Almost everything written for the public about bipolar disorder is about mania. Almost everything people with bipolar disorder actually live through is depression. That single mismatch explains the diagnostic delay, the wrong prescriptions, and the fact that most people arrive at the right answer six or nine years after the first episode.

The short answer: bipolar disorder is a recurrent mood disorder defined by episodes of mania or hypomania, usually alongside episodes of depression. Two prospective studies that followed people for more than a decade found that people with bipolar I were symptomatic 47.3 percent of weeks — 31.9 percent of them depressed and 8.9 percent manic or hypomanic — and people with bipolar II were symptomatic 53.9 percent of weeks, 50.3 percent of them depressed. It takes a median of about seven years from first symptoms to diagnosis. Lithium remains the first-line long-term medication and the best-evidenced one; structured psychological treatment added to medication cuts recurrence meaningfully; and most of the excess deaths in bipolar disorder are from physical illness rather than suicide, which makes ordinary medical care part of the treatment rather than an afterthought.¹ ² ³ ⁴ ⁵

Four doors

What it actually is

Mania is a distinct period of abnormally elevated, expansive or irritable mood with increased energy and goal-directed activity, lasting at least a week or requiring hospitalisation, causing marked impairment and sometimes psychosis. Hypomania is the same shape at lower intensity — at least four consecutive days, noticeable to others, but without marked impairment, psychosis or hospitalisation.

  • Bipolar I requires one manic episode, ever. Depression is nearly always part of the picture but is not required for the diagnosis.
  • Bipolar II requires at least one hypomanic episode and at least one major depressive episode, and no manic episode.

One manic episode makes it bipolar I regardless of everything else that follows. That is the whole of the definitional difference, and it is a much narrower difference than the way the two are usually described. The full comparison →

How common. Across eleven countries and 61,392 adults, lifetime prevalence was 0.6 percent for bipolar I, 0.4 percent for bipolar II, 1.4 percent for subthreshold bipolar and 2.4 percent for the spectrum as a whole. In a nationally representative US survey of 9,282 adults the figures were 1.0 percent, 1.1 percent and 2.4 percent.⁶ ⁷ Any figure quoted without saying which definition it uses is telling you very little.

The shape of the illness, measured week by week

This is the finding that reframes everything, and it comes from following people for more than a decade with weekly symptom ratings.

Where the weeks go Bipolar I (146 patients, mean 12.8 years)¹ Bipolar II (86 patients, mean 13.4 years)²
Depressed 31.9% 50.3%
Manic or hypomanic 8.9% 1.3%
Cycling or mixed 5.9% 2.3%
Symptomatic in total 47.3% 53.9%
Well 52.7% 46.1%

Two things follow. First, depression is the illness for most people, most of the time — by a factor of about three in bipolar I and about forty in bipolar II. Second, most of that symptomatic time is subsyndromal: in bipolar I, minor depressive and hypomanic symptoms combined were nearly three times more frequent than full episodes.¹ The between-episode weeks are not empty. They are where most of the disability lives, because psychosocial impairment rises with each increment of depressive symptom severity, while subsyndromal hypomanic symptoms are not disabling and may even improve functioning.⁸

A caution about generalising: both cohorts were enrolled at academic centres between 1978 and 1981, during an index episode, and treated with the pharmacology of that era. They describe a more severe population than a community sample.

Recovery, and the gap inside it

After a first hospitalisation for mania, 166 patients were followed for two to four years. Ninety-eight percent achieved syndromal recovery — no longer meeting criteria. Seventy-two percent achieved symptomatic recovery. Forty-three percent achieved functional recovery, defined as regaining their previous work and living situation.⁹

That 55-point gap between "no longer ill" and "life back" is the same gap that shows up across serious mental illness, and it is the reason a treatment plan aimed only at episodes is aimed at half the problem.

Recurrence is common even under good treatment. In the largest US effectiveness study, of 858 people who recovered, 48.5 percent had a recurrence within two years — more than twice as many depressive as manic recurrences. The strongest predictor was residual mood symptoms at the point of recovery.¹⁰ Which is a practical instruction: "mostly better" is a treatment target, not a finishing line.

