Few diagnostic distinctions carry higher stakes than separating bipolar disorder from borderline personality disorder (BPD). Both present with conspicuous mood instability, impulsivity and self-harm, yet their first-line treatments diverge sharply. Bipolar disorder is managed primarily with mood stabilisers and other pharmacotherapy; BPD responds best to structured psychotherapies such as dialectical behaviour therapy (DBT). Mislabel a client and the consequences are not trivial: a person with BPD prescribed mood stabilisers may be deprived of the psychotherapy that would actually help, while a person with bipolar disorder steered only into psychotherapy may be left without the pharmacological stabilisation they need. Compounding the difficulty, the two conditions co-occur often enough that “either/or” thinking is frequently the wrong frame. This article sets out the evidence-based discriminators, the comorbidity data and the role of structured assessment in resolving one of psychiatry’s most persistent diagnostic dilemmas.
In clinical practice I’ve seen errors go both ways, where people have been diagnosed with bipolar and some years later realise that it is in fact borderline. And vice versa. My hope is that this article will help Give clinicians tools that will clearly discriminate the two.
| Feature | Bipolar disorder | Borderline personality disorder |
|---|---|---|
| Duration of mood states | Days to weeks (mania ≥1 week; hypomania ≥4 days; depression ≥2 weeks) | Hours, rarely more than a few days; often several shifts per day |
| Trigger | Relatively autonomous of circumstances | Reactive, typically to interpersonal events |
| Nature of elevated mood | Euphoric/expansive or irritably grandiose; goal-directed | Brief relief or anger/anxiety-driven activation |
| Sleep | During manic or hypomanic episodes, educed need for sleep, feels rested | Variable |
| Identity | Stable between episodes | Persistently unstable self-image |
| Abandonment fear | Not characteristic | Central; frantic efforts to avoid abandonment |
| Self-harm pattern | Clusters around mood episodes | Chronic, interpersonally triggered affect regulation |
| Family history | Strongly heritable (~60–85%) | Moderately heritable (~40–60%) |
| First-line treatment | Mood stabilisers and pharmacotherapy | Structured psychotherapy (e.g., DBT) |
| Relevant screener | Mood Disorder Questionnaire (MDQ) | MSI-BPD; BSL-23 |
The overlap is genuine, not merely superficial. Impulsivity, emotional dysregulation, suicidality, deliberate self-harm and substance misuse are shared features that draw clinicians toward either diagnosis (Bayes, Parker, & Paris, 2019). The confusion intensifies when bipolar II disorder is in the frame, because hypomanic episodes are by definition less severe and non-psychotic, making them harder to distinguish from the affective storms of BPD.
Bipolar disorder is among the most frequently misdiagnosed conditions in mental health practice. The condition routinely escapes timely identification: a landmark meta-analysis reported an average gap of approximately six years between bipolar symptom onset and correct diagnosis and management (Dagani et al., 2017).
Yet the surface similarity conceals a structural difference that is the single most useful discriminator: the temporal grain of the mood disturbance and its relationship to circumstance.
Questionnaires designed for bipolar tracking, like the PMQ-9 and PHQ-9, when graphed over time, can show the mood cycles clearly.
In bipolar disorder, mood episodes are relatively sustained and comparatively autonomous of immediate circumstances. A manic or hypomanic episode unfolds over days to weeks (DSM-5 requires at least one week of mania, or four days of hypomania), and a depressive episode persists for a fortnight or more. These states represent a categorical departure from the person’s baseline that is observable to others and often runs its course somewhat independently of what is happening in the person’s life.
In BPD, affective instability operates on a far shorter timescale and is tightly coupled to interpersonal events. The DSM-5 criterion describes “marked reactivity of mood (e.g., intense episodic dysphoria, irritability, or anxiety) usually lasting a few hours and only rarely more than a few days.” The mood shifts within hours, frequently several times a day, and is characteristically reactive: a perceived slight, rejection or abandonment can precipitate a rapid plunge.
Empirical work supports this distinction. In a comparison of self-reported affective instability, individuals with borderline traits described instability that was significantly more interpersonal in origin than that reported by individuals with bipolar traits (Reich, Zanarini, & Fitzmaurice, 2012). Henry and colleagues (2001) found that the direction of affective lability also differs: borderline patients showed greater lability between euthymia and anger, whereas bipolar II patients showed greater lability between euthymia and depression. In short, ask not only how much the mood moves but how fast, in which direction, and in response to what.
A second discriminator concerns the quality of elevated mood. The hallmark of a manic or hypomanic episode is euphoric or expansive (or irritably grandiose) mood accompanied by a reduced need for sleep without consequent fatigue, increased goal-directed activity, pressured speech and inflated self-esteem. Decreased need for sleep is a particularly specific marker; people in a manic state feel rested after little sleep, which is rarely true of the agitated insomnia seen in BPD.
