Authoritative Definition and Systematic Review of Bipolar Disorder

Disclaimer: This article presents aggregated research findings. It is not a medical paper, cannot substitute for a physician, and should not be used as diagnostic evidence. The reasoning and conclusions in this article involved AI-assisted generation, have not undergone peer review or independent verification, and may contain errors.

TL;DR

Bipolar disorder (BD) is a group of chronic, highly heritable, potentially disabling affective psychiatric disorders characterized by recurrent episodes of mania, hypomania, and depression lasting days to months -- it is not "mood swings" or "moodiness," but a disease entity with well-defined symptomatological, temporal, and functional impairment thresholds.
Under the current criteria of DSM-5-TR (APA, 2022) and ICD-11 (WHO, 2022; CDDR 2024), the main clinical subtypes are bipolar I (at least one manic episode lasting >=1 week), bipolar II (at least one hypomanic episode + at least one major depressive episode, never a manic episode), and cyclothymic disorder (>=2 years of subthreshold fluctuations); the global bipolar spectrum lifetime prevalence is approximately 2.4% (Merikangas et al., 2011, Arch Gen Psychiatry).
The key distinction from borderline personality disorder (BPD) and major depressive disorder (MDD) is that mood changes in bipolar disorder occur in discrete "episodes," lasting days to months, and are accompanied by objective changes in activity/energy, rather than "affective instability" triggered by interpersonal conflict and fluctuating reactively within minutes to hours.

Key Findings

The diagnostic "threshold" is the duration of episodes and changes in activity/energy, not the intensity of mood itself. Both DSM-5 (2013) and ICD-11 (2022) added "sustained increase in activity or energy" as a gateway criterion (Criterion A / essential feature) for mania/hypomania, not merely "elevated/irritable mood."
Bipolar I: the minimum threshold is a single manic episode lasting >=1 week (or any duration requiring hospitalization); bipolar II requires >=1 hypomanic episode (DSM-5-TR: >=4 days; ICD-11: "lasting several days / at least several days") + >=1 major depressive episode (>=2 weeks) with no lifetime history of a full manic episode.
Cyclothymic disorder: >=2 years (adolescents >=1 year) of subthreshold fluctuations, with symptom-free periods not exceeding 2 months.
Rapid cycling is a "specifier" in both DSM-5-TR and ICD-11, defined as >=4 mood episodes within 12 months; ultra-rapid and ultradian cycling (Kramlinger & Post, 1996, Br J Psychiatry) are research concepts not formally adopted by DSM-5-TR or ICD-11.
Etiology: bipolar disorder is one of the most heritable of all psychiatric disorders (twin heritability ~ 85%; McGuffin et al., 2003, Arch Gen Psychiatry); the latest GWAS (Mullins et al., 2021, Nature Genetics) identified 64 genome-wide significant risk loci in 41,917 cases, enriched in synaptic signaling, calcium channels (CACNA1C), and targets of antipsychotic and antiepileptic drugs.
Core misconceptions: bipolar disorder =/= "drastic mood swings within a single day," =/= borderline personality disorder, =/= simply "being temperamental." The "mood fluctuations" experienced by the general population differ qualitatively, not merely quantitatively, from clinical episodes.

Details

1. Official Diagnostic Definitions

1.1 DSM-5-TR (American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision, APA, 2022)

DSM-5-TR places "Bipolar and Related Disorders" as an independent chapter, positioned between schizophrenia spectrum disorders and depressive disorders, reflecting its position as a bridge between the two categories in terms of symptomatology and genetics.