The delay

A systematic review of 59 studies and more than 40,000 people found a median delay to diagnosis of 6.7 years, a median delay in help-seeking of 3.5 years, and a median duration of untreated bipolar disorder of 5.9 years. A larger 2025 meta-analysis put the pooled mean duration of undiagnosed or untreated illness at 9.1 years.³ ¹¹ The authors of the first put the consequence plainly: the peak age of onset is 15 to 25, and diagnosis and guideline-recommended treatment are likely to be delayed until 25 to 35.³

The mechanism is not mysterious. People seek help when depressed, not when hypomanic; depression as the polarity of the first episode predicts a longer delay.¹¹ What to bring to an assessment, and why the questionnaires don't work →

What treats it

Medication is the foundation, and lithium is still the reference point. Across five trials and 770 participants, lithium beat placebo at preventing all relapses (relative risk 0.65, 95% CI 0.50–0.84) and manic relapses (0.62, 0.40–0.95), with a smaller and less certain effect on depressive relapses (0.72, 0.49–1.07).¹² A network meta-analysis of 33 maintenance trials in 6,846 participants concluded that "lithium should remain the first-line treatment when prescribing a relapse-prevention drug in patients with bipolar disorder, notwithstanding its tolerability profile."⁴ The UK guideline says the same in one line: "Offer lithium as a first-line, long-term pharmacological treatment for bipolar disorder."¹³

Lithium is also the drug with the most evidence on suicide: across 48 randomised trials and 6,674 participants, it reduced suicides (odds ratio 0.13, 95% CI 0.03–0.66) and deaths from any cause (0.38, 0.15–0.95) compared with placebo.¹⁴ That evidence is not unanimous — a randomised trial in 519 veterans who had recently had a suicide-related event was stopped for futility, finding no difference when lithium was simply added to existing care.¹⁵ The reasonable reading is that lithium's protective effect is real and works largely through preventing episodes, not that adding it late to a complicated regimen rescues an acute situation.

It requires monitoring, and the monitoring is not optional: blood levels, kidney function, thyroid function and calcium on a defined schedule.¹³ The questions to ask a prescriber →

Structured psychological treatment, added to medication, works. A component network meta-analysis of 39 trials and 3,863 participants found that manualised adjunctive therapies were associated with lower recurrence than control treatments (odds ratio 0.56, 95% CI 0.43–0.74), and — the finding worth acting on — that psychoeducation with guided practice of illness-management skills delivered in a family or group format beat the same content delivered individually (odds ratio 0.12, 95% CI 0.02–0.94).⁵ No modality was established as superior to another. What the psychotherapies actually do →

Sleep and rhythm are treatment, not lifestyle advice. In the trial that established interpersonal and social rhythm therapy, the people assigned to it in the acute phase survived longer without a new episode regardless of what they received afterwards, and their social rhythms were more regular.¹⁶ The printable mood and sleep log →

The part that gets missed

Bipolar disorder shortens life by about nine years for women and eight and a half for men, and the biggest single contributor is not what people assume. All-cause mortality is roughly doubled; the suicide standardised mortality ratio is around 14, applied to a small denominator, while the natural-cause ratio of about 1.6 applies to a very large one.¹⁷ ¹⁸ In a 23-year Swedish hospital cohort, "for bipolar disorder, most excess deaths were from natural causes."¹⁹

And there is a finding inside the Swedish national data that reads like an instruction. The association between bipolar disorder and death from chronic physical disease was weaker among people who already had a diagnosis of that physical condition than among those who did not — which the authors read as "suggesting that better provision of primary medical care may effectively reduce premature mortality among persons with bipolar disorder."¹⁸

So: a primary care doctor, blood pressure, cholesterol, glucose, smoking, and the metabolic effects of whatever you are taking, are part of bipolar treatment. Nobody says this loudly enough.

