Bipolar Disorder: Getting the Diagnosis Right
Bipolar disorder takes an average of several years to diagnose correctly, and people frequently collect two or three other diagnoses along the way. That is not carelessness. It follows directly from how the condition presents.
Understanding why it is missed is the most useful thing in this article, because the same reasoning tells you what to look for.
🔍 The great majority of people with bipolar disorder first come for help during a depressive episode, not a manic one. Depression is what hurts, so it is what gets reported. The highs are experienced as feeling well, feeling productive, finally being themselves — nobody makes an appointment about that, and many people do not consider it part of an illness at all. So the doctor sees depression, treats depression, and the underlying condition is not identified until something forces it into view, sometimes years later.
⚠️ Why the misdiagnosis matters
If bipolar depression were simply depression under a different name, the delay would be an administrative problem. It is not, because the treatment differs.
Antidepressants given on their own to someone with bipolar disorder can push them into a manic or mixed state, and can establish a pattern of more frequent episodes that then persists. Guidelines are consequently cautious about antidepressant use in bipolar disorder and clear that it should not be used alone in bipolar I.
So a misdiagnosis is not a delay in reaching the right treatment. It is a period of receiving a treatment that may actively worsen the course of the illness.
There is a further consequence worth stating. Someone whose depression does not respond to one antidepressant after another is often labelled as having treatment-resistant depression and moved through progressively more aggressive options. In a proportion of those people the depression is not resistant — it is bipolar, and it is being treated with the wrong class of drug.
🚩 Clues that a depression may be bipolar
None of these is diagnostic alone. Several together should prompt the question.
| Feature | Why it points that way |
|---|---|
| First episode before the mid-twenties | Bipolar depression begins earlier on average than unipolar |
| Family history of bipolar disorder | The strongest single risk factor; heritability is high |
| Onset after childbirth | The postpartum period is a recognised trigger for a first bipolar episode |
| Sleeping and eating more, not less | Excessive sleep and increased appetite are commoner in bipolar depression than in unipolar |
| Marked physical slowing | Heavy limbs, moving and speaking slowly |
| Psychotic features | More frequent in bipolar depression |
| Many short episodes | Repeated distinct episodes rather than one long continuous one |
| No response to several antidepressants | Worth reconsidering the diagnosis rather than escalating |
| Becoming agitated, irritable or sleepless on an antidepressant | One of the most important clues, and frequently recorded as a side effect rather than a signal |
| A seasonal pattern | Commoner in bipolar disorder |
📊 The forms it takes
| Defined by | Note | |
|---|---|---|
| Bipolar I | At least one full manic episode, lasting a week or requiring hospital admission. Depression usually occurs but is not required for the diagnosis | Mania may include psychosis and typically causes obvious disruption |
| Bipolar II | At least one hypomanic episode of several days, plus at least one major depressive episode. Never full mania | Not a milder illness. The depressive burden is greater, and the risk to life is comparable |
| Cyclothymia | Two years or more of fluctuating high and low periods that never reach full episodes | Chronic, wearing, and frequently mistaken for a personality trait |
The middle row corrects a widespread assumption. Bipolar II is often described as bipolar-lite because the highs are less dramatic, and people with it spend considerably more of their lives depressed than people with bipolar I. Treating it as the minor version leads to under-treatment of the phase that causes most of the harm.
🙂 The episode nobody reports
Hypomania is the diagnostic key and it is almost never volunteered, because it does not feel like illness.
What it involves: needing markedly less sleep and not feeling tired, more energy, faster thoughts, more talkativeness, more confidence, more sociability, more ideas, more spending, more risk-taking, and often genuine productivity. It lasts days rather than hours and represents a clear change from the person’s usual self.
Asked whether they have ever had a period of feeling unusually elevated, many people say no, entirely sincerely. Asked whether there has ever been a period of a few days when they needed far less sleep than usual and still had energy, when they got a great deal done, when other people commented that they seemed different or asked them to slow down — the answer is often yes.
👥 This is why an account from someone else changes the diagnosis more often than any test does. Partners, parents and close friends notice hypomania clearly — the sleeplessness, the talking, the plans, the spending, the irritability when interrupted. The person experiencing it notices feeling well. Bringing someone who has known you for years to an assessment, or giving permission for them to be contacted, is the single most useful practical step in getting this right. It is also worth writing down what happened in specific terms, since the details are hard to recall accurately in a consulting room.
⚡ Mixed states, which are the dangerous ones
The picture of alternating highs and lows is incomplete. Features of both can occur together, and this is common.
What it looks like: profound low mood and hopelessness alongside agitation, racing thoughts, irritability, restlessness and an inability to sit still — despair with energy behind it.
