Bipolar depression and diagnostic pitfalls

01OverviewDefinition, clinical context and the essential points that orientate the chapter.

A current depressive episode does not reveal whether the longitudinal disorder is unipolar or bipolar. Low mood, anhedonia, fatigue, guilt, sleep and appetite change, cognitive impairment, psychomotor change and suicidal thinking occur in both. Features such as earlier onset, recurrent episodes, postpartum episodes, atypical sleep or appetite change and family history may raise suspicion, but none establishes bipolarity. A defensible diagnosis requires evidence of a previous manic or hypomanic episode or an evolving specialist formulation.

Ask behaviourally anchored lifetime questions. Instead of asking only whether the person has been high, ask about periods when they needed little sleep, talked unusually rapidly, took on multiple projects, became markedly more sociable or irritable, spent beyond their means or behaved in ways others found uncharacteristic. Establish whether change was sustained, observable, episodic and followed by consequences. Antidepressant-associated activation is relevant evidence but is not by itself a complete diagnosis.

Collateral information can recover episodes the patient experienced as successful or has difficulty recalling. With consent, ask a trusted person about sleep, pace, spending and baseline. Review previous notes, prescriptions, admissions and perinatal records. Explain what information may be sought and the limits of confidentiality; if serious risk justifies disclosure without consent, share only what is necessary and document why. Use professional language support, and explore cultural meanings before classifying exuberance, spirituality or family concern as psychopathology.

Assess the current depression in full. Describe severity through symptoms, psychosis, catatonia, nutrition, self-care and concrete function. Ask directly about self-harm, suicide, agitation and mixed features. Examine medication adherence, recent dose changes, substances, thyroid or neurological symptoms, sleep disorders and reproductive context. A questionnaire can measure current symptoms but cannot distinguish bipolar from unipolar depression or predict individual suicide.

Treatment must address current need while preserving diagnostic humility. For established moderate or severe bipolar depression, follow the bipolar pathway rather than automatically prescribing unipolar antidepressant monotherapy. Discuss psychological and pharmacological options, previous response, overdose toxicity, reproductive safety and physical comorbidity. Share what is known, what remains uncertain and what new evidence would change the plan, with close review for activation or deterioration.

Key points

Bipolar depression can look identical to unipolar depression in cross-section; diagnosis depends on the lifetime course, not a special depressive symptom checklist.
Ask every depressed adult about previous overactivity, disinhibition, elevated or irritable mood, reduced sleep need, increased speech, projects, spending and treatment-related activation.
NICE advises considering specialist assessment when previous overactivity or disinhibition lasted four days or more and urgent referral for suspected mania, severe depression or danger.
Retrieve records and agreed collateral because past hypomania may have felt productive, occurred years earlier or been forgotten during the present low mood.
Assess current suicide risk, psychosis, catatonia, mixed activation, self-care, dependants, substances and access to medication before focusing on diagnostic terminology.
Check thyroid, anaemia, sleep, neurological, reproductive, medicine and substance explanations selectively from the history and examination; routine tests do not diagnose bipolar disorder.
For moderate or severe bipolar depression without current bipolar medication, NICE offers fluoxetine with olanzapine or quetiapine alone, with olanzapine alone or lamotrigine as preference-sensitive alternatives.
If lithium is already prescribed, check the plasma concentration and optimise an inadequate level before adding the NICE-recommended depressive treatment selected with the patient.
Monitor for emerging hypomania, worsening depression, adverse effects and suicide risk, and limit quantities where overdose toxicity is a material concern.

02AetiologyUnderlying causes, associations and risk factors, with why each one matters.

Underlying bipolar vulnerability

Depression occurs within a familial, polygenic and episodic mood disorder whose expression reflects development, prior episodes and environmental exposures.

Episode precipitants

Sleep disruption, childbirth, loss, illness, substances and treatment change can precipitate depression or mixed symptoms without defining the underlying diagnosis alone.

Maintaining adversity

Pain, isolation, debt, disrupted roles and consequences of previous mania can sustain hopelessness and withdrawal after the biological episode begins.

03PathophysiologyThe causal sequence from the underlying abnormality to symptoms and harm.

1
Mood-network polarity shifts
Bipolar illness involves changing regulation across reward, salience, arousal and cognitive-control systems rather than a fixed excess or deficit of one neurotransmitter.
2
Circadian and sleep feedback
Sleep and daily-rhythm disruption can precede, express and perpetuate depressive or mixed states, linking behavioural routine with episode vulnerability.
3
Withdrawal and reduced reward
Reduced activity and perceived failure limit rewarding experience, while negative predictions reinforce avoidance and functional loss during bipolar depression.
4
Mixed activation pathway
Depressive cognition can coexist with motor or mental activation, increasing capability, impulsivity and distress without producing a purely elevated presentation.

