Study Finds Cognitive Deficits in People Who Had COVID

BIANNA GOLODRYGA: Well, vaccinations continue to reduce infections. Hospitalizations and deaths. But some scientists are now concerned about the long-term impacts of the virus on the brain. Dr. Adam Hampshire is one of them. He is a cognitive neuroscientist at Imperial College London. Here he is breaking down his findings with our Hari Sreenivasan.

(BEGIN VIDEO CLIP)

HARI SREENIVASAN: Bianna, thanks. Dr. Adam Hampshire, thank you so much for joining us. First kind of a top-level overview and then we’ll kind of get into what your search was showing. But what did you find through surveys in the U.K.?

DR. ADAM HAMPSHIRE, READER, COGNITIVE NEUROSCIENCE, IMPERIAL COLLEGE LONDON: So, essentially, what we found was that people who had recovered from COVID-19, they showed a degree of underperformance in terms of a variety of aspects of cognitive abilities. So, just to expand on that a little bit, what I mean by that is they perform far less rather than we would expect from people just like them.

SREENIVASAN: And what kinds of underperformance? What kinds of cognitive impairment are we talking about?

HAMPSHIRE: Yes. So, the tool that we were using for cognitive assessment, it’s designed to try and measure multiple different aspects of cognition. For example, we have the executive functions like planning and reasoning, there’s attention to how good you are at focusing on a particular task, there’s memory functions and working memory functions, also language. Now, what we’ve observed is that some of those seem to be pretty much normal, but some of the higher sort of executive planning and language functions, those are where people were underperforming.

SREENIVASAN: Let’s get to how you came across this. It seems like there was also a little bit of a serendipity here, that you were already doing a massive survey before the pandemic really took hold.

HAMPSHIRE: Yes, it was a massive coincidence, if truth be known. So, I had set up a collaboration with BBC Two – Horizon here in the U.K., where what we were essentially trying to do is map different dimensions of human cognitive ability across the U.K. population. And that study, because it was promoted through BBC, it was very, very successful in terms of numbers. So, actually, in terms of throughout the whole of 2020, we tested about 390,000 people, primarily in the U.K. And, you know, just to explain, this isn’t a short assessment. You know, People volunteer, they’re not paid and they’d put in 40 minutes of time into the cognitive assessment, plus a detailed questionnaire afterwards. So, it’s very coincidental, we had just set the study up. It was running for a couple of months and then the pandemic took off in the U.K. And what happened, essentially, at that point in time is the number of colleagues of mine from different universities wrote to me and they said, you know, maybe this could be useful, we could try and work out what the impact of the pandemic is on people’s mental health in general, but also you should be able to capture some people who have been ill. And I had been thinking along similar lines. So, we sort of grouped together and collaborated and put together some extended questionnaires that looked at how people’s daily lives had been affected, what was happening in terms of their mental health, but also, when they’ve been ill, if they suspected that was COVID-19, if that had been confirmed with biological tests and also what happened, for example, did they end up in the hospital, did they have breathing difficulties, did they end up on a ventilator. So, it’s very rich data. And we were — I mean, if you consider how fast the pandemic took off, it would have been pretty much impossible to set up a study to do this in advance. So, it was very, very lucky.

SREENIVASAN: So, what were the things that people were reporting and, more important, what were the correlations or associations that you could start to make of people who had had COVID, who had been hospitalized or not? What were they falling short on? Is it a memory test? Is it about their emotional IQ? How do you sort it?

HAMPSHIRE: So, people essentially appear to be underperforming in particular cognitive domains. The domains that we saw the largest underperformance included some of the higher what we refer to as semantic functions. So, how good you are at dealing with complex reasoning based around often difficult words. Like having logical reasoning, for example, metaphors. Other things people struggled with was spatial planning and focused attention. Now, these are — these fall within the realms of what we’d often refer or generally as an executive function. Things that are important for us to break down difficult problems and to make good decisions, essentially.

SREENIVASAN: One of the concerning things when you look into the details of your study is that we’re not just talking about people who had to go to the hospital. You’re also looking at, maybe because you had such a large data set, people were quarantined, who stayed at home, who might have had a very mild, if not even a non-noticeable form of COVID.