Two things make this state matter more than any other.
It carries the highest risk to life. Depression removes the wish to continue and the energy of the mixed state removes the inertia that would otherwise prevent acting on it. This combination is the most dangerous in the whole condition.
Antidepressants tend to make it worse, adding activation to a state that already has too much.
Anyone who is deeply depressed and simultaneously agitated, wired, unable to be still, with thoughts racing, needs urgent assessment rather than a routine appointment.
🔄 Rapid cycling
Four or more distinct episodes within a year is described as rapid cycling. It is a pattern rather than a separate illness, it can come and go, and it is harder to treat.
Three associations are worth knowing because two are correctable: antidepressant exposure, an underactive thyroid, and alcohol or drug use. Anyone whose episodes have become frequent should have thyroid function checked and their antidepressant reviewed before the pattern is accepted as intrinsic.
🧭 What it gets confused with
| Condition | How it differs |
|---|---|
| Emotionally unstable personality | Mood shifts within hours and are usually triggered by something interpersonal — a rejection, an argument. Bipolar episodes last days to weeks and are largely autonomous, arriving without a trigger. Both can be present in the same person |
| ADHD | Constant since childhood rather than episodic. Distractibility, restlessness and impulsivity are the baseline rather than a change from it |
| Substance use | Stimulants produce states closely resembling mania, and withdrawal resembles depression. A period of abstinence is often needed before the picture is clear |
| Thyroid disease | Overactivity mimics mania, underactivity mimics depression. A simple blood test |
| Steroid medication | Can cause both elevated and depressed states, occasionally severe |
The first row causes the most trouble in practice, and the distinguishing question is duration and trigger: hours and reactive on one side, days to weeks and autonomous on the other.
💊 What the treatments are for
Different drugs address different phases, and matching them is most of the skill.
| Phase | Typically used |
|---|---|
| Acute mania | An antipsychotic, lithium, or valproate — sometimes in combination |
| Bipolar depression | Quetiapine, lamotrigine, certain other antipsychotics. Not an antidepressant alone |
| Long-term prevention | Lithium has the strongest evidence; quetiapine and others are alternatives |
| Preventing depressive episodes specifically | Lamotrigine, which works well downward and poorly upward |
Lithium
Lithium has been in use for decades and remains the reference treatment, for a reason that no other option matches: it is the only psychiatric medication with strong evidence for reducing suicide risk, an effect that appears to be separate from its effect on mood episodes. In a condition where risk to life is the central danger, that is not a small distinction.
It works well for prevention of both directions and for mania, and somewhat less well for acute depression.
What it requires in return is monitoring, and this is not optional.
- Blood levels checked regularly, because the gap between an effective and a toxic level is narrow
- Kidney function, thyroid function and calcium, since long-term use affects all three
- Awareness of what raises the level — anti-inflammatory painkillers, several blood pressure medications, dehydration, vomiting, diarrhoea, a low-salt diet, hot weather and heavy sweating
🔴 Lithium toxicity is a medical emergency and its early signs are easy to dismiss. A coarse trembling of the hands, vomiting and diarrhoea, unsteadiness, slurred speech, blurred vision, drowsiness and confusion. It develops readily during any illness causing dehydration, and when an anti-inflammatory painkiller is bought over the counter and taken alongside — a genuinely common route. Anyone taking lithium should tell every prescriber and pharmacist, avoid anti-inflammatory painkillers unless specifically approved, drink more in hot weather and during illness, and seek urgent help for those symptoms rather than waiting.
Valproate
Effective for mania and for prevention, and carrying one restriction that overrides everything else.
🔴 Valproate must not be taken by anyone who could become pregnant unless there is no alternative and strict conditions are met. Exposure in pregnancy causes physical malformations in around one in ten babies and neurodevelopmental problems in a substantially higher proportion — among the highest risks of any medicine in routine use. Regulators in multiple countries have imposed formal requirements including highly effective contraception, annual specialist review and signed acknowledgement. If you are a woman of childbearing age and have been given valproate without that conversation having happened, raise it before your next dose.
Lamotrigine
Particularly useful for preventing depressive episodes, which matters given how much of the illness is spent depressed. It does little for mania.
Two practical points determine whether it succeeds.
The dose must be increased very slowly over several weeks. This is not caution for its own sake — rapid increases raise the risk of a rare but serious skin reaction. Any rash appearing while starting or increasing lamotrigine means stopping and seeking advice the same day, not waiting to see whether it settles.