04Clinical features and red flagsSymptoms, examination findings, patterns of presentation and time-critical warnings.

Depression in longitudinal context

The depressive syndrome is accompanied by credible previous mania or hypomania, or remains under specialist evaluation because activation evidence is incomplete.

Past reduced sleep need

Distinct episodes of little sleep without fatigue plus increased energy carry more diagnostic weight than chronic insomnia alone.

Episodic behavioural change

Uncharacteristic spending, sociability, sexuality, irritability or projects clustered in time and resolved, rather than reflecting a lifelong stable trait.

Mixed depressive activation

Low mood or hopelessness coexists with racing thoughts, agitation, increased energy, impulsivity or reduced sleep need, raising immediate risk.

Bipolar treatment signal

Previous antidepressant activation, response to bipolar treatment or repeated postpartum episodes supports further assessment but remains evidence within a wider history.

Red flags requiring action

Suicidal preparation, access to lethal means, recent self-harm, severe hopelessness or an agitated mixed state requires immediate individual formulation and intervention.
Reduced sleep need, racing thoughts, agitation and impulsivity during apparent depression may indicate mixed features and greater capability for harmful action.
Delusions of guilt, poverty, bodily non-existence or deserved punishment, accusatory voices or catatonic signs indicate severe bipolar depression requiring urgent specialist care.
Confusion, fluctuating attention, focal neurology, fever, new seizure or abrupt late-life onset suggests delirium or neurological disease rather than an uncomplicated mood episode.
Depression, insomnia, irritability or psychosis arising during pregnancy or after birth requires rapid perinatal assessment and medication-risk review.

05InvestigationsWhat to request, why it matters and how to interpret it.

Investigation order

Read from the initial assessment onwards. Tests may run in parallel in urgent care; first-line, preferred, confirmatory, definitive and gold-standard labels appear only when the chapter explicitly states them.

01
Lifetime mood chronologyFirst step
Why
Map depressive, elevated, irritable and mixed periods, inter-episode recovery, duration, triggers and functional consequences.
Interpretation and limitations
A coherent episodic course supports bipolar formulation; uncertain dates or retrospective descriptions should be labelled uncertain instead of forced into criteria.
02
Collateral and record triangulation
Why
Confirm prior sleep, activity, speech, risk, admission, prescriptions and observed baseline change.
Interpretation and limitations
Concordant independent examples strengthen confidence. Disagreement prompts further enquiry and source documentation, not automatic dismissal of either account.
03
Current mental-state and risk assessment
Why
Assess depressive severity, psychosis, catatonia, mixed activation, cognition, insight, self-harm and functional danger.
Interpretation and limitations
Current urgency is driven by suicide, psychosis, self-neglect and activation, even while the precise longitudinal diagnosis remains provisional.
04
Medication and substance timeline
Why
Link mood change to antidepressants, corticosteroids, stimulants, alcohol, cannabis, other drugs and withdrawal.
Interpretation and limitations
Temporal association informs causality but does not automatically distinguish a substance-induced episode from exposure revealing underlying bipolar vulnerability.
05
Targeted physical-health screen
Why
Assess thyroid, anaemia, sleep, neurological, metabolic, reproductive and other phenotype-specific contributors.
Interpretation and limitations
Choose examination and tests for the suspected mechanism. Normal thyroid or blood results do not prove that the depression is primary or bipolar.

06Differential diagnosisRealistic alternatives and the features that help distinguish them.

Unipolar depression

The current phenotype may be identical; absence of established past mania or hypomania keeps unipolar depression plausible and requires longitudinal follow-up.

Substance or medicine effect

Alcohol, stimulants, cannabis, corticosteroids and antidepressant changes can cause depression, agitation or elevation and require a time-linked exposure history.

Trauma or personality-related distress

Reactive affective shifts and interpersonal crises can resemble cycling, but chronology, trauma phenomena and sustained episode features help distinguish processes that may coexist.

Endocrine or sleep disorder

Thyroid disease, anaemia, sleep apnoea and reproductive endocrine change can produce fatigue, cognition and mood symptoms needing targeted assessment.

07ManagementImmediate care, first-line treatment, alternatives and escalation.

01Depression with bipolar cluesRetrieve course before routine prescribingFirst stepCurrent depression is accompanied by previous activation, disinhibition or reduced sleep need.

1Establish current severity, suicide risk, psychosis and mixed features and provide urgent care first when danger or severe illness is present.
2Build a lifetime timeline using behavioural examples, records and consented collateral and review substances, medicines and medical explanations.
3EscalationSeek specialist bipolar assessment and agree interim safety, sleep and treatment plans that include monitoring for activation and clear escalation routes.

02Established bipolar depressionUse the bipolar treatment sequenceModerate or severe depression occurs in a person with a confirmed bipolar disorder diagnosis.