A. Manic Episode -- the necessary condition for bipolar I

A. A distinct period of abnormally and persistently elevated, expansive, or irritable mood and abnormally and persistently increased goal-directed activity or energy, lasting at least 1 week (or any duration if hospitalization is required), present most of the day, nearly every day. B. During the period of mood/activity disturbance, >=3 of the following symptoms (>=4 if mood is only irritable) are present to a significant degree and represent a noticeable change from usual behavior: (1) Inflated self-esteem or grandiosity (2) Decreased need for sleep (e.g., feels rested after only 3 hours of sleep) (3) More talkative than usual or pressure to keep talking (pressured speech) (4) Flight of ideas or subjective experience that thoughts are racing (5) Distractibility (6) Increase in goal-directed activity or psychomotor agitation (7) Excessive involvement in activities that have a high potential for painful consequences (e.g., unrestrained buying sprees, sexual indiscretions, foolish business investments) C. The mood disturbance causes marked impairment in social/occupational functioning, necessitates hospitalization, or there are psychotic features. D. The episode is not attributable to the physiological effects of a substance or another medical condition (Note: a full manic episode emerging during antidepressant treatment but persisting beyond the physiological effect of that treatment may be considered a manic episode and warrant a bipolar I diagnosis).

The mnemonic DIG FAST (Distractibility, Indiscretion, Grandiosity, Flight of ideas, Activity, Sleep, Talkativeness) is widely used in English-language clinical teaching.

B. Hypomanic Episode -- the core of bipolar II

Same A and B criteria as mania, but:

Minimum duration is >=4 days (most of the day, nearly every day);
Does not cause marked functional impairment, does not require hospitalization, and has no psychotic features (if any of these are present, it is upgraded to mania).

C. Major Depressive Episode

Over >=2 weeks, nearly every day for most of the day, depressed mood or loss of interest/pleasure must be present (at least one of the two is required), with a total of >=5 symptoms, including changes in weight/appetite, insomnia/hypersomnia, psychomotor agitation/retardation, fatigue, feelings of worthlessness/guilt, diminished concentration, and recurrent thoughts of death or suicidal ideation.

D. Three main subtypes

Subtype	DSM-5-TR required criteria	ICD-11 code
Bipolar I (BD-I)	>=1 manic episode; depressive episodes are not required	6A60
Bipolar II (BD-II)	>=1 hypomanic episode + >=1 major depressive episode; never had a manic or mixed episode	6A61
Cyclothymic Disorder	>=2 years (adolescents >=1 year) of recurrent hypomanic-like and depressive-like symptoms that never meet full episode criteria; symptoms present >=50% of the time period; symptom-free periods <=2 months; never met full criteria for any episode	6A62

E. Important DSM-5/DSM-5-TR changes

Elimination of "mixed episode" as a standalone diagnosis, replaced by the "with mixed features" specifier: >=3 non-overlapping depressive symptoms during a manic/hypomanic episode, or >=3 non-overlapping (hypo)manic symptoms during a depressive episode (APA, Highlights of Changes from DSM-IV-TR to DSM-5). Irritability, distractibility, and psychomotor agitation are excluded from the mixed features specifier because they can occur in both poles.
Other available specifiers: with anxious distress, with rapid cycling, with melancholic features, with atypical features, with mood-congruent/mood-incongruent psychotic features, with catatonia, with peripartum onset, with seasonal pattern.

1.2 ICD-11 (WHO, International Classification of Diseases, 11th Revision, 2022; CDDR 2024)

ICD-11 places bipolar disorder under the "Mood Disorders" chapter (6A6x), with criteria that are highly convergent with but not identical to DSM-5 (Severus & Bauer, 2020, Int J Bipolar Disord; Reed et al., 2019, World Psychiatry 18(1):3-19).

Official verbatim definition (WHO ICD-11, 2022, CC BY-ND 3.0 IGO):

Manic episode: "An extreme mood state lasting at least one week (unless shortened by treatment), characterized by euphoria, irritability, or expansive mood, along with increased activity or a subjective experience of increased energy" -- both gateway symptom types (mood + activity/energy) must be present. If there is a prior history of manic/mixed episodes, the 1-week duration requirement no longer applies.
Hypomanic episode: "A mood state lasting at least several days" -- ICD-11 intentionally avoids the DSM-5 hard threshold of ">=4 days" to reduce false negatives (Severus & Bauer, 2020).
Depressive episode: Nearly every day for most of the day, >=2 weeks; at least 5 symptoms, with at least one from the "affective cluster" (depressed mood or loss of interest; ICD-11 no longer counts "fatigue" in the affective cluster, a change that is more stringent than ICD-10; Reed et al., 2019).
Mixed episode (6A60.9 / 6A60.A): Manic and depressive symptoms "co-occurring or rapidly alternating within the same day," most of the day, nearly every day, lasting >=2 weeks (may be shortened by treatment; the 2-week requirement may not apply in cases of recurrence).
Cyclothymic disorder (6A62): "At least 2 years of persistent mood instability, with recurrent hypomanic-like and depressive-like symptoms present more of the time than not"; no history of manic/mixed episodes; depressive symptoms never meet full episode criteria.