Two interactions change the dose substantially. Valproate roughly doubles lamotrigine levels, so the dose is halved when they are combined. Oestrogen-containing contraceptives roughly halve lamotrigine levels, so starting or stopping the pill requires a dose adjustment — and the hormone-free week each month causes levels to swing, which some people notice as a cyclical pattern in how they feel.
Antipsychotics
Several are effective for mania, some for bipolar depression, and some for long-term prevention. They generally work faster than lithium in acute mania.
Their main cost is metabolic — weight gain, raised blood sugar and lipids — which varies considerably between them and needs monitoring rather than being discovered years later. Movement side effects and sedation vary too. Choosing among them is largely a matter of matching the side effect profile to the person.
😴 Sleep is both trigger and warning
The relationship between sleep and this condition is unusually strong and unusually useful.
Sleep loss can trigger an episode. A short night from travel, a new baby, shift work or a deadline can precipitate mania in someone susceptible. This is one of the more reliable triggers there is.
Reduced need for sleep is often the first sign of one. Not insomnia — that is lying awake wanting to sleep. This is sleeping four hours and waking up energetic. That distinction is the earliest reliable warning available, and it usually appears days before anything else is obvious.
Two things follow. Protecting sleep is an active part of treatment rather than general advice. And a change in sleep need is worth treating as a signal requiring action — contacting whoever manages the treatment, not waiting to see.
📅 What holds it stable
Regular daily rhythm. Consistent times for waking, sleeping, eating and activity reduce relapse measurably. This is a genuine intervention with evidence behind it rather than generic advice.
Knowing your own early signs, written down while well. They tend to be consistent for each person — more emails at night, more spending, more ideas, more irritability, a particular kind of restlessness — and are far easier to recognise in a list made in advance than to notice while they are happening.
An agreed plan for what happens when those signs appear: who is told, what changes, who is contacted. Made when well, followed when not.
Alcohol and drug use addressed, since both worsen the course substantially and interact with treatment.
Involving someone close, with permission. The article for family members covers that side.
🛑 The reason treatment stops
Stopping medication when well is the commonest cause of relapse, and it happens for understandable reasons: feeling recovered, side effects, missing the energy of the highs, disliking being someone who takes medication, or simply that a period of stability makes the illness feel unreal.
Two things are worth knowing before that decision is made.
Stopping abruptly is much worse than stopping gradually. This is particularly marked with lithium, where sudden discontinuation carries a sharply raised risk of relapse in the following weeks.
Restarting may not work as well. There is evidence that some people who stop lithium and relapse do not respond as well when it is restarted, which makes an impulsive stop a decision with lasting consequences.
Wanting to stop is a legitimate thing to raise. Doing it alone and quickly is where the harm lies.
🩺 When to seek help
- Thoughts of ending your life, or a plan — immediately, and this is not something to sit on
- Deep depression with agitation and racing thoughts — the mixed state, urgently
- Needing far less sleep than usual and feeling energetic — an early warning, act on it
- Behaviour with lasting consequences — spending, driving, sexual risk, resigning a job
- Losing touch with reality — beliefs others do not share, hearing things
- Becoming agitated or sleepless after starting an antidepressant — report it rather than enduring it
- Coarse tremor, vomiting, unsteadiness or confusion on lithium — possible toxicity, urgent
- Any rash while starting or increasing lamotrigine — same day
- Taking valproate and able to become pregnant, without having had the risk discussion
- Depression that has not responded to several antidepressants — worth revisiting the diagnosis
Treatment is stocked in the mood stabiliser and antidepressant category, including Lithobid (lithium), and in the antipsychotic category, including Seroquel (quetiapine) and Zyprexa (olanzapine). None of these is a medicine to start, change or stop without a prescriber who knows the full history and can arrange the necessary monitoring.
❓ Frequently asked questions
Why does bipolar disorder take years to diagnose?
Because most people first seek help during depression, not a high. Depression is what hurts, so it is what gets reported, while the elevated periods feel like wellness, productivity or finally being oneself — nobody makes an appointment about that. The doctor sees depression and treats depression, and the underlying pattern stays hidden until something forces it into view.
Why does the misdiagnosis matter so much?
Because the treatment differs. Antidepressants given alone to someone with bipolar disorder can push them into a manic or mixed state and can establish a pattern of more frequent episodes that persists. So a misdiagnosis is not merely a delay in reaching the right treatment — it is a period of receiving one that may worsen the course of the illness.
What suggests my depression might be bipolar?
Several features together: onset before the mid-twenties, a family history of bipolar disorder, onset after childbirth, sleeping and eating more rather than less, marked physical slowing, psychotic features, many short episodes rather than one long one, a seasonal pattern, no response to several antidepressants, and becoming agitated, irritable or sleepless after starting one. None is diagnostic alone.
