar disorder).
  2. They’re so similar, there’s really only one thing that marks borderline, that’s abandonment fear and chronic emptiness — and the life consequences that follow, like chaotic relationships. All the other symptoms overlap both conditions. And besides, it’s really about treatment, and both need good therapy (as in DBT for borderline; and maybe bipolar alsoVan Dijk ).

Interestingly, when a group of patients with a diagnosis of bipolar disorder was studied for “maladaptive personality traits”, a substantial decrease was found when the patients’ mood disorder was treated.Peselow Kind of makes you wonder if the same might be true for patients diagnosed as “borderline”?

[Update August 2003: here’s a stunning research finding that in my opinion says a great deal about the nature of “borderline”. A team from the National Institutes of Mental Health, including Dr. Bob Post, who has published a lot about bipolar disorder, studied a woman with “borderline personality disorder” for nearly a year in their research hospital. They found that when this woman had depression, and especially when she had psychotic symptoms (e.g. becoming paranoid, hearing voices), she had increased levels of an antibody to thyroid tissue in her bloodstream.Geracioti Huh?

What is the connection here? Well, we know that one type of thyroid disease is associated with these “autoantibodies” — antibodies directed toward one’s own tissues, in this case thyroid tissue. That is called Hashimoto’s thyroiditis. There is some connection between thyroid problems and mood problems, that’s clear, but the nature of the connection is not understood. So, it’s a mystery as to why this woman’s thyroid antibodies would vary along with her mood and other symptoms. Is the thyroid change causing the mood change? Or is it the other way around? Or is some third problem causing both at the same time? Just keep watching for more information on how thyroid, which is similarly mysteriously involved in bipolar disorder, affects complex mood conditions.]

Overlap between diagnostic criteria

There is tremendous overlap in the symptom patterns of bipolar and borderline personality disorder (Borderline PD).  Notice in the table below that almost every symptom is found in each column:

Borderline PD (DSM list) Bipolar (broad view)
unstable self
transient paranoid ideation
chronic emptiness
abandonment fearEnergy
(sex, substances, self-harm)Mood
affective instability
reactive mood
episodic dysphoria
irritability, intense anger
suicide attempts (~10%)
unstable self
psychosis, esp. paranoid/grandiose
(spending, sex, substances, risk sports)Mood
affective instability
“rejection hypersensitivity”
irritability, intense anger
suicide attempts (~10%)

Some doctors believe self-harm is “diagnostic” of borderline PD, or worse yet, synonymous: all borderlines cut, and anybody who cuts is borderline. It’s hard, let me  tell you, to convince them it isn’t that simple. Here’s a case history of cutting in a patient who had bipolar disorder. I’ve seen several myself. Read one of the very helpful websites on self-harm if you need more information on this common behavior.

However, I think there are indeed two symptom that differentiate the bipolar and borderline (to the extent there’s any point in doing so; more on that in a moment). You noticed the big yellow blank in the table, yes? People with profound fear of abandonment, and a feeling of chronic emptiness, have a different struggle in life from those who don’t have these problems. Interestingly, these two symptoms also distinguish PTSD from borderline (only found in the borderline group, not very much in PTSD).Cloitre

By contrast, “cutting” is not specific to borderline personality disorder.Pope  In my practice, I’ve had patients who do not have chronic emptiness or abandonment fears who do have self-harm behaviors, which show up when they are extremely agitated. These patients all seem to have figured out that cutting or other forms of self-harm (I had one patient who hit herself on the head with a rolling pin) somehow helped them cope with the intensity of the rest of their symptoms. I have no doubt now that self-harm behaviors are an attempt to “treat” the severe agitation somehow; but that behavior seems rather quickly to disappear when the agitation is controlled.


There are good treatments for both conditions. Borderline PD is usually treated with psychotherapy as the main tool, with medications as needed or to the extent that they are helpful. Bipolar disorder is treated just the other way around: start with medications as the core ingredients in treatment, but using psychotherapy wherever it might be helpful.


