May 11, 2022

expert reaction to study looking at neuropsychiatric conditions after severe COVID-19 and after other severe acute respiratory infections

A study published in JAMA Psychiatry looks at neuropsychiatric consequences of severe COVID-19 and other severe acute respiratory infections.

Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

“This is an impressive study, that uses appropriate statistical methods, but it needs some care in its interpretation.  It’s an observational study, and there were a lot of differences between the three groups who were compared in the main statistical analysis (people who had been hospitalised with Covid-19, people who had been hospitalised with a different serious acute respiratory infection, and everyone else) apart from whether they had been in hospital and, if so, what for.  The researchers made a lot of appropriate statistical adjustments to allow for differences in people’s personal characteristics (age, sex, ethnicity and a lot more), previous health conditions, and more.  Also, because they compared the three groups over the same time period, differences in the availability of health care could well have affected them all in broadly similar ways.

“One can never be absolutely certain about the interpretation of findings of an observational study.  But I don’t seriously doubt the overall finding that, in the period when the research was done, the chance of having a new diagnosis of a neuropsychiatric condition or a new prescription for neuropsychiatric medication was higher for people who had been hospitalised either for Covid-19 or for some other serious respiratory disease, than for the general public, and also that the risks of these neuropsychiatric events were similar in people who had been hospitalised for Covid or for some other serious respiratory disease.

“However, the research can’t tell us whether that would have been the position in another period of time.  The data all came from the period between late January 2020 and early July 2021 – that is, most of the hospitalisations would have occurred during the first two pandemic waves in England.  At that time there was generally a lot of disruption to health services, and indeed to life in general because of lockdowns and other restrictions.  So what the research can’t tell us is what caused the increased psychiatric risks for people hospitalised with Covid-19 or some other serious respiratory disease.  It can’t tell us what might happen in the future, when, we all hope, many fewer are being hospitalised with Covid-19 than was the case in those first two waves, and the current backlog of provision of some health services has decreased.  (The researchers make this clear in the Limitations section of their research paper by describing the conditions of the pandemic that might have affected their findings.)

“So we can’t just say that, in general, serious Covid-19 has much the same neuropsychiatric consequences as other very serious respiratory illness.  Maybe it does, maybe it doesn’t.

“We do know from the other parts of the research results that psychiatric conditions after hospitalisation do depend on the time period involved.  The researchers aimed to compare psychiatric events after serious acute respiratory disease, apart from Covid-19, in two periods of time – January 2015 to January 2020, before the pandemic, and late January 2020 to early July 2021 during the pandemic.  But they decided not to carry out detailed statistical analysis to make this comparison, because the risks of some of the neuropsychiatric events they recorded were too different in the two periods.  That might well be because of the special conditions during the pandemic, though this research can’t confirm that.

“Also, the researchers compared (for the pandemic period) the risks of psychiatric events between people who had been hospitalised for a heart attack (acute myocardial infarction – this is in Table 3 of the research paper), people who had been hospitalised with Covid-19 and people who had been hospitalised with a different serious acute respiratory infection.  They summarise these results in their paper simply by saying that the risks ‘varied’, and they certainly did.  The risk of a new diagnosis of anxiety was greater in patients who had had a heart attack than in either group that had been in hospital for a respiratory infection.  The risks of a new dementia diagnosis, a new antidepressant prescription, and a new antipsychotic prescription were greater for the respiratory patients (both groups) than the heart attack patients.  Other risks were comparable in all three groups.  That does all indicate that the extra risk of psychiatric events in the people hospitalised for Covid-19 or for another serious respiratory infection, compared to the rest of the public, wasn’t simply because they had been in hospital for a serious disease, because the pattern of risk was different than for the heart attack patients.  But again, this applies only to the first year and a half of the pandemic, only to England, and the research can’t tell us the reasons.”

Dr Max Taquet, NIHR Oxford Health BRC Senior Research Fellow, University of Oxford, said:

“This interesting study by Clift et al. used electronic health records data to quantify the risks of neuropsychiatric sequelae in the year after being hospitalised for COVID-19 among 32,000 adult patients.  It adds to a large body of evidence showing that COVID-19 is associated with an increased risk of neuropsychiatric diagnoses.  Interestingly, the study found that the risks were similarly elevated in patients hospitalised with COVID-19 and those hospitalised with another severe acute respiratory infection.  This suggests that, in severe infections, neuropsychiatric consequences of both COVID-19 and other severe respiratory infections might in part share common causes.

“As this study is limited to hospitalised adult patients, it cannot inform us on the risk in non-hospitalised individuals (which is the overwhelming majority of patients with COVID-19) nor in children.  Whether the neuropsychiatric risks have remained the same since the emergence of the Omicron variant remains an open question since all patients in this study were diagnosed before July 2021.”

‘Neuropsychiatric Ramifications of Severe COVID-19 and Other Severe Acute Respiratory Infections’ by Ashley Kieran Clift et al. was published in JAMA Psychiatry at 16:00 UK time on Wednesday 11 May 2022.

DOI: 10.1001/jamapsychiatry.2022.1067

Declared interests

Prof Kevin McConway: “I am a Trustee of the SMC and a member of its Advisory Committee.  My quote above is in my capacity as an independent professional statistician.”

Dr Max Taquet: “I work in the same university as the authors but we do not directly collaborate. I also work on similar studies trying to identify, quantify, and specify the neurological and psychiatric consequences of COVID-19.”