The short answer: Published estimates cluster between roughly 6 and 9 years, and they do not all measure the same thing. A 2022 meta-analysis of 59 studies found a median delay of 6.7 years between the first mood episode and a bipolar diagnosis. A 2025 meta-analysis put the pooled mean time spent undiagnosed or untreated at 9.1 years. The main reason is that depression, not mania, is usually what brings a person to care.

How long is the delay, and where does the 10-year figure come from?

Three separate meta-analyses give the most defensible numbers, and each defines the interval differently.

The widely quoted "10 years" is not from any of those. It traces to a 2003 report in the Journal of Clinical Psychiatry on a 2000 survey of 600 people recruited through National Depressive and Manic-Depressive Association support groups. In that sample, more than one third said they waited 10 years or more for an accurate diagnosis. That is a real published finding, but the respondents were self-selected support group members. It is not a population estimate, and it should not be repeated as one.

The 2022 review also noted the practical shape of the problem. Bipolar disorder peaks in onset between ages 15 and 25, while diagnosis and guideline-recommended treatment usually arrive between 25 and 35.

Why does depression usually get diagnosed first?

People seek help when they feel bad. Depressive episodes cause distress, so they generate appointments. Hypomania often does not.

The diagnostic threshold is also short and easy to miss in hindsight. Under DSM-5 criteria, a manic episode requires at least one week of symptoms, while a hypomanic episode requires only four consecutive days. Four elevated days two years ago rarely comes up in a 15-minute visit unless someone asks directly.

Under-reporting is well documented. In the 2003 survey, respondents acknowledged under-reporting their manic symptoms, and more than half also felt their doctor's limited understanding of bipolar disorder delayed the diagnosis. The 2025 meta-analysis found that a depressive first episode predicted a longer delay, while bipolar I disorder and psychotic symptoms predicted a shorter one. Obvious mania gets noticed. Quiet hypomania does not.

If you are trying to separate the two patterns, see our guides on bipolar versus unipolar depression and the subtle signs of bipolar II.

How many people receive a different diagnosis first?

In the 2003 survey, 69% of respondents reported being misdiagnosed, most often with unipolar depression. Those people consulted an average of four physicians before the correct diagnosis. A 2011 review in L'Encephale reported that over 60% of people with bipolar disorders received between one and four prior diagnoses.

Cross-sectional data point the same way. The BRIDGE study, published in Archives of General Psychiatry in 2011, assessed 5,635 adults who were in a current major depressive episode. Of those, 16.0% met DSM-IV-TR criteria for bipolar disorder, and 47.0% met broader bipolarity specifier criteria.

Follow-up data matter too. A 2017 meta-analysis of 56 prospective studies found that 22.5% of people diagnosed with major depressive disorder later developed bipolar disorder, with the greatest risk in the first five years. Family history of bipolar disorder (odds ratio 2.89) and psychotic symptoms (odds ratio 4.76) predicted that transition. For the depression side of this picture, depression.md covers diagnosis and treatment in more detail.

Do screening questionnaires like the MDQ actually work?

They help, within limits. A 2015 meta-analysis of 53 accuracy studies (21,542 people) reported the following at recommended cutoffs in psychiatric settings: the HCL-32 had 81% sensitivity and 67% specificity, the Mood Disorder Questionnaire (MDQ) had 66% sensitivity and 79% specificity, and the Bipolar Spectrum Diagnostic Scale had 69% sensitivity and 86% specificity.

Performance drops outside specialist care. In that same review, the MDQ at a cutoff of 7 had only 43% sensitivity in primary care or general population samples, with 95% specificity. A 2019 head-to-head meta-analysis of 9 studies found summary sensitivities of 82% for the HCL-32 and 80% for the MDQ, with specificities of 57% and 70%.

The screening yield is still striking. A 2005 study in a family medicine clinic found that 21.3% of patients taking an antidepressant for depression screened positive on the MDQ, and nearly two thirds of those had never been diagnosed with bipolar disorder. In that setting the MDQ had 0.580 sensitivity and 0.930 specificity.

Guidance is cautious about using these tools as diagnostic instruments. NICE guideline CG185 says plainly: do not use questionnaires in primary care to identify bipolar disorder in adults. A positive screen is a reason to ask more questions, not an answer.

Does the delay actually cause harm?

The 2025 meta-analysis found longer undiagnosed or untreated periods were associated with early onset, a depressive first episode, lifetime suicide attempts, comorbid anxiety and alcohol use disorders, and a family history of bipolar disorder. Those are associations, not proof of cause. The same review reported conflicting results for outcomes measured after diagnosis.

The case for acting early is stronger. A 2016 review of 8 primary studies and 2 meta-analyses (8,942 patients) found that treatment earlier in the illness course produced better response, lower relapse rates, longer time to recurrence, better remission, and better psychosocial and employment outcomes. The authors flagged a high risk of bias in most included studies.

The condition itself is not rare or mild. NIMH reports, using National Comorbidity Survey Replication data, that 2.8% of US adults had bipolar disorder in the past year and 4.4% experience it at some point. Among past-year cases, 82.9% had serious impairment.

What shortens the wait?

Four things show up repeatedly in the evidence and the guidelines.

  1. Someone asking the right question. NICE CG185 tells primary care clinicians to ask adults presenting with depression about previous periods of overactivity or disinhibited behaviour, and to consider specialist referral if those periods lasted four days or more.
  2. Collateral history. NICE recommends encouraging people to invite a family member or carer to give a corroborative history. Partners and parents usually remember the elevated weeks that the person does not.
  3. A life-course mood history, not a snapshot. NICE asks clinicians to document mood history, episode triggers, relapse patterns, symptoms between episodes, and family history. Mood charting between visits turns recall into a record.
  4. Treating treatment response as data. In the BRIDGE study, manic or hypomanic states emerging during antidepressant therapy was one of the markers linked to bipolarity. Read more on why antidepressants can worsen bipolar disorder.

Structure helps as well. The 2022 review found that access to early intervention services was associated with a shorter duration of untreated bipolar disorder.

The bottom line

The honest answer is a range, not a single number. Median delay to diagnosis is about 6.7 years in the best pooled data, mean time undiagnosed or untreated is about 9.1 years, and the popular 10-year claim comes from one self-selected support group survey in 2000. Most of that time is spent carrying a depression diagnosis, because depression is what sends people to the clinic and hypomania is what they forget to mention. Bringing a written mood history and a family member to the appointment is the most practical thing a person can do to shorten it.

Last updated: September 2026. This article is for informational purposes only and does not constitute medical advice. Only a qualified clinician can diagnose bipolar disorder, and you should not start or stop any psychiatric medication without talking to your prescriber.