First published April 9, 2021.
The short answer: in a large study of health records published in April 2021, about one in three people (33.6%) received a neurological or psychiatric diagnosis in the six months after COVID-19. Most of those diagnoses were anxiety and mood disorders, and many were not new problems. For about one person in eight (12.8%), it was the first such diagnosis of their life. Strokes and dementia were much rarer, but they were noticeably more frequent among people who had been severely ill.
The figure travelled fast in headlines, often reworded as "one in three COVID patients suffer brain disorders". That version is not quite right. Below is what the researchers measured, what the numbers mean, how the study compares with flu, and why later work matters for anyone still dealing with symptoms.
What the Oxford team studied
The research was led by psychiatrist Maxime Taquet and colleagues at the University of Oxford and was published in The Lancet Psychiatry. Rather than recruiting patients, the team analysed electronic health records from the TriNetX network, which at the time held anonymised data on more than 81 million people, mostly in the United States.
From those records they identified 236,379 people who had been diagnosed with COVID-19 and were still alive six months later. They then looked for 14 kinds of diagnosis in that period, from stroke, brain bleeds, parkinsonism and dementia to anxiety, mood disorders, psychosis, substance misuse and insomnia.
The design is retrospective. It shows which diagnoses were recorded after infection, not why they happened. That point matters for the headline figure, because the combined "neurological or psychiatric" category puts very different conditions in one bucket.
The key numbers, explained
Across the whole group, the main findings were:
- Any neurological or psychiatric diagnosis: 33.62% within six months.
- A first-ever diagnosis of this kind: 12.84%.
- Anxiety disorders: 17.39%, the most common single outcome.
- Mood disorders such as depression: roughly 14%.
- Ischaemic stroke: 2.10%.
- Dementia: 0.67%.
- Psychotic disorder: 1.40%.
- Intracranial haemorrhage (bleeding in the brain): 0.56%.
So the one-in-three figure is driven mainly by common mental health conditions, and a good share of those people had a history of similar problems. It does not mean that a third of patients developed brain damage.
Severity made a big difference
The risk climbed with how ill people had been. Among patients admitted to intensive care, 46.42% had a neurological or psychiatric diagnosis within six months, and 25.79% had a first diagnosis. In the same intensive-care group, 6.92% had an ischaemic stroke and 1.74% were diagnosed with dementia. Among people who had developed encephalopathy (an acute disturbance of brain function, such as delirium) while ill, 62.34% had a later diagnosis, with stroke at 9.35% and dementia at 4.72%.
The authors stressed, though, that the risk was not limited to severe cases. People who were never admitted to hospital also showed raised rates, just lower ones.
Compared with flu and other infections
The strongest part of the study is its comparison groups. The team matched COVID-19 patients with similar people who had influenza or other respiratory infections, adjusting for age, sex, ethnicity, existing illnesses and other factors.
After matching, COVID-19 was linked to a 44% higher risk of a neurological or psychiatric diagnosis than flu (hazard ratio 1.44) and a 16% higher risk than other respiratory infections (hazard ratio 1.16). For a first-ever diagnosis, the gap was wider: 78% higher than flu and 32% higher than other respiratory infections.
Taquet summed it up in interviews, saying to the Financial Times that "brain diseases and psychiatric disorders are more common after Covid-19 than after flu or other respiratory infections". He added that the next step was to see what happens beyond six months.
What the study cannot tell us
A common myth around this research is that it proved the virus directly injures the brain. It did not. Several explanations can sit side by side:
- Direct and indirect biological effects. COVID-19 can cause blood clotting problems, inflammation and low oxygen levels, all of which can affect the brain. The higher stroke rates in severe cases fit this picture.
- The stress of serious illness. Being critically ill, isolated in hospital, or afraid of dying can leave lasting anxiety, depression or trauma symptoms, especially in 2020, when the disease was new and frightening.
- Pandemic life itself. Lockdowns, job losses and bereavement raised stress for everyone, although the matched comparison with flu patients helps to separate this from the infection.
- How records are kept. Health-record studies depend on what doctors write down. People with COVID-19 may have been followed up more closely, so more problems may have been recorded.
The authors themselves called for prospective studies that follow patients over time to confirm the findings and explain the mechanisms.
Other research pointed the same way
Smaller studies published around the same time reported similar patterns. A team at the Gemelli hospital in Rome followed 381 people treated there for COVID-19 and reported that about 30% showed symptoms of post-traumatic stress disorder after recovery, a finding published in early 2021. Because that group was much smaller and drawn from one hospital, it is best read as a signal rather than a precise rate.
Researchers also began describing long COVID: symptoms such as fatigue, breathlessness, poor concentration ("brain fog") and sleep problems that persist for months after the initial infection. Long COVID and the diagnoses in this study overlap but are not the same thing. Someone can have brain fog without any formal neurological diagnosis, and someone can be diagnosed with anxiety without having long COVID.
What this means for you
For most people, the practical lessons are modest and reassuring as well as cautionary:
- Feeling anxious, low or exhausted after an illness is common and treatable. It is worth mentioning to a doctor rather than waiting it out.
- New symptoms such as sudden weakness, facial drooping, slurred speech or confusion are emergencies whether or not you have had COVID-19. Call emergency services straight away.
- If memory, concentration or mood problems persist for weeks after infection, talk to a doctor, who can look for treatable causes and point you to support services.
- People who care for someone recovering from a severe illness should watch for changes in mood or thinking and encourage them to seek help.
The Oxford study remains one of the most cited early pieces of evidence that COVID-19 can leave a mark beyond the lungs. Its main message is not that a third of patients were harmed in the brain, but that doctors and health systems should plan for mental and neurological care after infection, particularly for those who were seriously ill.
Sources: University of Oxford Department of Psychiatry: Taquet et al., 6-month neurological and psychiatric outcomes in 236,379 survivors of COVID-19; European Academy of Neurology summary of the study; Financial Times; original tip via CNN.
COMMENTS