Update 2014:  I used to have a bunch of individual studies listed here but recently there have been several massive reviews (“meta-analyses”), so those are now more reliable guides.Stoffers, Ingenhofen And they all seem to say the same thing:   antidepressants, mood stabilizers, and antipsychotics all work somewhat and none work really well, indeed not very well at all. Some medications work better on some symptoms than others: anger; impulsivity; anxiety; depression) . None is consistently the best across those various symptoms.

One recent (Cochrane) reviewStoffers concludes:

The available evidence indicates some beneficial effects with second-generation antipsychotics, mood stabilisers, and dietary supplementation by omega-3 fatty acids. However, these are mostly based on single study effect estimates. Antidepressants are not widely supported for BPD treatment…

Got that? Antidepressants are not the best class of medications for borderline personality disorder. None work terribly well. But evidence for benefit is greater for mood stabilizers than antidepressants.

So you can see why I emphasize: “what difference does it make what diagnosis you arrive at, borderline or bipolar?” As far as medications go, avoid antidepressants — either because the aren’t likely to work (borderline) or because they can make things worse (bipolar).  Instead, if there’s suspicion of borderline features (emptiness and abandonment fear, and the relationship chaos that results) then include DBT ( see below). If there are strong features of bipolar disorder (mood/energy cycling, clock disturbances, hints of hypomanic phases) then include bipolar education and regular sleep patterns.


The best studied technique for borderline personality disorder is “dialectic behavior therapy”, designed and studied initially by Dr. Marsha Linehan. Here is a brief and moderately technical overview; and a remarkable site that describes DBT from the user’s perspective, with lots of useful details including how to make your own flash cards, and links to Linehan’s lectures (some have even been transcribed). This technique is distinguished from the approaches which proceeded it in (at least) three ways:

  • It has randomized-trial evidence for effectiveness;
  • It has an understandable, logical, research-oriented rationale behind it.
  • It focuses on behaviors, and may not be sufficient treatment for feelings.

A group in Europe used the Linehan treatment approach, and got the same results as in Dr. Linehan’s original researchLinehan et al, namely a dramatic decrease in suicidal and self-harm behaviors, although they point out that the big improvements came for the patients with the most severe symptoms.Verheul They suggest that the Linehan DBT approach may be best suited for patient with severe self-harm and suicidal behavior, and that other therapies might be more appropriate for patients without these behaviors — because DBT does not seem to affect mood symptoms very much.
Thus, there may be even more reason, supported by the Verheul study, to think about medications for mood, as well as psychotherapies for mood (after DBT for self-harm and suicidality, if present). \=

Update 9/2009: other therapies which have emerged as effective for borderline in randomized clnical trials (RCTd):


 Does diagnosis affect treatment?

Imagine there really is a difference between these two conditions. We don’t really know that now. But imagine there is one, some difference in the structure of the limbic system, the emotion system of the brain, perhaps. And imagine that we had some great lab test that could tell the two apart perfectly (that’s rarely the case even with an excellent lab test, by the way, so don’t hold your breath). Now, suppose you really have “borderline”, but you get called
“bipolar “. What happens? You get treated primarily with medications. These might help, as those studies I mentioned a moment ago indicate. But you would still need some help with feelings of abandonment and emptiness (and the problems with relationships that come up when you have those feelings). Unless you had a really rigid psychiatrist or mental health system, you could then try to get a psychotherapy to address these.

What about the other way around? Suppose by our magic test you “really” have bipolar, but get diagnosed borderline PD. Well, until recently, this was the big problem. You’d get labeled as “personality disorder” and often your medical care, from primary care as well as mental health providers, would change accordingly. You’d get shunted to the bottom of the list of patients someone might want to take into a practice. You’d be told that your situation was basically unchangeable except with years of psychotherapy, and then discharged from the hospital no matter what your symptoms were, as was one of my patients only several months ago. (Even if you “really” have borderline PD you shouldn’t be treated this way anymore: if our mental health system were perfect (right), you would be found in some deliberate screening program to have borderline PD and placed in a treatment program designed for your condition. Hmm, sounds like how they treat diabetes, doesn’t it?).