Key differences between ICD-11 and DSM-5-TR:

ICD-11 uses "essential features" rather than numbered criteria A-D, encouraging clinical judgment (Angst et al., 2020).
ICD-11 uses "several days" for hypomania rather than "4 days"; DSM-5 retains "4 days."
ICD-11's cyclothymic disorder allows hypomanic periods to meet full episode criteria; DSM-5-TR stipulates that full episode criteria are never met.
Antidepressant-induced (hypo)mania can count toward a bipolar diagnosis in both systems, but the ICD-11 draft requires a prior depressive episode.

2. Core Clinical Features and Differential Diagnosis

2.1 Cardinal symptom cluster of mania

Mania is not "being happy" but a bio-psycho-behavioral syndrome:

Physiological changes: Markedly decreased need for sleep (typical: 3 hours per night yet feeling energized), increased libido and appetite, psychomotor agitation;
Cognitive changes: Flight of ideas, distractibility, grandiose delusions (can reach psychotic levels), loss of judgment;
Behavioral changes: Dramatic increase in goal-directed activity (staying up all night starting multiple projects), uncontrolled spending, sexual indiscretions, reckless driving;
Social changes: Pressured speech, excessive self-confidence, irritability, proneness to interpersonal conflict;
Anosognosia: In most cases, patients themselves do not recognize that they are ill.

2.2 Features of depressive episodes (bipolar depression vs. unipolar depression)

Although DSM/ICD use the same criteria for both unipolar and bipolar depression, there are clinical phenomenological differences (Goodwin & Jamison, Manic-Depressive Illness, Oxford, 2007):

Bipolar depression more commonly presents with: psychomotor retardation, hypersomnia (rather than insomnia), increased appetite, leaden paralysis, family history of bipolar disorder, early onset (<25 years), postpartum onset, antidepressant-induced (hypo)mania or rapid cycling.
Mixed features are more common than pure depression; Cafaro et al. (2023) analyzed the Stanley Foundation Bipolar Network cohort (n=903, 7-year follow-up with 14,213 visits), showing that the DSM-5 "mixed features specifier" has low sensitivity because it excludes irritability, agitation, and distractibility;
The depressive phase is the period of highest suicide risk (Schaffer et al., 2015, Bipolar Disorders, International Society for Bipolar Disorders Suicide Task Force meta-analysis).

2.3 Differentiation from major depressive disorder (MDD)

The key question: Has the patient ever experienced a (hypo)manic episode? This is the fundamental taxonomic divide (Leonhard, 1957). Clinicians should proactively ask:

Have you ever had a period of several days without sleep yet full of energy?
Have family or friends ever commented that "you seemed like a different person"?
Have you ever had a period of impulsive spending, sexual indiscretions, or rapidly starting multiple projects?
After taking antidepressants, did you experience a "switch" effect?

Hirschfeld et al. (2003, J Clin Psychiatry 64(2):161-74) in the NDMDA national bipolar disorder patient survey (n=600) found that 69% of bipolar disorder patients had been misdiagnosed, with the most common misdiagnosis being unipolar depression -- this is precisely why proactive screening for prior mania is critically important.

