Bipolar disorder vs borderline personality disorder: documenting the differential
Few differentials in outpatient psychiatry carry higher stakes than bipolar disorder versus borderline personality disorder. The treatments diverge sharply, both conditions present with mood instability, and a wrong label can follow a patient for a decade. The clinical work is hard enough. The documentation work is where many otherwise careful clinicians fall short: the chart has to show why you concluded what you concluded, not just what you concluded. This guide covers the differential and how to write it down.
Why they get confused
The overlap is real, not a product of sloppy assessment. Both conditions involve mood instability, impulsivity, irritability, and elevated self-harm and suicide risk. Both can present with agitation, subjective racing thoughts, and chaotic relationships. In a single cross-sectional visit, especially a crisis visit, the presentations can look nearly identical. A patient in the middle of an intense affective storm does not arrive wearing a diagnosis.
The differential is longitudinal. Almost every reliable distinguishing feature lives in the time course: how long mood states last, what triggers them, what the person is like between them, and when the pattern began. A snapshot cannot answer those questions. A history can.
The core differentiators
Every row below describes a classic pattern, not an absolute rule. Atypical presentations exist on both sides, and no single feature settles the question by itself. The features work as converging evidence.
| Feature | Classic bipolar pattern | Classic borderline pattern |
|---|---|---|
| Mood shifts | Sustained episodes lasting days to weeks, a clear departure from baseline | Rapid shifts over hours, typically triggered by interpersonal events |
| Sleep | Decreased need for sleep with preserved or increased energy during elevation | Insomnia with next-day fatigue; the person wants sleep and misses it |
| Between states | Return toward a recognizable baseline between episodes | Pervasive pattern with no true episodic offset |
| Self-experience | Grandiosity or expansiveness during elevated periods | Chronic emptiness, unstable self-image, abandonment sensitivity |
| Onset and course | Episodic course, often with an identifiable first episode | Pattern traceable to adolescence, running continuously since |
| Family history | Bipolar disorder in first-degree relatives adds meaningful weight | Less diagnostically specific |
Episodicity versus reactivity is the spine of the differential. Bipolar mood episodes are sustained states that represent a departure from the person's baseline and persist across contexts: the elevated week is elevated at work, at home, and alone. Borderline mood shifts are typically rapid, measured in hours, and tied to interpersonal triggers such as a perceived rejection, with mood often recovering when the interpersonal weather changes.
Sleep is the most underused question in this differential. Decreased need for sleep, meaning three or four hours followed by a morning of energy and productivity, is a strong pointer toward mania or hypomania. Insomnia with next-day exhaustion is common in nearly everything else, including BPD, depression, and anxiety. Asking "when you slept four hours, how did you feel the next day?" does more work than most screening instruments.
The DSM-5-TR anchors, summarized rather than quoted: a manic episode requires a distinct period of abnormally elevated, expansive, or irritable mood with increased energy or activity lasting at least one week, or any duration if hospitalization becomes necessary, accompanied by several characteristic symptoms such as decreased need for sleep, grandiosity, or pressured speech, and causing marked impairment. Hypomania requires a similar distinct period lasting at least four consecutive days, observable to others as a change from baseline, without marked impairment or psychosis. Borderline personality disorder, by contrast, is defined by a long-standing pattern, present since early adulthood, of unstable relationships, shifting self-image, and volatile emotion, together with marked impulsivity; at least five of nine criteria must be met. The structural difference matters: one diagnosis is built on episodes, the other on a pattern. Your documentation should reflect which structure you found.
They can and do co-occur
This is not an either/or differential. Comorbidity between bipolar disorder and BPD is well documented, and while studies vary widely on the rate, no serious reading of the literature supports treating the two as mutually exclusive. The practical documentation consequence: a both/and conclusion requires documenting each condition independently. If the chart says both, it should show discrete mood episodes meeting episode criteria and a pervasive borderline pattern that persists between and outside those episodes. A comorbid label supported only by "features of both" is the weakest entry in the chart, because it commits to two treatments' worth of implications on one condition's worth of evidence.
The assessment that earns the diagnosis
The differential is settled by evidence gathered over time, and the intake is where you set that process up honestly.
- Longitudinal history. Walk the timeline back to adolescence. Ask for the single best example of a "high" period and interrogate its duration, pervasiveness, sleep need, and trigger.
- Collateral information, where possible and consented. Family members often remember sustained episodes the patient minimizes, and just as often confirm that the "mood swings" have always lasted hours, not weeks.
- Screening tools, held loosely. The MDQ and similar instruments raise or lower suspicion; they never diagnose. A positive MDQ in a patient with prominent affective instability is expected, because the instrument captures instability from several sources. See the scales guide for the same discipline applied to symptom measures.
- Prospective mood charting. Daily mood and sleep logs over weeks turn a retrospective argument into observed data. Sustained multi-day elevation with reduced sleep need shows up on a chart in a way memory cannot.