1Review current bipolar medication, adherence, plasma lithium concentration when relevant, previous response, overdose access and reproductive safety.
2Offer a trained bipolar psychological intervention and select NICE pharmacological options collaboratively according to current treatment and preference.
3EscalationReview symptoms, function, suicide risk, activation and adverse effects early, stopping ineffective add-on treatment and escalating specialist care when needed.

03Diagnosis remains uncertainTreat need and preserve uncertaintyDepression is clinically significant but prior activation evidence is incomplete or conflicting.

1State the depressive syndrome, leading unipolar and bipolar formulations and the evidence supporting and opposing each possibility.
2Address safety, physical contributors and psychosocial needs and avoid making a screening score or family history the diagnostic verdict.
3Arrange longitudinal review and record the observations, collateral or treatment changes that should trigger diagnostic and management revision.

Key medicines and prescribing safety2 treatments · regimens, roles and cautions

A NICE-offered monotherapy option for moderate or severe bipolar depression when benefits and adverse effects fit the individual.

Quetiapine immediate-release

For bipolar depression, take 50 mg at bedtime on day one, 100 mg on day two, 200 mg on day three and 300 mg on day four; the recommended daily dose is 300 mg.

Check metabolic and cardiovascular baseline, sedation, orthostasis and falls risk; monitor weight, glucose, lipids and movement effects and review interactions, pregnancy and driving safety.

A preference-sensitive NICE alternative for bipolar depression; acute treatment is outside this product's licensed prevention indication and slow titration prevents immediate control.

Lamotrigine

When not taking valproate or enzyme-inducing medicines, use 25 mg once daily for weeks one and two, 50 mg daily for weeks three and four, then increase gradually toward 200 mg daily according to the product schedule.

Dosing changes substantially with valproate or enzyme inducers; stop and assess any rash urgently because serious skin reactions are associated with rapid titration and early treatment.

08ComplicationsImportant consequences, why they occur and why they matter clinically.

Suicide and self-harm

Hopelessness, prior attempts, mixed energy, intoxication and access to toxic medication can converge into high-lethality behaviour across the episode.

Switch into activation

Spontaneous course or treatment can coincide with hypomania, mania or mixed symptoms, requiring prompt reassessment of diagnosis, medication and safety.

Cumulative role loss

Repeated depression and consequences of mania can disrupt employment, education, relationships, housing and finances, then reinforce future episode vulnerability.

Physical-health burden

Inactivity, smoking, disrupted sleep, missed medical care and metabolic treatment effects contribute to cardiovascular, metabolic and respiratory morbidity.

09Monitoring and follow-upTreatment response, safety checks and longer-term review.

Review suicidal thinking, mixed activation, sleep need, psychosis, self-care and access to toxic medicines at every early contact and after treatment changes.
Track depressive symptoms and concrete function alongside emerging elevation, irritability, increased activity or reduced need for sleep.
For antipsychotic treatment, monitor pulse and blood pressure after dose changes, weight weekly for six weeks and at twelve weeks, and glucose or HbA1c and lipids at twelve weeks.
Check lamotrigine titration, interacting medicines and new rash or mucosal symptoms, and instruct the person not to restart after interruption without dose advice.
Review the working longitudinal diagnosis as collateral, records and future episodes add evidence, and communicate any revision across care settings.

10Special situationsVariants, exceptions and circumstances that change the usual approach.

Cross-section cannot classify course

No single depressive symptom proves bipolarity; the defining evidence lies in previous or future episodes of pathological elevation.

Hypomania may be valued

People may omit productive or pleasurable activation unless asked about sleep, pace and consequences in neutral behavioural language.

Mixed states need direct enquiry

A visibly depressed patient can have racing thoughts and increased capability, so energy and sleep questions remain essential.

Family history changes probability

A first-degree bipolar history should deepen assessment but cannot substitute for an episode history in the patient.

Formulation can stay provisional

Naming uncertainty and its review plan is safer than repeatedly switching categorical diagnoses without documenting the evidence.

11Common pitfallsFrequent interpretation and management errors.

01
Assuming recurrent depression is unipolar without asking about lifetime overactivity and disinhibition.
02
Diagnosing bipolar disorder solely from mood lability, family history or an online screening score.
03
Treating antidepressant-associated agitation as definitive hypomania without assessing the full syndrome and exposure timeline.
04
Waiting for diagnostic certainty before addressing suicide, psychosis, nutrition or mixed activation.
05
Using antidepressant monotherapy automatically for established bipolar depression instead of the recommended bipolar pathway.
06
Failing to seek records or collateral because the patient does not recall a previous activated period.
07
Overlooking thyroid, sleep, neurological, perinatal, medicine and substance contributors to the current depression.