2.4 Differentiation from borderline personality disorder (BPD)

The two are frequently confused but are distinct entities (Zimmerman & Morgan, 2013, Dialogues Clin Neurosci 15(2):155-69; Paris, Psychiatric News, 2024; Ghaemi, 2013, Psychiatry Investigation):

Dimension	Bipolar Disorder	Borderline Personality Disorder
Unit of mood change	Episodes: days to months	Reactive: minutes to hours
Triggers	Spontaneous (occasionally stress-induced)	Nearly always triggered by interpersonal conflict/abandonment
Accompanying physiological changes	Changes in sleep, energy, appetite, libido	Primarily emotional/cognitive
Family history	Strong (heritability ~85%)	Relatively weak
Self-identity	Stable during inter-episode periods	Chronically unstable
Treatment response	Mood stabilizers (lithium, valproate, etc.)	Psychotherapy (DBT) is first-line

Zimmerman & Morgan (2013, Dialogues Clin Neurosci) noted: "Approximately 10% of BPD patients also have bipolar I, and another approximately 10% have bipolar II; conversely, approximately 20% of bipolar II patients also carry a BPD diagnosis, but among bipolar I patients that proportion is only approximately 10%." Therefore, clinicians must exercise caution in differential diagnosis, avoiding both under-diagnosis of bipolar disorder and excessive use of mood stabilizers in BPD patients.

3. Epidemiology

Global lifetime prevalence (WHO World Mental Health Surveys; Merikangas et al., 2011, Arch Gen Psychiatry 68(3):241-251, 11 countries, n=61,392): BD-I = 0.6%; BD-II = 0.4%; subthreshold bipolar = 1.4%; bipolar spectrum total = 2.4%. Twelve-month prevalence was 0.4%/0.3%/0.8% respectively.
U.S. NCS-R (Merikangas et al., 2007, Arch Gen Psychiatry): lifetime BPS = 4.4%.
Mainland China: China Mental Health Survey (CMHS; Huang et al., 2019, Lancet Psychiatry, n=32,552 adults): BD-I lifetime = 0.4%, BD-II < 0.1%, BD total ~ 0.6% -- significantly lower than the West. Possible reasons include underreporting due to stigma, low sensitivity of the CIDI instrument for BD-II, and cultural differences in expression.
Age of onset: In most studies, median age of onset is approximately 18-25 years; Merikangas et al. (2011) showed that 50% of cumulative onset of bipolar spectrum occurred before age 25. Mean age at first manic episode is approximately 22 years for males and approximately 25-27 years for females (Kennedy et al., 2005, Am J Psychiatry, Camberwell 35-year cohort).
Sex differences: BD-I has comparable overall prevalence in males and females; BD-II, rapid cycling, and mixed episodes are more common in females; males more commonly present with mania-first onset and comorbid substance use disorders; females more commonly present with depression-first onset, seasonal pattern, peripartum and perimenopausal recurrence (Diflorio & Jones, 2010; Psychiatric Times, 2023).
Suicide: Lifetime suicide attempt rate ~33.9% (Dong et al., 2020, meta-analysis); suicide mortality risk is 20-30 times higher than the general population (Plans et al., 2019, J Affect Disord), and approximately 10-15% of BD patients die by suicide; lithium has a specific anti-suicidal effect (Cipriani et al., 2013, BMJ).

4. Episode Duration and Cycling Patterns

4.1 Typical natural course (untreated)

Manic episodes: median ~3.5 months;
Depressive episodes: median ~5.2 months (when defined as major depressive episodes, median approximately 15 weeks, with 75% remitting within 35 weeks; Solomon et al., NIMH-CDS cohort).
Both are significantly shortened with treatment.

4.2 Precise terminology for cycling patterns (Kramlinger & Post, 1996, Br J Psychiatry 168(3):314-323)

Pattern	Definition	Adopted by DSM-5-TR / ICD-11?
Rapid cycling	>=4 episodes meeting full criteria for manic/hypomanic/depressive/mixed episodes within 12 months	Adopted by both DSM-5-TR and ICD-11 as a specifier
Ultra-rapid cycling	>=4 episodes per month, or cycle length < several weeks (original definition: 48-hour cycle, Alarcon, 1985)	Not formally adopted (research terminology)
Ultradian cycling (ultra-ultra-rapid cycling)	Distinct polarity switches within 24 hours	Not formally adopted; highly controversial; overlaps with the concept of "mixed states"; first described by Kramlinger & Post (1996) based on systematic observation of 5 patients at NIMH

Clinical key point: DSM-5-TR does not require polarity switches; any 4 episodes meeting threshold (including repeated episodes of the same polarity) qualifies as rapid cycling. This contrasts with the common public misconception of calling "drastic mood swings within a single day" rapid cycling.