Getting treated in California

  • Ask for the assessment the guidelines describe — a full history of mood, of episodic changes in behaviour, of symptoms between episodes, family history, and a corroborative account from someone who knows you.²⁰ Questionnaires are explicitly not the way this is identified.²¹
  • Ask what the long-term plan is, not only what is being prescribed this month. Relapse prevention, early warning signs and a written plan are the guideline standard.¹³
  • Get the physical care. A primary care doctor who knows the diagnosis, and annual metabolic checks, are part of the treatment on the mortality evidence above.
  • If your plan denies a level of care, that is appealable and the review is free. Appeal a denial → · If nobody in-network has an opening →
  • If cost is the obstacle, county and sliding-scale routes exist and neither requires exhausting the other. What a sliding scale is → · What changed at your county →
  • Free peer support exists and is not a substitute for treatment. The Depression and Bipolar Support Alliance runs no-cost peer groups across California, online and in person. It costs nothing to try one while a referral is pending.
  • Know which prescriber you are seeing. A psychiatrist and a psychiatric nurse practitioner can both prescribe and monitor mood stabilisers in California. What a psychiatric NP is →
  • If a crisis is the immediate problem, 988 is free and 24/7, and there are routes that do not run through the police. Crisis without police →

Q&A

Q: Is bipolar disorder mostly mania? A: No. In the long prospective studies, people with bipolar I spent 31.9 percent of weeks depressed and 8.9 percent manic or hypomanic; people with bipolar II spent 50.3 percent depressed and 1.3 percent hypomanic.¹ ² The public image of the illness is drawn from its least common state.

Q: How long does it take to get diagnosed? A: A median of about 6.7 years to diagnosis across 59 studies, with a pooled mean of about 9.1 years for the duration of undiagnosed or untreated illness in a later meta-analysis.³ ¹¹

Q: Do I have to take medication forever? A: That is a decision for you and a prescriber, and it deserves the real numbers on both sides. What the evidence says about stopping is specific: after discontinuing stable lithium maintenance, more than half of new episodes occurred within ten weeks, and the risk of early recurrence — especially of mania — exceeded what the untreated course would predict.²² Rapid discontinuation is the specific hazard. Nothing here is a reason to stop anything on your own.

Q: Can therapy replace medication? A: The evidence base is for psychotherapy added to medication; that is how every trial was designed.⁵ It is a real addition, not a substitute.

Q: What about bipolar and ADHD, or bipolar and borderline personality disorder? A: Both overlap and both are commonly confused with it. Around one in five people with bipolar disorder also meets criteria for borderline personality disorder, and the differential has features that genuinely separate them. BPD or bipolar? → · The adult ADHD map →

Q: Is bipolar disorder genetic? A: Family history raises risk substantially, and it is one of the things a proper assessment asks about.²⁰ It is not deterministic, and a family history of bipolar disorder is not by itself a reason to diagnose it — the UK guideline says so explicitly for young people.²³

Ready to find a therapist who works alongside a prescriber? Filter by approach, schedule and payment route → · What to ask on the consult call →

In crisis? Call or text 988 — free, 24/7.

