Mood, Anxiety and Emotional Health
It is not all in your head, and it is not only in your head.
Depression and anxiety are real conditions with real biology, and the story most people were told about that biology is out of date. What replaced it is more interesting and more useful. It also means that a proper assessment includes questions about your thyroid, your sleep, your medications, your iron and your heart rhythm, and that skipping those is not being efficient. It is being incomplete.
Begin reading
What feeds into mood
Life, loss and load
Sleep and circadian rhythm
Physical illness
Medication effects
Usually assessed
Symptoms, duration and a questionnaire
Often not assessed
Thyroid function, iron, B12, calcium
Sleep apnoea and insomnia
Every medication and supplement being taken
Alcohol intake, and heart rhythm
A screen for previous periods of elevated mood
None of the second list is exotic and none of it is expensive. It is skipped, and skipping it is how treatable contributors go unfound for years.
Two things are true at once, and most writing on this subject picks one and abandons the other. Depression and anxiety are genuinely biological conditions that respond to treatment. And they are also embedded in a life, a body and a set of circumstances that a fifteen minute appointment cannot see.
Holding both is not a compromise. It is the only position the evidence supports.
This page covers what the serotonin story got wrong, and what replaced it. Why depression is a syndrome rather than a single disease. The medical conditions that present as mood and anxiety disorders. What inflammation and the gut do and do not explain. Which treatments actually have evidence behind them, and what should happen before anyone concludes that treatment is not working.
The central idea
Treatment resistant depression is sometimes untreated something else.
Which is a reason to look further, not a reason to stop what is working.
Free guide
Get Brain on Fire
The guide walks through what belongs in a proper history. Which foundational tests find the contributors that routine visits skip. And how to tell which symptoms need urgent assessment rather than an appointment.
Brain on Fire
The biology underneath mood, and what changes when it is addressed.
Where the popular explanation came from
The chemical imbalance story, and what actually replaced it.
For thirty years people were told that depression is caused by low serotonin. That explanation was always a simplification of a hypothesis, and it was never established. A large 2022 umbrella review concluded the evidence does not support it as the primary cause. This got reported as antidepressants having been debunked, which is not what it says and not what follows from it.
The old story and the current model
The old story
Depression is caused by low serotonin, and antidepressants top it back up
What the current model involves
Neuroplasticity and how readily circuits reorganise
Stress response systems and how they recover
Function across specific brain circuits rather than one chemical
Inflammatory signalling, in a subset of people
What does not follow from it
That antidepressants do not work. Their benefit was measured directly in trials rather than inferred from the theory.
A wrong mechanistic story and a genuinely effective treatment can coexist, and in medicine they frequently have. Aspirin worked for decades before anyone understood why.
What the honest version sounds like: antidepressants outperform placebo for moderate to severe depression, the average benefit is real but more modest than the marketing implied, individual responses vary widely, and nobody can predict in advance who will respond. That is a fair account. It is also a much better argument for measuring outcomes properly and revisiting the diagnosis when nothing moves than the old story ever was.
One label, several conditions
Depression is a syndrome, not a disease.
The diagnosis is made from a list of symptoms. Two people can both meet criteria while sharing almost none of the same experience: one cannot sleep and has lost their appetite, the other sleeps twelve hours and eats constantly. Grouping those together under a single word is defensible as a starting point and indefensible as an endpoint, and it is one reason trial results average out to modest effects when some subgroups do much better than others.
The distinctions that change what happens next are worth naming. Melancholic features, early waking and marked loss of pleasure, respond differently from atypical features. Anxious distress predicts a slower response. Seasonal pattern opens a specific treatment. Peripartum onset has its own risks and its own urgency.
And one distinction matters more than all the others combined, because getting it wrong causes harm rather than merely wasting time.
Anyone being treated for depression should be asked about periods of unusually elevated mood. Not everyone is.
Presentations that differ
Under one diagnosis
Presentations that differ under one diagnosis
Melancholic features: early morning waking and marked loss of pleasure
Atypical features: sleeping and eating more rather than less
Anxious distress: slower to respond to treatment
Seasonal pattern: opens a specific treatment
Peripartum onset: its own risks and its own urgency
The question that changes treatment
Has there ever been a period of several days with markedly reduced need for sleep, racing thoughts, unusual energy, or uncharacteristic risk taking, noticed by other people and not only by you?
A yes points toward the bipolar spectrum, where an antidepressant given alone can destabilise mood. This is a screening question, not a diagnosis, and it belongs in every mood assessment.
When the body is doing the talking
Anxiety, and the conditions that imitate it.
