“Mood disorder” is a category, not a diagnosis. Under that umbrella sit unipolar depression, bipolar I, bipolar II, cyclothymia, premenstrual dysphoric disorder, disruptive mood dysregulation in kids, and a handful of less common variants. From a distance they look similar enough that people use the words interchangeably. Up close they’re different illnesses with different treatments, and getting the diagnosis wrong is one of the more expensive mistakes in psychiatry.
The shared thread is that the system regulating your emotional baseline isn’t holding the line. For some people that means a flat, gray weight that won’t lift. For others it means the baseline keeps sliding into highs and lows that don’t match anything happening in their actual life. The categories exist because the patterns repeat. Once you’ve seen a few thousand patients, you recognize which mood story belongs to which illness, and that recognition changes what you prescribe.
Patients almost never walk in with the right label. They walk in with “I’ve been depressed for years” or, increasingly, “I’m pretty sure I have bipolar.” The job of the first appointment is figuring out which of these illnesses we’re actually looking at.
The spectrum, in plain English
Major depressive disorder is the one most people picture. Two weeks or more of low mood, loss of interest, sleep changes, appetite changes, slowed thinking, sometimes suicidal thinking. No history of mania or hypomania. Roughly 8% of US adults in any given year. Responds to SSRIs, SNRIs, therapy, and, in stubborn cases, things like bupropion, mirtazapine, ketamine, or ECT.
Bipolar I is depression plus full mania. Mania is a week or more (or any duration if it puts you in the hospital) of dramatically elevated or irritable mood, decreased need for sleep, racing thoughts, pressured speech, grandiosity, and impulsive behavior that’s genuinely out of character. Real mania is unmistakable when you’ve seen it. People max out credit cards, drive across state lines, start three businesses, sleep two hours and feel rested. About 1% lifetime prevalence.
Bipolar II is depression plus hypomania. Hypomania is the milder cousin: four days or more of elevated mood and energy, but not enough to wreck your life or land you in a psych unit. Patients usually love their hypomanias and don’t report them. They come in for the depressions. This is the diagnosis that gets missed the most, and missing it is how people end up on five years of antidepressants with no real improvement.
Cyclothymia is the low-amplitude, long-duration version. Two years of mood swinging between mild depressive symptoms and mild hypomanic symptoms, never quite hitting threshold for either. Chronic, exhausting, and because nothing ever looks dramatic, often untreated for a decade.
PMDD is its own beast. Severe mood symptoms, irritability, depression, and anxiety that appear in the luteal phase of the menstrual cycle and resolve within a few days of bleeding starting. Affects roughly 5% of menstruating people. Distinct disorder with its own treatment algorithm, usually SSRIs (often dosed only luteal-phase) and hormonal options. Disruptive mood dysregulation disorder lives in pediatrics, designed as a category to slow the overdiagnosis of pediatric bipolar that ran wild in the 2000s.
Why the diagnosis actually matters
This is the part of the work that keeps psychiatrists careful. If you give an antidepressant to someone with undiagnosed bipolar disorder, you can flip them into mania, mixed states, or rapid cycling. SSRI monotherapy in bipolar is one of the more reliable ways to destabilize a patient who was previously holding it together. It’s the reason I spend twenty minutes on family history and mood history at the first appointment before I write anything.
The flip side is also real. People with classic unipolar depression don’t need lithium, and putting them on lithium because somebody got cautious is its own kind of harm. Lithium has a narrow therapeutic window, requires blood draws, and can mess with your kidneys and thyroid over time. Worth it if you need it. Not worth it if you don’t.
Mania is the diagnostic question every psychiatrist asks, and most patients don’t volunteer the answer.