I have never felt euphoric. Can I still have it?
Yes, and the question is often asked the wrong way. Instead of whether you have felt unusually elevated, ask whether there has ever been a period of several days when you needed far less sleep and still had energy, got a great deal done, felt unusually confident, spent more, or when other people commented that you seemed different or asked you to slow down. Many people answer no to the first and yes to the second.
Is bipolar II a milder illness?
No, and the assumption leads to under-treatment. The highs are less dramatic, and people with bipolar II spend considerably more of their lives depressed than people with bipolar I, and the risk to life is comparable. Treating it as the minor version means under-treating the phase that causes most of the harm.
What is a mixed state and why is it dangerous?
Deep depression and hopelessness together with agitation, racing thoughts, irritability and an inability to sit still — despair with energy behind it. It carries the highest risk to life in the whole condition, because depression removes the wish to continue while the energy removes the inertia that would prevent acting. Antidepressants tend to make it worse. It needs urgent assessment.
How is this different from borderline personality disorder?
Chiefly by duration and trigger. Mood shifts in emotionally unstable personality happen within hours and are usually set off by something interpersonal, such as a rejection or an argument. Bipolar episodes last days to weeks and are largely autonomous, arriving without a trigger. Both can be present in the same person, which is why the distinction is drawn on pattern rather than on intensity.
Why is lithium still used after so many decades?
Because it does something nothing else matches: it is the only psychiatric medication with strong evidence for reducing suicide risk, an effect that appears separate from its effect on mood episodes. In a condition where risk to life is the central danger, that is decisive. It also prevents episodes in both directions well. What it requires in return is regular blood level, kidney, thyroid and calcium monitoring.
What raises lithium to dangerous levels?
Anti-inflammatory painkillers bought over the counter, several blood pressure medications, dehydration, vomiting or diarrhoea, a low-salt diet, hot weather and heavy sweating. Warning signs of toxicity are a coarse hand tremor, vomiting and diarrhoea, unsteadiness, slurred speech, blurred vision, drowsiness and confusion — a medical emergency needing urgent help rather than waiting.
Why the warnings about valproate?
Because exposure in pregnancy causes physical malformations in around one in ten babies and neurodevelopmental problems in a substantially higher proportion — among the highest risks of any medicine in routine use. Regulators in several countries now require highly effective contraception, annual specialist review and signed acknowledgement before it is used by anyone who could become pregnant. If that conversation has not happened, raise it before your next dose.
Why is lamotrigine increased so slowly?
Because rapid increases raise the risk of a rare but serious skin reaction, so any rash while starting or increasing it means stopping and seeking advice the same day. Two interactions also change the dose substantially: valproate roughly doubles lamotrigine levels, so the dose is halved alongside it, and oestrogen-containing contraceptives roughly halve them, so starting or stopping the pill requires an adjustment.
Can I stop treatment once I have been well for a long time?
It is a legitimate thing to raise and a bad thing to do alone. Stopping abruptly is much worse than tapering, and with lithium in particular sudden discontinuation carries a sharply raised risk of relapse in the following weeks. There is also evidence that some people who stop lithium and relapse respond less well when it is restarted, which makes an impulsive stop a decision with lasting consequences.
📑 Sources and editorial
- DSM-5 criteria for bipolar I disorder, bipolar II disorder and cyclothymia
- NICE guideline on the assessment and management of bipolar disorder
- CANMAT and ISBD guidelines for the management of bipolar disorder
- Studies of diagnostic delay and of misdiagnosis as unipolar depression
- Evidence on antidepressant-associated mood switching and cycle acceleration
- Meta-analyses of lithium and suicide risk reduction
- Literature on lithium toxicity, monitoring requirements and interacting medicines
- Regulatory guidance on valproate use in women of childbearing potential, including malformation and neurodevelopmental risk data
- Prescribing information for lamotrigine, including titration schedules and interactions with valproate and oestrogen-containing contraceptives
- Research on sleep disruption as a trigger and prodromal sign of manic episodes
- Trials of interpersonal and social rhythm therapy in relapse prevention
- Related reading: advice for family members, types of depressive state
- Related products: mood stabiliser category
- RXshop Editorial Team — reviewed by Thomas Walsh, MD, Psychiatrist and Mental Health Specialist
Medical Disclaimer: The information in this article is for educational and informational purposes only and is not a substitute for professional medical advice, diagnosis, or treatment. Always seek guidance from a qualified healthcare provider with any questions you may have regarding a medical condition, and before starting, stopping or changing any medication.