But back to our example: you’re “really” bipolar, but you get diagnosed “borderline”. Even if you were initially treated with psychotherapy, ideally it would be noticed that you were not improving fully, and might need medication treatment as well. And hopefully, since mood instability is your primary problem (not plain depression, nor psychosis), you would then be treated with mood stabilizers.

The point here is that you would not necessarily be lead into a treatment that can harm you with the “wrong” diagnosis, either way. You might well get a treatment that could be helpful, even if it is not the “core treatment” you will eventually need (and hopefully get).

However, diagnoses also help predict how things are likely to go in the future — “prognosis”. For borderline personality disorder, the prognosis is not as good: responding to a medication so well that symptoms disappear completely, which is frequent in bipolar disorder (at least one person in 4 gets that kind of response; and only about 1 in 5 shows no improvement, with the rest of the group in the middle somewhere). So one of the risks in labeling a person bipolar, who “really” is borderline (as if we knew that there really is a “real” difference at some biological level, which we don’t), is that it might raise too much hope of a good outcome.

Worse yet, people with borderline-ness are prone to idealizing their doctors and therapists. So raising hope (for example, by emphasizing “bipolar” and the possibility of a great response) plays right into that idealizing, with increased potential for great disappointment later when treatment does not work as hoped. That great disappointment can likewise be magnified by people with borderline-ness, sort of a mirror image of the initial idealization and hopefulness. This overidealization and later strong devaluation are well-recognized phenomena in borderline-ness. Since suicide attempts are also very common in people who meet the DSM diagnostic rules for “borderline personality disorder”, you can see the problem here: raising hopes that get dashed later can lead to an intense hopelessness to which the person is already prone, and hopelessness is a well-known risk factor for suicide.

Therefore may I caution: assuming that a person is not “really” borderline, and is instead “really” bipolar, can add risk. We should all watch out for that. At the same time, assuming that a patient is “really” borderline when he/she might have bipolar disorder also clearly adds risk: it lowers the hope that an excellent treatment response is possible, which is a very terrible thing for a doctor to do to a patient and her/his family. It also lowers the determination to keep trying for a better outcome (“she’s a borderline” has historically been a reason not to admit people to hospital, or discharge them more quickly than others, for example). Finally, missing bipolarity because of a diagnosis of “borderline” may lead to an emphasis on using antidepressant medications, which are not thought to make borderline worse; indeed, they can help quite a bit — but they can definitely make bipolar disorder worse, as you’ve seen or will see discussed repeatedly elsewhere on this website.

Bottom line: one cannot simply assume that everything which looks like borderline personality disorder “really” is bipolar disorder. In particular, one must be careful not to raise hopes of possible symptom-free life if there is some diagnostic uncertainty about bipolarity versus borderline-ness. At the same time, lowering expectations and treatment effort because a person appears to have borderline traits is a similar pitfall on the opposite side of a presumed best “middle ground” approach.

As you can see, overall my recommendation is that you avoid getting too stuck on a diagnostic label. Which one you get depends a great deal on the orientation of the therapist or doctor! Psychiatrists might be better diagnosticians, in theory, because the they have pliers as well as hammers (“when all you have is a hammer, everything looks like a nail”). But finding a psychiatrist who really does use her/his pliers just as much as his/her hammer can be difficult. Finding a psychiatrist at all can be difficult. In that case, a therapist who feels comfortable treating borderline personality disorder is a good starting place; you can use websites like this one to learn more about the two diagnoses and help guide your treatment from there.

Finally, if you’re wondering about what causes borderline personality disorder in the first place, here’s a brief essay introducing some new research in that area, about how attachment to parents, particularly moms, is mediated by particular brain chemicals.