HAMPSHIRE: Frankly, this is what surprised me in our particular study. So, one extreme, you have people who ended up in the hospital or on a ventilator, and we kind of expected that might have an impact on their cognition. We know that from past studies in other disorders, right? But the fact that we see a cognitive underperformance in people who remained at home and had positive biological tests, but no medical assistance, you know, that’s kind of surprising. But over that, in terms of their imaging literature, it’s also the case there’s some evidence accumulating in that same milder population. This is actually a big interest in the U.K. at the moment. So, can we follow people up who had milder symptoms and see how they’ve been affected and what happens in terms of the recovery. So, one of the things that’s being, I would say, most useful about my particular study is we validated, essentially, a set of cognitive assessment tools that can be used and then, they’re now being applied in studies where they’re trying to track people who only (INAUDIBLE) have had COVID-19.

SREENIVASAN: So, you essentially say, OK, here is a woman who is 32, here is 100 women who are 32 years old and in good health, et cetera, this is their baseline, this is kind of where we should be on this particular type of skill or test, and then here is somebody who is post-COVID, perhaps has been hospitalized, and how did they do on the same score, is that right?

HAMPSHIRE: That pretty much captures it. So, we take all of those factors into account and we measure how someone’s score deviates from what we would expect from people just like them.

SREENIVASAN: How do you know, for example, that with such a large population that some of these people couldn’t have had these effects because of other underlying conditions, because of perhaps their own aging process, perhaps they were showing signs of early Alzheimer’s? How do we know it’s COVID and not other factors?

HAMPSHIRE: Well, it’s a great question. I mean, so, in the context of our study, what we’ve observed is an association, it’s a strong association, it’s a worrying association and we need to investigate it further. But I would say, we haven’t absolutely sort of nailed the cause-and-effect relationship here, because we don’t have the longitudinal data. What we can do is factor lots of things into our equation. So, you mentioned age. Age is something that we specifically take account of. We take many things into account, including pre-existing conditions. So, we survey people about whether they’ve had psychiatric conditions, neurological conditions, also biological conditions that we know predispose people towards COVID-19 illness and its severity, such as diabetes, for example. All of that is in the model. And that’s one of the powers of having so much data. You can’t normally do that, but we have data from like nearly 400,000 people from last year.

SREENIVASAN: You know, one of the things that is happening right now is that you’ve got patients who are coming in to different hospitals and doctors all over the world with long COVID symptoms and, because of that, you also see different research institutions starting to study long COVID and what are the possible reactions all over the world now. You kind of starting to see some of those initial studies and research be published. And this is your field now. So, what are you intrigued by that your colleagues in the profession are starting to look at?

HAMPSHIRE: There’s a real convergence of evidence towards the possibility of there being cognitive consequences of COVID-19. And perhaps if I can unpack that a little bit, on a categorical level, on one hand there’s been quite a lot of work looking at people’s self-perceptions of how they’ve been affected. So, this is all quite subjective. It’s what, you know, someone feels essentially has happened to them. And many people are reporting that they’re having problems with things like concentration, brain fog, problem solving and finding the words. So, what we’ve done, and some other groups are starting to do, is looked at objectively measuring cognition. And so, there what we do, we don’t just ask people how they feel they’ve been affected, instead, we essentially challenge them and say, can you solve this difficult problem, can you, you know, concentrate on this difficult task, and we objectively measure what they can cope with cognitively. Now, what’s interesting is on a sort of domain level, in terms of different aspects of cognition, the results we’re seeing dovetail quite closely between those data types of what people self-perceive and what we objectively measure. In addition to that, there’s an emerging body of evidence from the brain imaging literature and there, what they’re trying to do is look at brain imaging markers and pathology using different imaging methods and trying to quantify different types of pathology. And, again, there’s quite a lot of evidence beginning to emerge that there can be an impact on the brain. Again —

SREENIVASAN: When you say that, let me break that down. So, you’re talking about the COVID-19 is actually having an effect on the physical structure of my brain?