The “highs” reported in BPD are usually different in kind. They tend to be brief reprieves of relief or transient elation tied to a positive interpersonal event, or states of anger- and anxiety-driven activation, rather than the autonomous, sleep-curtailed, expansive energy of hypomania. The MDQ’s subscale structure captures this: Positive Activation (increased energy, grandiosity, decreased need for sleep) is relatively specific to bipolar disorder, whereas Negative Activation (irritability, racing thoughts, distractibility) is elevated across many disorders of emotional dysregulation, including BPD (Carpenter, Stanton, Emery, & Zimmerman, 2020).
Three further domains tilt the balance toward BPD. Identity disturbance, a markedly and persistently unstable self-image, is a core BPD feature with no equivalent in bipolar disorder, where the sense of self is generally stable between episodes. Abandonment sensitivity, frantic efforts to avoid real or imagined abandonment, is similarly central to BPD and absent from the bipolar criteria. And while self-harm and suicidality occur in both, the pattern differs: in BPD, recurrent self-injurious behaviour is frequently a chronic, interpersonally triggered means of regulating unbearable affect, whereas in bipolar disorder suicidality clusters around discrete mood episodes.
Heritability provides corroborating, though not decisive, evidence. Bipolar disorder is among the most heritable psychiatric conditions, with twin-study estimates commonly in the region of 60 to 85 per cent (Johansson et al., 2019). BPD is also heritable but generally to a more moderate degree, with twin-study estimates for the diagnosis and its temperamental components frequently falling around 40 to 60 per cent (Skoglund et al., 2021). A clear family history of bipolar disorder modestly raises the prior probability of a bipolar diagnosis, but heritability alone cannot adjudicate an individual case, particularly given the documented genetic overlap between the two conditions.
This is where clinicians most often go wrong, because the relationship between the two disorders is not purely one of differential diagnosis: they genuinely co-occur, and they are genuinely confused for one another.
On co-occurrence, the most authoritative figures come from a systematic review and meta-analysis by Fornaro and colleagues (2016). Pooling across studies, they reported that approximately 21.6 per cent of people with bipolar disorder also met criteria for BPD, and conversely that approximately 18.5 per cent of people with BPD also met criteria for bipolar disorder. Comorbidity was notably higher in bipolar II samples, where rates of comorbid BPD reached around 37.7 per cent. In other words, roughly one in five clients carrying either diagnosis will qualify for both, and the comorbid presentation is associated with a more severe course and greater suicide risk.
On misdiagnosis, the work of Zimmerman and colleagues is foundational. In a sample of psychiatric outpatients, Zimmerman, Ruggero, Chelminski, and Young (2008) found that the majority of patients with a prior clinician diagnosis of bipolar disorder did not meet criteria for it on structured interview, and that overdiagnosis was a real phenomenon. Crucially, in the follow-up analysis, Ruggero, Zimmerman, Chelminski, and Young (2010) demonstrated that patients with BPD were significantly more likely to be overdiagnosed with bipolar disorder: those who had been incorrectly told they had bipolar disorder were roughly four times more likely to have BPD than those without a prior bipolar diagnosis. The authors argued that the cross-sectional emotional instability of BPD is readily mistaken for the episodic mood swings of bipolar disorder, particularly when clinicians rely on a loose, “mood swings equal bipolar” heuristic rather than carefully characterising episode duration and trigger.
The practical lesson is twofold. First, a substantial minority of clients legitimately have both conditions, so screening positive on a bipolar instrument does not exclude BPD, and vice versa. Second, the direction of error in routine practice tends to favour over-attribution of mood instability to bipolar disorder, making deliberate assessment of BPD features a necessary corrective.
Because the discriminating features are subtle and prone to clinician bias, structured measures add real value, used as adjuncts to a thorough clinical interview rather than as standalone arbiters.
For bipolar screening, the Mood Disorder Questionnaire (MDQ) is the most widely used self-report instrument, screening for a lifetime history of manic and hypomanic symptoms and requiring symptom clustering and functional impairment. Its Positive versus Negative Activation subscales are particularly useful in this differential, since high Negative Activation with minimal Positive Activation should prompt consideration of BPD, PTSD or ADHD rather than bipolar disorder. The MDQ sits within NovoPsych’s broader bipolar disorder assessment questionnaires collection.
For BPD, the McLean Screening Instrument for BPD (MSI-BPD) offers a brief screen mapped to the DSM criteria, while the Borderline Symptom List (BSL-23) provides a dimensional measure of borderline symptom severity suitable for both diagnostic support and outcome monitoring. Administering instruments for both constructs is the most defensible approach, precisely because comorbidity is common and a single positive screen settles nothing.