Anxiety produces physical symptoms, which is why it is so often diagnosed from them. The difficulty is that several medical conditions produce exactly the same physical symptoms. Someone experiencing a racing heart, sweating, tremor and a sense of dread cannot tell from the inside which is which. Neither can anyone else without asking.
Several features should prompt a second look. Anxiety that arrives without any psychological context. Anxiety that starts abruptly in someone with no history of it. Anxiety that is strongly episodic, with a clear beginning and end. A first episode after fifty. Or anxiety alongside physical findings such as weight loss, a persistently fast pulse, or new high blood pressure.
This is not an argument that anxiety is usually something else. It usually is not. It is an argument that the check is cheap and the conditions are treatable. Being told to manage stress while an arrhythmia goes unrecorded is a specific and avoidable failure.
Panic attacks that begin after fifty with no history deserve a workup, not a breathing exercise.
Worth excluding
Before settling on anxiety
Worth excluding before settling on anxiety
Conditions that imitate it
Thyroid overactivity
Heart rhythm disturbance, including atrial fibrillation and supraventricular tachycardia
Low blood sugar episodes
Stimulants, including caffeine and decongestants, and withdrawal from alcohol or sedatives
Asthma, anaemia and vestibular disorders
Rare but classic: adrenal tumours producing catecholamines
Features that prompt a look
First onset after fifty
No psychological context at all
Weight loss, a persistently fast resting pulse, or new high blood pressure
Strictly episodic, with a clear beginning and end
Most anxiety is anxiety. The point is that the exclusions are inexpensive, the conditions are treatable, and the cost of missing one is measured in years.
The layer that gets skipped
Physical contributors to mood and anxiety.
None of these is an alternative explanation that replaces psychology. Each is a contributor that can lower the floor, blunt a response to treatment, or produce symptoms in its own right. Several are usually present together. Finding and correcting them does not make the therapy or the medication unnecessary, it makes both work better, and it is the part of the assessment that a short appointment cannot fit.
Both directions, and both mistaken
Hypothyroidism produces low mood, fatigue, slowed thinking and weight gain. Overactivity produces anxiety, irritability, palpitations and insomnia. Subclinical patterns and thyroid autoimmunity add a further layer that a single reading may not capture. Thyroid Health covers what the numbers mean.
Before anaemia appears
Low ferritin is associated with fatigue, poor concentration and low mood, and symptoms can be present while the complete blood count is still normal. It is common in menstruating women, in people with heavy periods, and after pregnancy. A normal haemoglobin is not a reason to skip ferritin.
Deficiency, and the low normal band
Both are associated with depressive symptoms and cognitive change, and neurological effects can precede any blood count abnormality. Metformin, long term acid suppression, restricted diets and alcohol all lower status. Methylmalonic acid and homocysteine help when a level sits in the low normal band.
Strong association, weaker intervention data
Low vitamin D is consistently associated with depression across observational studies. Supplementation trials in people who are not deficient have largely not improved mood. Correcting genuine deficiency is worth doing on its own merits. Presenting vitamin D as a treatment for depression goes past the evidence, and saying so is part of taking the evidence seriously.
Frequently mistaken for depression
Fragmented sleep and repeated overnight oxygen dips produce fatigue, low mood, irritability and poor concentration. Depression and apnoea overlap substantially, and apnoea is often the one that has never been looked for. Snoring, witnessed pauses, unrefreshing sleep and morning headache warrant a sleep study.
Not only a symptom
Insomnia predicts the onset of depression rather than only accompanying it, and treating it directly improves depressive symptoms. Cognitive behavioural therapy for insomnia is the first line treatment and outperforms sedative medication over the longer term. Treating sleep as a downstream detail wastes one of the more effective available interventions.
A review that takes ten minutes
Corticosteroids can produce marked mood change including elevation. Interferon therapy reliably induces depressive symptoms. Some hormonal contraceptives affect mood in a minority of users, which deserves acknowledgement rather than dismissal. Sedatives, opioids and several others contribute. Withdrawal states matter as much as the drugs themselves.
Both are self treatment that backfires
Alcohol is a depressant that fragments sleep and worsens anxiety on the following days through withdrawal. Cannabis is frequently used for anxiety and can worsen it, and heavy use is associated with poorer outcomes in mood disorders. Intake is systematically underreported, so this deserves an actual conversation rather than a tick box.
Perimenopause in particular
The perimenopausal window carries elevated risk of depressive symptoms, including in women with no prior history, and it is regularly attributed to circumstances instead. Premenstrual dysphoric disorder and the postpartum period are the other two windows where timing itself is diagnostic information. Hormone Health covers the physiology.