5. Neurobiological Basis

5.1 Genetics

Familial aggregation: First-degree relatives of bipolar patients have a 5-10 fold higher prevalence than the general population (Craddock & Sklar, 2013, Lancet).
Twin heritability: McGuffin et al. (2003, Arch Gen Psychiatry 60(5):497-502) reported BD heritability of 0.85 (95% CI 0.73-0.93); Kieseppa et al. (2004, Am J Psychiatry 161(10):1814-1821, Finnish national twin cohort) reported 0.93. Combined estimates range from 60-85%.
GWAS: Mullins et al. (2021, Nature Genetics 53:817-829, Psychiatric Genomics Consortium, 41,917 cases vs. 371,549 controls) identified 64 genome-wide significant loci, of which 33 were novel; risk alleles were enriched in synaptic signaling pathways, genes expressed in prefrontal cortex and hippocampal neurons, and targets of antipsychotics, calcium channel blockers, antiepileptics, and anesthetics; key genes include CACNA1C, ANK3, TRANK1, POU3F2, among others.
Common variant SNP-h-squared is approximately 20%; polygenic risk scores (PRS) explain approximately 4% of phenotypic variance in independent samples, indicating a highly polygenic architecture with genetic overlap with schizophrenia and major depression.
Rare variants: e.g., 16p11.2 copy number variants suggest larger effect sizes.

5.2 Neurochemistry

Dopamine hypothesis (Ashok et al., 2017, Molecular Psychiatry 22(5):666-679): During mania, striatal D2/D3 receptor availability is elevated and reward circuitry is hyperactive; during bipolar depression, striatal dopamine transporter (DAT) levels are elevated and dopaminergic transmission is reduced. This "dopamine receptor-transporter homeostasis imbalance model" explains the seemingly contradictory clinical evidence that antipsychotics are effective for mania while dopamine agonists (e.g., pramipexole) are effective for bipolar depression.
Monoamine overview: 5-HT, NE, and DA systems are all involved;
Glutamate/GABA imbalance: Elevated cortical glutamate and reduced GABAergic inhibition during mania (Sigitova et al., 2017, Psychiatry Clin Neurosci);
Mechanisms of action of lithium, valproate, lamotrigine, etc. involve GSK-3-beta, the inositol phosphate pathway, voltage-gated sodium channels, and glutamate modulation.

5.3 Neuroimaging

Structural MRI (ENIGMA-BD Working Group, Hibar et al., 2016/2018, Mol Psychiatry; Ching et al., 2022, Biol Psychiatry, 1232 longitudinal cases): Reduced volumes of the amygdala, hippocampus, and thalamus; increased ventricular volume; reduced cortical thickness of the prefrontal cortex (especially dlPFC, vlPFC, OFC) and temporal cortex; decreased fractional anisotropy in white matter tracts (especially fibers connecting the prefrontal cortex to the limbic system).
Functional MRI (Phillips & Swartz, 2014, Am J Psychiatry; Chase & Phillips, 2016): Abnormalities in three major circuits -- emotional processing, emotion regulation, and reward processing -- featuring deficient ventrolateral prefrontal cortex-amygdala regulatory circuitry combined with left ventral striatum-vlPFC reward circuit hyperactivation.
Longitudinal MRI in children and adolescents shows further amygdala volume abnormalities after the first manic episode, suggesting a dual neurodevelopmental and neurodegenerative process.

5.4 Metabolism/systems biology

Mitochondrial dysfunction: A core hypothesis long advocated by Kato et al., involving impaired oxidative phosphorylation and ATP production;
Elevated oxidative stress, reduced neurotrophic factors (BDNF), and elevated neuroinflammatory markers (IL-6, TNF-alpha);
Circadian rhythm disruption: Variants in CLOCK, BMAL1, PER, and other genes; disruption of the sleep-wake cycle can trigger mania (Wehr et al., classic observations).
These mechanisms collectively support the "neuroprogression" hypothesis -- recurrent episodes can worsen functional and cognitive impairment (Scaini et al., 2020, Braz J Psychiatry 42(5):536-551).

