LONDON – The largest study to date, involving 236,379 confirmed cases, shows that 1 in 3 survivors of COVID-19 was diagnosed with a neurological or psychiatric condition within six months of contracting the infection. For 13% of those patients, it was their first such diagnosis.

The more serious neurological problems occurred in those who were more seriously ill as a result of COVID-19. For example, of those admitted to intensive care, 7% subsequently had a stroke and 2% were diagnosed with dementia.

Anxiety disorders, affecting 17% of patients, and mood disorders, affecting 14%, were the most common conditions.

Psychiatric and neurological conditions were more common in COVID-19 patients than in patients who had flu or other respiratory tract infections during the same time period, implying there is a specific impact of being infected with SARS-CoV-2.

The study, by researchers at Oxford University, draws on electronic health records from the U.S. Trinetx network, which has data on 81 million people, both with and without health insurance. It looked at the incidence of 14 neurological and psychiatric outcomes.

Patients older than 10 who became infected after Jan. 20, 2020, and were still alive on Dec. 13, 2020, were included in the analysis. They were compared to 105,579 patients diagnosed with flu and 236,038 diagnosed with any respiratory tract infection – ranging from sinusitis to pneumonia – and including flu.

In addition to making a direct comparison of subsequent psychiatric and neurological conditions, the comparison to flu and respiratory tract infections was intended to control for social factors, such as loss of employment, loss of social support and living through a pandemic, which could cause stress and subsequent mental health problems.

The results indicate brain disease and psychiatric disorders are more common after COVID-19, even when patients are matched for risk factors, said Max Taquet, clinical fellow in psychiatry at Oxford University and co-author of the paper published in Lancet Psychiatry.

“We now need to see what happens beyond six months. The study cannot reveal the mechanisms involved, but does point to the need for urgent research to identify these, with a view to preventing or treating them,” Taquet said.

After taking into account age, sex, ethnicity and co-morbidities, there was overall a 44% greater risk of neurological and psychiatric diagnoses after COVID-19 than after flu, and 16% greater risk than with all respiratory tract infections.

Although the researchers attempted to control for social factors, they acknowledge circumstantial, along with biological mechanisms, are likely in play.

“A lot of mental health problems are to do with stress, rather than the direct effect of the virus on the brain, or on the immune system,” said Paul Harrison, professor of psychiatry at Oxford University, who led the study.

It is also the case that patients who recovered from COVID-19 might be more likely to be diagnosed with a psychiatric or neurological condition because they had more follow-up than flu or respiratory tract infection patients, Harrison noted, saying it is important not to overestimate the excess risks associated with COVID-19. “However, there is an excess risk,” he said.

In an earlier study, Harrison and his colleagues delivered evidence that people who survive COVID-19 infection are at increased risk of being diagnosed with a psychiatric disorder, with almost 1 in 5 survivors getting a diagnosis of anxiety, depression or insomnia in the three months following a positive test for the virus.

Of those, 1 in 4 had not had a psychiatric disorder diagnosis before COVID-19.

In addition, people with a pre-existing psychiatric disorder were 65% more likely to be diagnosed with COVID-19, even when known risk factors, such as diabetes, hypertension and obesity were taken into account.

That study, also using Trinetx, covered 62,354 patients who had a diagnosis of COVID-19 between Jan. 20 to Aug. 1, 2020.

In the new study involving 236,379 cases of COVID-19, 190,077 did not need hospital treatment, 46,302 were hospitalized, 8,945 had intensive care and 6,229 had some form of encephalopathy.

It is known SARS-CoV-2 can enter the brain, mostly likely through the olfactory bulb, said co-author Masud Husain, professor of cognitive neurology at Oxford University. To date, there is little evidence of the virus getting into neurons, but it has been shown to cause inflammation in the surrounding tissues. In terms of the observed neurological conditions that is “probably important,” and in addition, “the virus can cause clotting that can impact the brain,” Husain said.

Post-mortem investigations are becoming very important in trying to establish the underlying mechanisms of how SARS-CoV-2 affects the brain, and more studies are due to be published in the next few months, noted Husain.

The six-month follow-up study was conducted before concerns were raised about a possible link between Astrazeneca plc’s COVID-19 vaccine and serious blood clotting, including cerebral venous sinus thrombosis, and the researchers said they did not look for that condition.

Long COVID-19

There was some overlap between psychiatric and neurological problems, with some patients having more than one diagnosis, but it is not possible to tell from the data if one leads to another, or if one condition is more likely to co-occur with another.

Harrison said it also is not possible to map the findings of this study over others investigating Long COVID, such as the U.K.’s Phosp-Covid study, which is looking at the prevalence of more than 40 different symptoms, including cognitive impairment and insomnia, in patients hospitalized with COVID-19.

“Long COVID includes some features [of psychiatric and neurological conditions], for example brain fog. I expect there is going to be quite a lot of overlap, but we didn’t look at it,” Harrison told BioWorld.

It also was not possible to assess risk at baseline, thought the relative risk of getting a first diagnosis of a psychiatric or neurological condition was higher.

Both recurrent and first diagnoses are presented in the paper because they are complementary sources of information, Harrison said. “But if you wanted to be more purist on the effect of COVID-19, maybe initial diagnosis is more important.”