- Time. Sometimes the honest documented answer at intake is a provisional diagnosis with a stated plan to clarify. That is not indecision. It is the assessment working as designed.
For the intake documentation itself, see the 90791 vs 90792 guide.
What the documented reasoning looks like
The patient below is fictional. This is a single assessment paragraph from an intake note, written to show the differential reasoning on the page.
Assessment (intake, fictional patient)
A.R. is a 26-year-old presenting with "mood swings" and a prior chart label of bipolar II disorder assigned during an urgent evaluation two years ago. Today's history does not support discrete mood episodes: described highs last hours rather than days, are consistently tied to interpersonal events, and have never involved decreased need for sleep. When she sleeps four hours she reports exhaustion the next day, not energy. No period of elevated mood with increased activity sustained across four or more days could be elicited. Collateral from her mother (by phone, with consent) corroborates: no sustained elevated periods, but "ups and downs since middle school" centered on relationships. The longitudinal pattern since early adolescence includes chronic emptiness, intense fear of abandonment, recurrent superficial cutting after perceived rejection, and unstable self-image. MDQ screen today is positive (9 of 13 symptom items endorsed, with co-occurrence and at least moderate problem level reported); interpreted as a screen only, and in this context most consistent with the affective instability the instrument also captures. Family history is negative for bipolar disorder in first-degree relatives. Diagnosis: borderline personality disorder, provisional. Rule out bipolar II disorder, retained on the differential because hypomanic periods are frequently under-reported retrospectively. Plan to clarify: prospective daily mood and sleep charting for eight weeks, repeat collateral if any sustained elevation emerges, and re-examination of the diagnosis at each visit. Deferring maintenance mood-stabilizer initiation pending clarification; safety planning documented separately.
Documenting the differential
Five habits separate charts that show reasoning from charts that show conclusions:
- Name the competing diagnosis explicitly. "Rule out bipolar II" in the assessment tells every future reader the question was seen and taken seriously.
- State the evidence for and against, in the chart. The absent findings, such as no decreased sleep need and no sustained episodes, are often the most persuasive lines in the note.
- Use provisional and rule-out language honestly. Provisional means you expect the criteria will be met but the information is incomplete. It is a tool for accuracy, not a hedge to avoid commitment.
- Revisit and update as longitudinal data arrives. A provisional diagnosis with no follow-up entries is a promise the chart never kept. When the mood charting comes back, document what it showed and what changed.
- Never let a first-visit label fossilize. The most common failure mode is not the initial miss; it is ten subsequent notes copying the miss forward.
Why the differential matters for treatment
Speaking generally and educationally: the management approaches differ, which is why the differential deserves this effort. Mood-stabilizing pharmacotherapy is central to bipolar disorder treatment, while structured psychotherapies such as DBT are generally considered first-line for BPD, with medications typically targeting specific symptoms rather than the disorder itself. Evidence continues to evolve on both sides, and individual treatment decisions belong to the treating clinician and patient.
Frequently asked questions
Can someone have both bipolar disorder and BPD?
Yes. Comorbidity is well documented. The documentation standard is to establish each independently: episode evidence for the bipolar diagnosis and a pervasive pattern, present outside mood episodes, for BPD.
Does a positive MDQ mean the patient has bipolar disorder?
No. The MDQ is a screen. It raises suspicion and earns a more careful episode history; it does not diagnose. Affective instability from other sources, including BPD, can produce positive screens, which is why the score should always be documented alongside its interpretation.
Which direction is misdiagnosis more common?
The literature discusses errors in both directions. A commonly discussed concern is patients with borderline personality disorder carrying bipolar labels, in part because "mood swings" invites the bipolar frame. The reverse also occurs. The safeguard is the same either way: document the evidence, not just the label.
How long should I take to settle the diagnosis?
As long as the evidence requires, documented provisionally along the way. Weeks of prospective mood charting plus collateral is a reasonable arc for many ambiguous presentations. A chart that shows an honest open question with a plan is stronger, clinically and medicolegally, than a premature answer.
Do AI scribes handle differential diagnosis?
They handle the documentation of it. OneStep Scribe drafts the assessment with the reasoning you voiced during the visit, including the rule-out language and the clarification plan, so the thinking you did out loud lands in the note. The differential itself is always the clinician's.
OneStep Scribe is an AI scribe built for psychiatric clinicians. It listens to the visit and drafts the full note, including an assessment that captures the differential reasoning you articulated, for your review and signature. Every account is NPI-verified.
Start a 14-day free trialThis article is clinician education about documentation and assessment. It is not diagnostic or treatment guidance for any individual, and it is not clinical, legal, or billing advice for any specific patient. It does not guarantee any clinical, billing, or audit outcome. All patients described are fictional. CPT is a registered trademark of the American Medical Association. Always verify against current diagnostic manuals, payer policies, and your compliance advisors.