6. Historical Evolution of the Concept

Period	Key Figure(s)	Contribution
Ancient Greece-Rome	Aretaeus of Cappadocia (1st-2nd century CE)	First recorded observation that "melancholia and mania alternate in the same person" -- the earliest clinical observation of the bipolar phenomenon in history.
1854	Jules Baillarger (France) and Jean-Pierre Falret (France)	Independently proposed "folie a double forme" and "folie circulaire," respectively.
1899/1913	Emil Kraepelin	In the 6th edition of his Compendium, proposed "manic-depressive insanity (manisch-depressives Irresein / MDI)", unifying circular insanity, simple mania, melancholia, and mixed states into a single category, in opposition to "dementia praecox" (later schizophrenia). Kraepelin's MDI concept was broader than today's BD, encompassing recurrent unipolar depression.
1957	Karl Leonhard (Germany)	In Aufteilung der endogenen Psychosen, divided MDI into unipolar and bipolar, where the former involves only recurrent depression or only mania, and the latter involves both depression and mania. Family history studies subsequently confirmed this distinction.
1966	Jules Angst (Switzerland) and Carlo Perris (Sweden)	Independently validated the unipolar-bipolar dichotomy through family studies.
1980	DSM-III	The American Psychiatric Association formally adopted Leonhard's dichotomy; "bipolar disorder" first appeared as an official term in the DSM.
1976 onward	Hagop S. Akiskal	Proposed the "bipolar spectrum": BD-I -> BD-II -> BD-III (antidepressant-induced hypomania) -> BD-IV (recurrent depression on a hyperthymic temperament background) -> cyclothymic and hyperthymic temperaments. Emphasized that subthreshold bipolar is severely underdiagnosed.
1994/2013/2022	DSM-IV -> DSM-5 -> DSM-5-TR	Introduced the rapid cycling specifier, mixed features specifier, and added "activity/energy" as a required gateway symptom.
2022	ICD-11 officially in effect	Adopted a three-subtype classification similar to DSM-5, but retained ICD's characteristic "essential features" narrative approach.

As Ghaemi (2013, Psychiatry Investigation) stated: "Post-DSM-III American psychiatry is not neo-Kraepelinian but neo-Leonhardian" -- that is, it retreated from Kraepelin's broad unifying concept to the unipolar/bipolar dichotomy, while Akiskal's "spectrum" movement represents a return to Kraepelin.

7. Common Misconceptions (What Bipolar Disorder is NOT)

It is not "mood ups and downs within a single day." The minimum thresholds for clinical mania/hypomania/depression are >=1 week/>=4 days/>=2 weeks, respectively. Saying "my colleague is happy one moment and unhappy the next -- classic bipolar" is almost never correct.
It is not borderline personality disorder. BPD mood changes occur over minutes to hours, are nearly always triggered by interpersonal conflict, and involve chronic identity instability; BD follows a longitudinal pattern of episode-remission cycles with objective changes in activity/energy/sleep.
It is not "inability to control one's temper" or a "personality flaw." It is a highly heritable, polygenic brain disease that can affect brain structure and function.
It does not only have "two poles" -- it has a broad "middle ground": subsyndromal symptoms, mixed features, comorbid anxiety/substance use, and cognitive deficits are all part of the illness course.
It is not synonymous with "creativity." Although some studies show slightly higher rates of bipolar spectrum in artists, the vast majority of patients experience significant functional impairment and suffering; romanticizing bipolar disorder delays help-seeking.
It is not untreatable. Lithium (since Cade's 1949 report) remains the gold standard mood stabilizer and significantly reduces suicide; psychotherapy (CBT, IPSRT, family-focused therapy) has well-established effects on relapse prevention during the maintenance phase.
Rapid cycling does not mean "switching back and forth within a single day." The official definition of rapid cycling is >=4 episodes meeting threshold criteria within 12 months, while multiple switches within 24 hours belongs to "ultradian cycling," which is not a formally recognized diagnostic entity in DSM-5-TR / ICD-11 and is often a manifestation of mixed states, BPD, ADHD, or substance effects.