Sources

  1. Judd LL, Akiskal HS, Schettler PJ, et al., "The long-term natural history of the weekly symptomatic status of bipolar I disorder," Archives of General Psychiatry 59(6), 2002, 530–537 — 146 patients, mean 12.8 years; "symptomatically ill 47.3% of weeks… Depressive symptoms (31.9% of total follow-up weeks) predominated over manic/hypomanic symptoms (8.9% of weeks) or cycling/mixed symptoms (5.9% of weeks)" — europepmc.org.
  2. Judd LL, Akiskal HS, Schettler PJ, et al., "A prospective investigation of the natural history of the long-term weekly symptomatic status of bipolar II disorder," Archives of General Psychiatry 60(3), 2003, 261–269 — 86 patients, mean 13.4 years; "symptomatic 53.9% of all follow-up weeks: depressive symptoms (50.3% of weeks) dominated the course over hypomanic (1.3% of weeks) and cycling/mixed (2.3% of weeks)" — europepmc.org.
  3. Scott J, Graham A, Yung A, Morgan C, Bellivier F, Etain B, "A systematic review and meta-analysis of delayed help-seeking, delayed diagnosis and duration of untreated illness in bipolar disorders," Acta Psychiatrica Scandinavica 146(5), 2022, 389–405 — 59 studies, more than 40,000 individuals; "The median DHS, DD and DUB were 3.5…, 6.7… and 5.9 years"; "diagnosis and guideline recommended interventions… are likely to be delayed until age 25-35 years" — europepmc.org.
  4. Miura T, Noma H, Furukawa TA, et al., "Comparative efficacy and tolerability of pharmacological treatments in the maintenance treatment of bipolar disorder: a systematic review and network meta-analysis," The Lancet Psychiatry 1(5), 2014, 351–359 — 33 randomised trials, 6,846 participants; "lithium should remain the first-line treatment when prescribing a relapse-prevention drug in patients with bipolar disorder, notwithstanding its tolerability profile" — doi.org.
  5. Miklowitz DJ, Efthimiou O, Furukawa TA, Scott J, McLaren R, Geddes JR, Cipriani A, "Adjunctive psychotherapy for bipolar disorder: a systematic review and component network meta-analysis," JAMA Psychiatry 78(2), 2021, 141–150 — 39 trials, 3,863 participants; "manualized treatments were associated with lower recurrence rates than control treatments (OR, 0.56; 95% CI, 0.43-0.74)"; "Psychoeducation with guided practice of illness management skills in a family or group format was associated with reducing recurrences vs the same strategies in an individual format (OR, 0.12; 95% CI, 0.02-0.94)" — pmc.ncbi.nlm.nih.gov.
  6. Merikangas KR, Jin R, He JP, et al., "Prevalence and correlates of bipolar spectrum disorder in the world mental health survey initiative," Archives of General Psychiatry 68(3), 2011, 241–251 — 61,392 adults in 11 countries; "The aggregate lifetime prevalences were 0.6% for bipolar type I disorder (BP-I), 0.4% for BP-II, 1.4% for subthreshold BP, and 2.4% for BPS" — doi.org.
  7. Merikangas KR, Akiskal HS, Angst J, et al., "Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey replication," Archives of General Psychiatry 64(5), 2007, 543–552 — 9,282 US adults; "Lifetime (and 12-month) prevalence estimates are 1.0% (0.6%) for BP-I, 1.1% (0.8%) for BP-II, and 2.4% (1.4%) for subthreshold BPD" — doi.org.
  8. Judd LL, Akiskal HS, Schettler PJ, et al., "Psychosocial disability in the course of bipolar I and II disorders: a prospective, comparative, longitudinal study," Archives of General Psychiatry 62(12), 2005, 1322–1330 — 158 bipolar I and 133 bipolar II patients followed a mean of 15 years; "At each level of depressive symptom severity, BP-I and BP-II are equally impairing"; "Subsyndromal hypomanic symptoms are not disabling in BP-II, and they may even enhance functioning" — doi.org.
  9. Tohen M, Zarate CA, Hennen J, et al., "The McLean-Harvard First-Episode Mania Study: prediction of recovery and first recurrence," American Journal of Psychiatry 160(12), 2003, 2099–2107 — 166 patients followed 2–4 years; "By 2 years, most subjects achieved syndromal recovery (98%…); 72% achieved symptomatic recovery… Only 43% achieved functional recovery" — doi.org.
  10. Perlis RH, Ostacher MJ, Patel JK, et al., "Predictors of recurrence in bipolar disorder: primary outcomes from the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD)," American Journal of Psychiatry 163(2), 2006, 217–224 — of 858 who recovered, "416 (48.5%) of these individuals experienced recurrences, with more than twice as many developing depressive episodes (298, 34.7%) as those who developed manic, hypomanic, or mixed episodes (118, 13.8%)"; residual symptoms at recovery as the key predictor — doi.org.
  11. Keramatian K, Pinto JV, Tsang VWL, Chakrabarty T, Yatham LN, "Duration of untreated or undiagnosed bipolar disorder and clinical characteristics and outcomes: systematic review and meta-analysis," British Journal of Psychiatry 227(5), 2025, 622–632 — "The pooled mean DUBD across all studies was 9.10 years"; depressive first-episode polarity among the predictors of longer delay — europepmc.org.