Bidirectional, and treated as one directional
Pain and depression each increase the risk and severity of the other, and they share overlapping circuitry. Treating one without the other produces partial results in both. The medications used for pain add sedation and cognitive effects of their own.
Symptoms that track the day
Large glucose swings produce irritability, shakiness, anxiety and fatigue that follow a daily pattern. Diabetes and depression co-occur at higher rates in both directions. Symptoms with a reliable relationship to meals are worth mapping before they are attributed to temperament. Blood Sugar and Metabolic Health goes further.
Uncommon, and worth the basic check
Hypercalcaemia from an overactive parathyroid gland causes low mood, fatigue and cognitive change. Cortisol excess produces depression, anxiety and insomnia along with physical signs. Adrenal insufficiency causes profound fatigue and low mood. These are not common, and calcium and basic chemistry are already on most routine panels.
A few more belong in the history without needing a card. Undiagnosed coeliac disease and other malabsorption states produce deficiencies that show up as mood symptoms. Postural orthostatic tachycardia syndrome is frequently labelled anxiety before it is measured. Persistent symptoms after infection carry a mood component that is real and not a reinterpretation of the physical one. And traumatic brain injury, including injuries years earlier, changes mood regulation in ways that are consistently underweighted.
On what a workup is for
Checking your thyroid does not mean nobody believes you. It means somebody is looking properly.
The alternative is being managed for years around a contributor that one blood test would have found.
Real, partial, and widely overstated
Inflammation and mood, honestly.
This is one of the more solid pieces of biology in psychiatry, and it is also one of the most oversold. What is established: a subset of people with depression have elevated inflammatory markers. Giving healthy volunteers an inflammatory challenge reliably produces low mood, fatigue, social withdrawal and slowed thinking within hours. Interferon therapy induces depression in a substantial share of the people who receive it. Inflammation can cause depressive symptoms, and that is not in dispute.
What is not established is that it explains depression generally. Most people with depression do not have elevated markers. Trials of anti-inflammatory drugs in depression show benefit concentrated in participants who had raised inflammation to begin with, which is a genuinely important finding and also a narrow one. It points toward matching treatment to biology rather than toward an anti-inflammatory approach for everyone.
The gut sits in the same category. Communication between gut, microbiome and brain is real, and the mechanisms are being mapped properly. Human trials of probiotics for depression remain small, short and inconsistent, and the effect sizes claimed in marketing are not the effect sizes in the trials. Gut Health covers what is settled and what is not.
A mechanism that is real in some people is not a model that explains everyone.
Established against claimed
Inflammation and mood
Established against claimed
Established
A subset of people with depression have elevated inflammatory markers
Inflammatory challenge in healthy volunteers produces low mood and withdrawal within hours
Interferon therapy induces depression in a substantial share of recipients
Anti-inflammatory trials show benefit concentrated in those with raised markers
Claimed beyond it
That inflammation explains depression in general
That a consumer panel can confirm it in an individual
That probiotics are an established treatment
That diet alone resolves a mood disorder
The useful reading is that biology varies between people with the same diagnosis, and that measuring it may eventually direct treatment. That is a research direction, not a product.
What actually has support behind it
What the evidence supports.
Ranked roughly by the weight of human evidence rather than by how novel it sounds. Nothing here is presented as a replacement for care you are already receiving, and none of it should be started or stopped on the strength of a web page. The pattern worth noticing is that the best supported interventions are largely not products.
Comparable to medication, and it lasts
Cognitive behavioural therapy, behavioural activation and interpersonal therapy all have substantial trial evidence. For mild to moderate depression the effect is broadly comparable to medication, and gains persist better after treatment stops. For severe depression, the combination outperforms either alone. Access is the limiting factor far more often than efficacy.
Genuine effect, honest caveats
Meta-analyses find a moderate antidepressant effect, larger in trials with more vigorous activity and supervision. Smaller and better blinded trials tend to show smaller effects, which is worth stating rather than hiding. Even discounted, the effect is real, and it acts on sleep, metabolic health and inflammatory signalling at the same time.
One of the highest yield interventions
Cognitive behavioural therapy for insomnia improves both sleep and depressive symptoms, and it outperforms sedative medication over time. If apnoea is present, treating it changes daytime mood and cognition. Sleep is treated as a symptom to be waited out far more often than it is treated as the intervention it can be.
Effective, imperfect, and worth understanding
Antidepressants outperform placebo for moderate to severe depression, with a modest average benefit and wide individual variation. Response cannot currently be predicted in advance. They deserve a fair hearing, a proper trial at an adequate dose, structured measurement of response, and a review of the diagnosis if nothing moves. Stopping abruptly can produce a difficult discontinuation syndrome, so changes belong with the prescriber.