The distinguishing work happens in the history, not the symptoms-in-the-room. The mood you’re sitting with on appointment day tells me almost nothing. What I want is the longitudinal arc. Have you ever had a stretch of three or more days where you needed almost no sleep and felt fantastic? Have you ever started projects in a frenzy and abandoned them a week later? Has anyone in your family been diagnosed with bipolar disorder, been hospitalized for a manic episode, completed suicide, or been described as “the brilliant uncle who lost everything”? Family history matters more in mood disorders than almost anywhere else in psychiatry. The heritability of bipolar I sits around 70-85%.
Screening tools help, sort of. The MDQ and the HCL-32 catch some bipolar II that would otherwise be missed. They miss plenty too. The art is in the follow-up questions, and the follow-up questions take time, which is part of why a fifteen-minute primary care med-check is a bad setting for sorting this out.
The “I think I have bipolar” appointment
This visit happens at least once a week. Usually it’s somebody in their twenties or thirties who’s been reading online and is convinced they have bipolar II because their mood changes. The honest answer is that most of them don’t.
What I see more often is some combination of ADHD, borderline personality features, trauma, sleep deprivation, cannabis use, or unipolar depression with anxious distress. All of those can produce mood that feels like it’s swinging. The difference is timescale. Bipolar moods come in episodes lasting days to weeks. ADHD and personality-driven mood shifts can rotate in an afternoon. If you went from fine to devastated because your boss sent a curt email, that’s not hypomania crashing into depression. That’s emotional reactivity, which is a different problem with different treatment.
I had a woman last spring, late twenties, who’d been told she had bipolar II by an urgent care provider after a ten-minute visit. She’d been on lamotrigine for two years. We went through her history. No hypomanic episodes she could describe. No family history. The “mood swings” were almost entirely sleep-driven, she was working overnight shifts and a second job. We tapered the lamotrigine, fixed the sleep, treated her underlying anxiety with sertraline, and she’s been steady since.
The other thing worth saying: wanting to have bipolar is a thing too. It can feel like an explanation for years of feeling out of control. Real bipolar is a serious illness with real costs, including a lifetime suicide rate around 15%. The treatment options when someone does have it are genuinely good. When somebody doesn’t, the label can stick for decades and shape decisions about jobs, relationships, and pregnancy.
What treatment actually looks like
SSRIs and therapy first
Sertraline 50-200mg, escitalopram 10-20mg, fluoxetine 20-60mg. Four to six weeks before you feel anything real. CBT alongside if you can swing it. Augment with bupropion or mirtazapine if first-line doesn’t hold.
Mood stabilizers, not antidepressants
Lithium remains the gold standard for bipolar I, with mortality data nothing else touches. Lamotrigine for bipolar II depression. Quetiapine, lurasidone, and valproate fill the rest of the menu. Antidepressants only with a stabilizer on board.
SSRI, often luteal-only
Sertraline or fluoxetine dosed continuously or only during the luteal phase. Drospirenone-containing OCPs help some patients. GnRH agonists for severe refractory cases. CBT helps the irritability piece.
Across all of these, sleep is non-negotiable. Sleep deprivation triggers manic episodes in bipolar patients more reliably than almost any other variable. Alcohol use makes every mood disorder worse and complicates every medication. Cannabis is a coin flip; some patients are fine, others get destabilized in ways they don’t connect to the weed until somebody points it out.
Therapy belongs in the picture for almost everyone. CBT for depression has 40 years of data behind it. Interpersonal and social rhythm therapy (IPSRT) was built specifically for bipolar, and the rhythm part matters: bedtime, wake time, meals, and social anchors on a regular schedule keep mood more stable than almost anything else. DBT helps when emotional reactivity is the real engine.
The last thing worth saying. Mood disorders mostly don’t resolve and stay resolved. They remit. They come back. The goal of treatment isn’t to cure them, it’s to make the remissions longer and the relapses shorter and less destructive. Patients who accept that arc do better than patients who keep trying to be done with the illness. Stable on medication for ten years is a better life than four episodes and three hospitalizations off medication. That tradeoff is most of the work.