Mental health: what it is, what threatens it, and how to heal

Mental health is more than the absence of illness. It's the capacity to feel, think, connect, and act in ways that let you handle stress, relate to others, and pursue what matters. Like physical health, it fluctuates — and like physical health, it responds to care.

• Persistent low mood, anxiety, or emotional numbness
• Sleep changes — insomnia or sleeping far too much
• Withdrawal from people or activities you used to enjoy
• Difficulty concentrating or making decisions
• Appetite or weight changes
• Using alcohol or substances to cope
• Feeling worthless, hopeless, or 'not yourself'
• Thoughts of self-harm or suicide (seek help immediately)

• Genetics and family history
• Chronic stress, trauma, or grief
• Physical illness or hormonal changes
• Isolation, loneliness, or unstable relationships
• Substance use
• Sleep deprivation and poor nutrition
• Life transitions — job loss, moves, parenthood, retirement

Recovery looks different for everyone. Below are the evidence-based and complementary approaches most often used — often in combination.

There is a persistent idea that good mental health means feeling good most of the time, and it causes a surprising amount of harm. By that definition, grief is a disorder, anxiety before something important is a malfunction, and anger at genuine injustice is a symptom. A more useful definition is closer to range and recoverability: the capacity to feel the full spectrum of human emotion in proportion to what is happening, and to return to baseline afterwards rather than getting stuck.

Under that definition, the concerning signals are different from the ones people usually watch for. Not 'am I sad?' but 'has this lasted long past its cause?'. Not 'am I anxious?' but 'is the anxiety arriving without a trigger, or refusing to leave after the trigger is gone?'. Not 'am I struggling?' but 'has my life quietly become smaller in order to accommodate the struggle?'. Duration, proportion and functional impact are what clinicians actually assess, and they are what you can usefully assess in yourself.

This framing also removes a barrier that stops people getting help. You do not need to qualify for a diagnosis to deserve support, and you do not need to be in crisis for treatment to be worthwhile. The evidence for early, low-intensity intervention is strong precisely because problems are easier to shift before they have become a settled architecture.

Good mental health is not a flat line. It is a wide range with a reliable way back.

Before any specific therapy or medication, there is a layer that determines how much traction anything else gets. It is deeply unfashionable because it cannot be packaged or sold, and it is consistently the highest-yield place to start.

Sleep comes first, because insufficient sleep produces, within days, symptoms that are functionally indistinguishable from anxiety and depression in a healthy person: emotional volatility, impaired concentration, negative interpretation bias, reduced tolerance for stress. Fixing sleep does not cure mental illness, but attempting to treat mental illness on four hours a night is like rehabilitating a knee while running on it daily.

Then movement, daylight and food, in that order of evidence. Then connection — the single most robust predictor of long-term wellbeing across the entire epidemiological literature, and the one most likely to erode invisibly during a difficult period. And finally purpose: something that requires you and rewards effort. These five are not a substitute for professional treatment when treatment is needed. They are the surface it has to be built on, and neglecting them is why some people conclude that therapy or medication 'didn't work' for them.

• Fixed wake time, seven days a week — the highest-leverage single habit
• Daylight within an hour of waking
• Movement most days, at whatever intensity you will actually sustain
• One real conversation with another human being, regularly and deliberately
• Something to be bad at and slowly improve in — mastery is a mood input

The largest single predictor of whether therapy works is not the school of therapy — it is the quality of the working relationship between you and the therapist. This is one of the most replicated findings in psychotherapy research, and it has an immediately practical implication: if you have tried therapy once, disliked the person, and concluded therapy doesn't work for you, you have tested one relationship rather than the whole intervention.

It is entirely reasonable to treat the first two or three sessions as an assessment running in both directions. Do you feel understood rather than processed? Does the therapist explain their approach and what progress would look like? Do you leave with something to think about or try? Are you able to disagree with them without the relationship becoming tense? A good therapist expects these questions and welcomes them.

Modality still matters for some presentations — CBT and its variants have the deepest evidence base for anxiety disorders and depression, EMDR and trauma-focused CBT for PTSD, DBT for emotional dysregulation, family-based approaches for eating disorders in adolescents. A reasonable strategy is to choose the modality with the best evidence for your situation, and then, within that, choose the person you can actually talk to.

• Match modality to problem, then prioritise fit with the person
• Ask what progress will look like and when you will review it together
• Two or three sessions is a fair trial of fit; a poor fit is not a failed treatment
• Cost barriers are real — ask about sliding scales, group programmes and public services

If distress lasts more than two weeks, disrupts daily life, or includes thoughts of self-harm, reach out. Early help almost always shortens the road.

The period immediately after improvement is the most under-managed stage in mental health, and the most common point of relapse. Feeling better removes the urgency that drove the effort, so the appointments stop, the routines loosen, the medication gets skipped, and the difficult conversations get postponed. Several months later the same pattern returns and it feels inexplicable.

People who stay well tend to do a small number of unremarkable things consistently. They know their own early warning signs, in specific rather than general terms — not 'I get low' but 'I stop replying to messages, and my sleep shifts an hour later'. They keep one or two non-negotiable anchors regardless of how well things are going. They review rather than abandon treatment, tapering with a clinician rather than stopping abruptly. And they have at least one person who is allowed to tell them that something looks different.

None of this is dramatic and none of it requires the illness to be present to be worth doing. That is the point. Maintenance is what turns a recovery into a durable one, and it is the part of the process where the smallest amount of ongoing effort buys the largest amount of protection.

Getting better is one project. Staying better is a different one, and it needs its own plan.

• World Health Organization — Mental Health
• NIMH — Mental Health Information