  12. Geddes JR, Burgess S, Hawton K, Jamison K, Goodwin GM, "Long-term lithium therapy for bipolar disorder: systematic review and meta-analysis of randomized controlled trials," American Journal of Psychiatry 161(2), 2004, 217–222 — five trials, 770 participants; relative risks 0.65 (all relapses), 0.62 (manic), 0.72 (depressive) — doi.org.
  13. National Institute for Health and Care Excellence, "Bipolar disorder: assessment and management," clinical guideline CG185 — recommendation 1.7.7, "Offer lithium as a first-line, long-term pharmacological treatment for bipolar disorder"; recommendation 1.4.1 on risk management and relapse-prevention plans; recommendations 1.10.14 to 1.10.24 on lithium monitoring — nice.org.uk.
  14. Cipriani A, Hawton K, Stockton S, Geddes JR, "Lithium in the prevention of suicide in mood disorders: updated systematic review and meta-analysis," BMJ 346, 2013, f3646 — 48 randomised trials, 6,674 participants; "Lithium was more effective than placebo in reducing the number of suicides (odds ratio 0.13, 95% confidence interval 0.03 to 0.66) and deaths from any cause (0.38, 0.15 to 0.95)" — doi.org.
  15. Katz IR, Rogers MP, Lew R, et al., "Lithium treatment in the prevention of repeat suicide-related outcomes in veterans with major depression or bipolar disorder: a randomized clinical trial," JAMA Psychiatry 79(1), 2022, 24–32 — stopped for futility after 519 veterans; "No overall difference in repeated suicide-related events between treatments was found (hazard ratio, 1.10; 95% CI, 0.77-1.55)" — doi.org.
  16. Frank E, Kupfer DJ, Thase ME, et al., "Two-year outcomes for interpersonal and social rhythm therapy in individuals with bipolar I disorder," Archives of General Psychiatry 62(9), 2005, 996–1004 — 175 participants; "participants assigned to IPSRT in the acute treatment phase survived longer without a new affective episode (P = .01), irrespective of maintenance treatment assignment" — doi.org.
  17. Hayes JF, Miles J, Walters K, King M, Osborn DP, "A systematic review and meta-analysis of premature mortality in bipolar affective disorder," Acta Psychiatrica Scandinavica 131(6), 2015, 417–425 — 31 studies; all-cause SMR 2.05 (95% CI 1.89–2.23) with I² 96.2%; suicide SMR 14.44; natural-cause SMR 1.64 — pmc.ncbi.nlm.nih.gov.
  18. Crump C, Sundquist K, Winkleby MA, Sundquist J, "Comorbidities and mortality in bipolar disorder: a Swedish national cohort study," JAMA Psychiatry 70(9), 2013, 931–939 — 6,618 people with bipolar disorder in a cohort of 6,587,036; "Women and men with bipolar disorder died 9.0 and 8.5 years earlier on average than the rest of the population"; the weaker association among those with a prior diagnosis of the physical condition, "suggesting that better provision of primary medical care may effectively reduce premature mortality" — doi.org.
  19. Ösby U, Brandt L, Correia N, Ekbom A, Sparén P, "Excess mortality in bipolar and unipolar disorder in Sweden," Archives of General Psychiatry 58(9), 2001, 844–850 — 15,386 people with a hospital diagnosis of bipolar disorder, 1973–1995; "For bipolar disorder, most excess deaths were from natural causes" — doi.org.
  20. National Institute for Health and Care Excellence, clinical guideline CG185, recommendation 1.3.2 — the full assessment, including "symptoms between episodes" and "encourage people to invite a family member or carer to give a corroborative history"; recommendation 1.3.3 on differential diagnoses — nice.org.uk.
  21. National Institute for Health and Care Excellence, clinical guideline CG185, recommendation 1.2.3 — "Do not use questionnaires in primary care to identify bipolar disorder in adults" — nice.org.uk.
  22. Suppes T, Baldessarini RJ, Faedda GL, Tohen M, "Risk of recurrence following discontinuation of lithium treatment in bipolar disorder," Archives of General Psychiatry 48(12), 1991, 1082–1088 — 14 studies, 257 patients; "More than 50% of new episodes of illness occurred within 10 weeks of stopping"; "Risk of early recurrence of bipolar illness, especially of mania, evidently is increased following discontinuation of lithium use and may exceed that predicted by the course of the untreated disorder" — doi.org.
  23. National Institute for Health and Care Excellence, clinical guideline CG185, recommendation 1.11.6 — "Do not make a diagnosis of bipolar disorder in children or young people on the basis of depression with a family history of bipolar disorder but follow them up" — nice.org.uk.

Paid for by participating therapists. Inclusion is computed from availability data — never purchased. No ads, no data sold.