The bipolar versus BPD distinction rewards careful attention to the grain of mood disturbance. Sustained, autonomous episodes lasting days to weeks, with euphoric and sleep-curtailed highs, point toward bipolar disorder; rapid, interpersonally reactive shifts measured in hours, set against a backdrop of identity disturbance and abandonment fear, point toward BPD. But the clinician’s task is not always to choose one over the other. With roughly one in five clients qualifying for both, and with BPD a documented driver of bipolar overdiagnosis, the most rigorous stance is to assess both constructs deliberately, anchor the formulation in episode duration and trigger, and let structured measurement sharpen, not replace, clinical judgement.
For broader guidance on distinguishing overlapping presentations, see our pillar resource on Differential Diagnosis for Mental Health Conditions, and related guides on BPD vs Complex PTSD and ADHD vs Anxiety.
Access 600+ scored, norm-referenced assessments to support accurate differential diagnosis.
Bayes, A., Parker, G., & Paris, J. (2019). Differential diagnosis of bipolar II disorder and borderline personality disorder. Current Psychiatry Reports, 21(12), 125. https://doi.org/10.1007/s11920-019-1120-2
Carpenter, R. W., Stanton, K., Emery, N. N., & Zimmerman, M. (2020). Positive and Negative Activation in the Mood Disorder Questionnaire: Associations with psychopathology and emotion dysregulation in a clinical sample. Assessment, 27(2), 219–231. https://doi.org/10.1177/1073191119851574
Dagani, J., Signorini, G., Nielssen, O., Bani, M., Pastore, A., de Girolamo, G., & Large, M. (2017). Meta-analysis of the interval between the onset and management of bipolar disorder. The Canadian Journal of Psychiatry, 62(4), 247–258.
Fornaro, M., Orsolini, L., Marini, S., De Berardis, D., Perna, G., Valchera, A., Ganança, L., Solmi, M., Veronese, N., & Stubbs, B. (2016). The prevalence and predictors of bipolar and borderline personality disorders comorbidity: Systematic review and meta-analysis. Journal of Affective Disorders, 195, 105–118. https://doi.org/10.1016/j.jad.2016.01.040
Henry, C., Mitropoulou, V., New, A. S., Koenigsberg, H. W., Silverman, J., & Siever, L. J. (2001). Affective instability and impulsivity in borderline personality and bipolar II disorders: Similarities and differences. Journal of Psychiatric Research, 35(6), 307–312. https://doi.org/10.1016/S0022-3956(01)00038-3
Hirschfeld, R. M. A., Williams, J. B. W., Spitzer, R. L., Calabrese, J. R., Flynn, L., Keck, P. E., Jr., Lewis, L., McElroy, S. L., Post, R. M., Rapport, D. J., Russell, J. M., Sachs, G. S., & Zajecka, J. (2000). Development and validation of a screening instrument for bipolar spectrum disorder: The Mood Disorder Questionnaire. American Journal of Psychiatry, 157(11), 1873–1875. https://doi.org/10.1176/appi.ajp.157.11.1873
Johansson, V., Kuja-Halkola, R., Cannon, T. D., Hultman, C. M., & Hedman, A. M. (2019). A population-based heritability estimate of bipolar disorder in a Swedish twin sample. Psychiatry Research, 278, 180–187. https://doi.org/10.1016/j.psychres.2019.06.010
Reich, D. B., Zanarini, M. C., & Fitzmaurice, G. (2012). Affective lability in bipolar disorder and borderline personality disorder. Comprehensive Psychiatry, 53(3), 230–237. https://doi.org/10.1016/j.comppsych.2011.04.003
Ruggero, C. J., Zimmerman, M., Chelminski, I., & Young, D. (2010). Borderline personality disorder and the misdiagnosis of bipolar disorder. Journal of Psychiatric Research, 44(6), 405–408. https://doi.org/10.1016/j.jpsychires.2009.09.011
Skoglund, C., Tiger, A., Rück, C., Petrovic, P., Asherson, P., Hellner, C., Mataix-Cols, D., & Kuja-Halkola, R. (2021). Familial risk and heritability of diagnosed borderline personality disorder: A register study of the Swedish population. Molecular Psychiatry, 26(3), 999–1008. https://doi.org/10.1038/s41380-019-0442-0
Zimmerman, M., Ruggero, C. J., Chelminski, I., & Young, D. (2008). Is bipolar disorder overdiagnosed? Journal of Clinical Psychiatry, 69(6), 935–940. https://doi.org/10.4088/jcp.v69n0608