Recommendations

For clinicians:

For patients presenting with recurrent depression, proactively screen for prior (hypo)manic episodes using tools such as the MDQ (Mood Disorder Questionnaire), HCL-32, or Bipolarity Index. Hirschfeld et al. (2003, J Clin Psychiatry 64(2):161-74) in the NDMDA national survey (n=600) found that 69% of bipolar patients had been misdiagnosed, most commonly as unipolar depression.
Do not diagnose bipolar disorder based solely on "mood instability"; establish the longitudinal temporal structure of "episodes," focusing on objective changes in activity/energy and decreased need for sleep as the two most discriminating indicators (this is the core change in ICD-11 gateway criteria).
Avoid excessive use of mood stabilizers or antipsychotics in BPD patients; Paris (2024) and Zimmerman & Morgan (2013, Dialogues Clin Neurosci) both warn about inappropriate medication resulting from BPD-BD misdiagnosis.
Conduct systematic suicide risk assessment for all BD patients, with priority consideration of lithium for patients with a history of suicidal behavior.
Early diagnosis and psychoeducation are critical for prognosis; Pinto et al. (2021, J Affect Disord, HOPE-BD Canadian multi-center naturalistic observational study, n=319, 12 Canadian research centers) reported a median diagnostic delay of 5.0 years (IQR 1.0-12.0) for BD-I and 11.0 years for BD-II; earlier age of onset and lifetime suicide attempts were both associated with longer delays.

For patients and families:

Maintaining stable sleep rhythms is one of the most powerful non-pharmacological interventions for preventing manic relapse;
Keep a life chart / mood chart to help identify prodromal periods;
Do not discontinue medication on your own, even when symptoms have remitted. Suppes et al. (1991, Arch Gen Psychiatry 48(12):1082-1088) in a meta-analysis of 14 studies (n=257 BD-I patients) showed that after an average of 30 months of treatment followed by discontinuation, over 50% of patients experienced a new episode within 10 weeks, with a median time to loss of remission of only 5.0 months.

Thresholds/benchmarks for triggering action:

If >=4 days of decreased sleep + increased energy + pressured speech occur -> contact a psychiatrist immediately;
If >=4 episodes of any type occur within 12 months -> apply the rapid cycling specifier and reassess treatment (avoid antidepressant monotherapy);
If a "switch" effect occurs after antidepressant use -> reassess whether this represents undiagnosed BD.

Caveats

Diagnostic criteria continue to evolve: Although DSM-5-TR (2022) and ICD-11 (2022, complete CDDR published 2024) are highly convergent, differences remain in hypomania duration thresholds (DSM 4 days vs. ICD "several days") and the definition of cyclothymic disorder. Clinical diagnosis should also reference local authoritative guidelines (e.g., China's Guidelines for the Prevention and Treatment of Bipolar Disorder, 2nd Edition, 2018, Lu Lin et al.).
Controversy over the classification of subthreshold bipolar persists: Akiskal's bipolar spectrum advocates for broad diagnosis, but critics (Paris et al.) raise concerns about overdiagnosis risk -- currently no biomarker can adjudicate.
Chinese epidemiological data (CMHS, 2019) are far below the global average, and the specific reasons (true difference vs. diagnostic instrument issues) remain unresolved.
Neurobiological findings are mostly correlational, not causal: GWAS loci have small effect sizes, neuroimaging abnormalities show substantial heterogeneity, and no biomarker currently exists for individual-level diagnosis -- clinical diagnosis remains primarily phenomenology-based (Reed et al., 2019, World Psychiatry 18(1):3-19).
This document does not constitute medical advice; any diagnosis and treatment must be carried out by a qualified psychiatrist.
Regarding "ultradian cycling": Kramlinger & Post (1996, Br J Psychiatry) first described it based on close observation of 5 patients at NIMH, an extremely small sample; subsequent large-sample validation is limited; some experts (Zimmerman, Goldberg) argue that the construct lacks content validity and discriminant validity and is easily confused with mixed states, BPD, and ADHD. When citing this term, its research-only, non-official diagnostic status should be clearly stated.