Underrated because it is not a treatment
Loneliness and social isolation are associated with depression, anxiety and worse physical outcomes, and the association runs in both directions. Behavioural activation works partly by restoring contact and activity. This is not a soft recommendation appended to the end, it is one of the more powerful variables in the whole picture.
Specific, cheap, and often forgotten
Bright light therapy is established for seasonal depression and has evidence in non-seasonal depression as well. Regular light exposure in the morning, consistent sleep timing and reduced evening light all support circadian regulation, which is disturbed in a large share of mood disorders. Anyone with a bipolar spectrum diagnosis should discuss light therapy with their clinician first.
The intervention people least want to hear
Alcohol worsens sleep architecture, produces rebound anxiety, interacts with medication and lowers the ceiling on any other treatment. A trial period without it is one of the few changes that reliably reveals how much of the picture it was responsible for, and the answer is frequently more than expected.
Promising, with a short evidence base
A Mediterranean style pattern has trial evidence for improving depressive symptoms, though from a small number of studies with unavoidable blinding problems. Omega-3s with a higher EPA proportion show benefit in some trials, mainly as an addition rather than a replacement. Worth doing. Not worth presenting as equivalent to the four items above it.
Three notes on the edges. Transcranial magnetic stimulation is established for depression that has not responded to medication and is widely available. Electroconvulsive therapy remains the most effective treatment that exists for severe depression, and modern practice bears little resemblance to its reputation, which is a case of stigma rather than evidence driving under-use. Ketamine and psilocybin assisted approaches are genuinely promising and belong in supervised clinical settings with proper screening, not in the informal market that has grown around them. All three are conversations for a psychiatrist.
The sequence
How I approach mood and anxiety symptoms.
This describes a clinical process carried out with a person, alongside mental health care rather than instead of it. What I add is the physical layer, done thoroughly and early, because that is the layer most often left incomplete and it changes how well everything else works.
Get help now
Some of this is not a scheduling question.
If you are having thoughts of suicide or you are in immediate danger, call or text 988 to reach the Suicide and Crisis Lifeline in the United States, or call 911. If you are outside the United States, call your local emergency number. You do not have to be certain that it is serious enough to call.
Thoughts of ending your life
A plan, or access to means
Thoughts of harming someone else
New confusion or disorientation
Hearing or seeing things others do not
Beliefs others find alarming
Days without needing sleep, with racing thoughts
New psychiatric symptoms with seizures or abnormal movements
Severe agitation, fever and muscle rigidity on medication
Symptoms starting within weeks of childbirth
Inability to eat, drink or care for yourself
Withdrawal from alcohol with tremor or confusion
Three of these deserve naming plainly. Symptoms beginning in the weeks after childbirth, particularly confusion, agitation or unusual beliefs, are a psychiatric emergency and treatable. A period of markedly elevated mood, reduced need for sleep and racing thoughts needs urgent review before any antidepressant continues. And severe agitation with fever, sweating and muscle rigidity in someone taking serotonergic medication needs emergency assessment the same day.
Free guide
Do not let the physical layer go unchecked
Brain on Fire walks through what belongs in a proper history. Which foundational tests find the contributors that routine visits skip. And which symptoms need urgent assessment rather than an appointment.
Own your biology
Both things are true.
Your symptoms are not a failure of character, and they are also not fully explained by a single chemical. They sit at the meeting point of a life, a body and a nervous system, and every one of those three can be examined.
The most useful thing you can do is refuse the false choice. Keep the care that is helping, and insist that the physical layer gets checked properly rather than assumed to be fine.
Biological from imbalanced Depression has real biology. That biology was never one neurotransmitter running low.
Syndrome from disease One diagnosis covers several different conditions, which is why average results understate what some people get.
Contributor from cause A low ferritin rarely explains everything. It reliably explains something.
Non-response from failure Nothing moving is information. It should reopen the assessment, not just change the prescription.
Bring what has already been tried.
No pressure, and nothing to buy. Bring when it started, what has been tried and for how long, every medication and supplement including the over the counter ones, how you actually sleep, and any results you already have. This works alongside your mental health care, not instead of it.
Common questions
Questions about mood and anxiety.
Short, plain answers to what people ask most.
Is depression caused by a chemical imbalance?
What blood tests should be done for depression or anxiety?
Can thyroid problems cause depression or anxiety?
Why is my anxiety not responding to treatment?
Should I stop my antidepressant?
Does inflammation cause depression?
Can diet and supplements treat depression?
How do I know if it is anxiety or something physical?
