People usually arrive at this question the same way: they read something frightening. An article claiming depression shrinks the brain, or that every week of untreated psychosis causes permanent loss.
Then they try to work out whether it is true, and find the internet unhelpfully confident in both directions.
Here is a straight answer.
The short version
Claims that untreated mental illness causes definite, permanent brain damage are not supported by current evidence.
What is reasonably established is narrower: some mental illnesses are associated with measurable differences in the brain, and in psychosis, longer delay before treatment is associated with somewhat poorer outcomes. Whether the illness causes those differences, or they precede it, or both arise from something shared, is genuinely unresolved.
That is less dramatic than the headlines. It is also the honest position, and we think a clinic should say so rather than use a frightening claim to push people toward treatment they should choose for better reasons.
What the research actually shows
Differences are observed
Studies comparing groups of people with certain conditions to groups without do find average differences in brain structure and function. This is real.
Three caveats are usually lost in the retelling. These are average differences between groups, not findings that apply to any individual. They are typically small. And they cannot, by themselves, tell you what caused what.
The direction of causation is unclear
This is the heart of it. If people who have been ill longer show more difference, that could mean the illness caused it. It could equally mean that people whose brains differ in some way at the outset tend to have longer or harder illnesses. Observational studies cannot separate these.
The National Institute of Mental Health describes schizophrenia as involving a combination of genetic, environmental and brain-related factors — a picture of interacting contributors rather than a simple sequence.
The psychosis delay literature
The most-studied version of this question concerns how long someone is psychotic before treatment begins. Across many studies, longer delay is associated with somewhat poorer outcomes, and reviews generally describe that association as small to modest but fairly consistent.
Two explanations compete in the literature: that untreated psychosis is directly harmful, or that something else influences both how long someone takes to reach treatment and how well they do. Large trials have produced mixed results — at least one well-known randomised study did not find delay predictive of outcome in its main analysis.
We go into this further in untreated schizophrenia.
What about depression?
You will see claims that depression shrinks the hippocampus. Group differences have been reported. Whether repeated episodes cause this, and whether any change is permanent, is not settled — and some evidence suggests brain measures can change with recovery, which argues against a simple story of accumulating damage.
What is better established
Rather than speculative neurology, these are the effects of untreated illness that are well supported — and they are the ones that should actually influence your decision.
- Cognitive symptoms during illness are real. Concentration, memory and decision-making are commonly affected while someone is unwell. These generally improve as the illness is treated, which is the opposite of damage.
- Physical health deteriorates. Sleep, activity, nutrition and attendance at medical appointments all suffer, and those have measurable consequences.
- Substance use frequently develops, and heavy sustained alcohol use does have well-documented effects on the brain. If you are worried about brain health specifically, this is the mechanism with the strongest evidence behind it.
- Life opportunities are lost — education, work, relationships. Not neurological, and often the most lasting cost.
Notice that the best-evidenced route from untreated mental illness to brain harm runs through alcohol, not through the illness directly.
Why the overstatement is harmful
It is tempting to think a frightening claim at least gets people through the door. In practice it does the opposite in two ways.
It makes people who have been unwell for years conclude they have missed their chance. We hear this directly — families asking whether it is "too late" after a decade. It is not, and that belief delays treatment further.
And it damages trust. A patient who is told something alarming, then discovers it was overstated, reasonably wonders what else they were told for effect.
The World Health Organization frames mental disorders as treatable conditions. That framing is both accurate and more useful than fear.
The reason to treat early anyway
Nothing above is an argument for waiting. The case for early treatment stands on firmer ground:
Symptoms are distressing now, and treatment relieves them. Earlier treatment means less disruption to education, work and relationships. The first contact with services being a calm appointment rather than a crisis admission genuinely shapes how someone engages for years. And the longer patterns of avoidance run, the more there is to unpick.
Those reasons do not require anyone to be frightened about their brain.
Where the frightening version comes from
It is worth understanding how a modest research finding becomes an alarming headline, because it helps you read the next one.
A study reports a small average difference in a brain measure between two groups. The paper itself is careful: it describes an association, notes that the design cannot establish causation, and reports an effect size most readers would consider slight.
The press release drops the caveats. The headline converts "associated with" into "causes". A second-hand article converts a group average into something that applies to the individual reader. By the fourth retelling, a cautious finding about a cohort has become a claim that your illness is destroying your brain.
No single step in that chain is outright fabrication. The cumulative distortion is still severe, and this question attracts it more than most, because fear travels further than nuance.
How to check a claim like this yourself
Four questions will get you most of the way.
Is it a group average or an individual prediction? Almost all brain-imaging findings in psychiatry are the former, and a difference between group means tells you very little about any one person.
Does the design permit a causal claim? Observational studies, which is most of this literature, can show that two things occur together. They cannot establish which came first, and honest papers say so.
How large is the effect? "Statistically significant" is not a synonym for "large". With a big enough sample, very small differences reach significance.
Who is saying it, and what do they gain? Facilities that benefit from urgency sometimes have an interest in overstating permanence. We have an interest here too, which is part of why this article is written the way it is.
If a page tells you something is certain while the underlying literature is in fact divided, that is a reason to discount the page rather than to be frightened by it.
Frequently asked questions
So does untreated mental illness cause brain damage or not?
Current evidence does not establish that it does. Associations between illness and brain measures exist; causation is unresolved. Treat confident claims in either direction with caution.
Is it too late if someone has been ill for years?
No. People who have been unwell for a long time recover with treatment. Length of illness may make treatment slower, not futile.
Does depression permanently affect memory?
Concentration and memory problems are common during depression and generally improve with treatment. Persistent difficulty after recovery is worth assessing, as it may have another cause.
What about alcohol?
Heavy sustained alcohol use does have well-documented effects on the brain. This is the clearest mechanism by which untreated mental illness can lead to physical harm, because drinking so often develops alongside it.
Should I worry about brain scans?
Brain imaging is not used to diagnose common mental illnesses. It is used to exclude other conditions where there is a clinical reason. A clinician not ordering a scan is not cutting corners.
What to take from this
If you came here frightened by something you read, the accurate position should be reassuring: the alarming version is not established, and people who have been unwell for a long time do get better.
If you came here deciding whether to seek help, the reason to do so is that the symptoms are costing you something now — which is sufficient. SSHIMOH's clinical team assesses across the range of conditions discussed here, and an appointment commits you to nothing.