HAMPSHIRE: There’s evidence to suggest that this might be the case. Now, perhaps I can break that down a little bit more, because the way in which people are affected is likely to differ, for example, based on the severity of their illness. Let’s say, for example, at the far extreme, that you went to hospital and you’re put on a ventilator, there are people who have quite serious neurological complications under those conditions, and actually, that’s not specific to COVID-19.

SREENIVASAN: Sure.

HAMPSHIRE: That’s (INAUDIBLE). Perhaps you have people with multiple organ failure, et cetera, right? So, that can have neurological impact. There’s other groups that are arguing there could be a direct impact of the virus on the brain. Now, there’s a growing body of evidence gained to support that that may be happening. And in addition to that, there’s effects of things like hypoxia, so lack of oxygen, which can lead to death of some neurons and it can impact the brain. And, of course, there’s also general inflammatory responses. So, this is broad spectrum of different ways in which an illness like this could affect the brain. What we don’t really understand yet is how the aspects of cognition that we seem to be seeing could be affected, how those correlates with these different types of change in the brain.

SREENIVASAN: So, really, I guess, the first clue for that is when people lose their sense of smell. I mean, smell doesn’t actually happen in my nose, it happens in my brain, right? So, if there’s something — if the virus is able to get to the point of my brain that is telling my nose or not telling my nose what the smell is around me, then what is in the neighborhood of the olfactory lobe? What else could it be affecting?

HAMPSHIRE: Well, you’ve sort of hit the nail on the head there in terms of all of the theories out there. And I should clarify, I’m not doing research into that theory myself. But there’s the idea that there could be an impact through the sort of gustatory and olfactory system into the brain and then a kind of spread infecting other systems. And actually, some of the cognitive systems, they overlap with those systems in the brain. So, they’re in the neighborhood, as you say.

SREENIVASAN: If this work in pointing to you that the effects of COVID-19 are not like the things that we’re already used to, what is the bigger picture kind of impact on that? How does the health care system, how do policymakers need to think about long-term potential effects of COVID that, again, we’re only a year and a half into this, right, and researchers love to have the ability to have time where they can go and go back and forth to patients? But what if there are other effects down the line?

HAMPSHIRE: Well, that’s the big concern, isn’t it? And at the moment, we just don’t know. And that’s why more research is needed. And, you know, we need that longitudinal research, we need to be tracking people over time and we need the combination of different types of measures and information so that we can get a better idea of the underlying biological mechanisms, essentially.

SREENIVASAN: Here is the other part. OK, so there are these changes from COVID. Are any of them improving if you’ve had a cognitive deficit? Have you seen at all people say, well, it was pretty bad for a month or so, but I seem to be getting better now?

HAMPSHIRE: I would say at the moment there is some evidence beginning to emerge that there might be a sort of slow recovery trajectory post-COVID- 19, and I certainly hope that’s the case. I’ve seen one, I think, study out of the U.S. where they’ve started to show some evidence that that’s the case. And there’s another one I’m involved with where they’ve looked at people over longer periods of time who had mild symptoms and they seem to be recovering over time. It’s important to note that it’s likely to be more complicated than that, though, because whether people recover, the degree to which they recover is going to relate to how ill they were, the impact that there’s been on them and also things like their age.

SREENIVASAN: Here you are doing this research, here are your colleagues doing this research and you’re finding things that, for you, are concerning, for you that say require more research, and outside, you’ve got people who are still resistant to a vaccine, who don’t think this is real, who don’t think this will have any long-term effects on them, who feel invincible, who think this is the flu. I mean, when you meet your friends and family and they figure out what you do, what do you say to them?

HAMPSHIRE: Well, I always say the same thing essentially as it relates to our data, which is we see quite a clear relationship, whereby if a person was more ill, then we tend to see a greater degree of cognitive underperformance. Now, if you have the vaccine, it is either going to stop you getting ill or it’s going to greatly reduce the severity of the illness.

SREENIVASAN: Dr. Adam Hampshire of the Imperial College London, thanks so much for joining us.

HAMPSHIRE: Thank you.