In this study, we present data from a cohort of 485 individuals who reported having a confirmed positive COVID-19 test, and who were assessed using a comprehensive and widely validated battery of cognitive tests that measures aspects of memory, attention, verbal abilities, problem-solving and reasoning. Cognitive scores in multiple broad domains were related to participants’ self-reported COVID-19 physical and mental health experiences, including infection severity, extent of recovery, and measures of anxiety and depression.
The results unequivocally confirm the existence of cognitive impairments in the aftermath of COVID-19 infection. There are several important novel findings here. First, there is striking domain specificity of the impairments. Speed of processing was most markedly impaired, with verbal abilities, reasoning and global cognition scores also impaired, whereas a measure of memory performance was unaffected. Second, when all physical, cognitive and mental health factors were taken into account, two distinct subjective symptom patterns emerged. On the one hand, there exists a collection of ‘physical symptoms’, including fatigue, pain and limitations in performing everyday physical activities, that tend to vary together and are strongly associated with COVID-19 infection severity. Thus, unsurprisingly, more severe disease and older age are associated with poorer physical well-being post-infection. On the other hand, there exists a second set of ‘mental health’ symptoms that include depression, anxiety and self-reported limitations in emotional well-being that tend to co-occur and are unrelated to disease severity. Third, the cognitive deficits are strongly and consistently associated with the physical sequelae of the disease, rather than the mental health symptoms. That is, better physical health was correlated with faster processing speed, better verbal ability, and overall cognitive performance, while no associations were found between these measures and the mental health & wellness factor.
Given the global nature of the COVID-19 pandemic and the fact that it has disproportionately (but not exclusively) affected the elderly and those from low educational and SES backgrounds, it is of interest to explore how these factors interact with the physical and mental health outcomes that we have identified. In our primary analyses, age, sex, post-secondary education, and SES were accounted for by adjusting cognitive scores for the known effects of these variables (estimated from a large control sample), yet even when these factors were included as covariates in a supplementary analysis the same pattern of results was observed, confirming that the observed relationships between physical health and cognitive scores were not driven by a residual effect of any these factors. Furthermore, the lack of any correlations between these socio-demographic variables and corrected cognitive scores (Figure S4) suggests that this relationship is a core characteristic of COVID-19 infection, rather than a secondary effect related to the demographic profile of those who have been most commonly infected.
It is important to understand that we are not describing two types of people in the post COVID-19 infection population, but two distinct factors that contribute to and characterise the post COVID-19 syndrome. Indeed, the fact that these measures were clearly dissociable in terms of the demographic variables that they correlated with, suggests that they represent two distinct and separable, though not mutually exclusive, effects of COVID-19 infection. For example, our linear regression models (Table S4) showed that physical health was negatively correlated with age (with an average decline of approximately 0.1 SDs per decade), whereas mental health & wellness increased with age (by 0.1 SDs for every 10 years). While it is perhaps to be expected that older COVID-19 survivors would be most affected in terms of their physical and cognitive outcomes (given the greater likelihood of co-morbidities in that group), the fact that it is the young that have been most severely affected in terms of their mental health and well-being is surprising. Completion of post-secondary education was also associated with significantly better mental health, but not physical health outcomes, and males reported better physical health, yet there was no difference between males and females in terms of mental health & wellness. Again, the differing patterns of correlations between socio-demographic variables and the physical and mental health factors further confirm the existence of two distinct outcomes of COVID-19 infection that are dissociable in multiple ways.
The fact that measures of mental health such as anxiety and depression were not associated with cognitive outcomes in the context of the COVID-19 pandemic is surprising, as numerous studies have shown an association between anxiety, depression and cognition in the pre-pandemic era. For example, one study of 4582 participants30 showed that generalized anxiety is associated with memory and verbal fluency deficits, particularly in young adults. Similarly, a systematic review and meta-analysis by Semkovska et al.31 confirmed that deficits in selective attention, working memory and long-term memory persist in major depression and worsen with repeated episodes. However, it is important to clarify that these, and most other studies that have examined the relationship between mental health and cognition, have focussed on clinical populations; that is, patients who have been diagnosed with a major mental health condition, such as depression or anxiety. In our study, the fact that no association was observed between measures of mental health and cognition may be due to a predominance of detectible, yet sub-clinical, mental health issues among the COVID-19 survivors. For instance, we found that only about one third of our participants had probable anxiety or depression, which is consistent with estimates of the prevalence of these disorders in the general population during the COVID-19 pandemic43. Regardless, our study highlights the importance of carefully examining the relationship between physical wellness, mental health and cognition in other patient populations, to determine what is driving any observed cognitive impairments.
Based on our data it is not possible to disentangle the effects on mental health of COVID-19 infection and the pandemic more generally. Factors such as job security, financial instability, home-schooling, social isolation and an elevated sense of community fear have undoubtedly affected the mental well-being of people throughout the world, irrespective of whether they have received a positive COVID-19 diagnosis. Theoretically, the same argument could be made to explain some of our physical and cognitive outcome measures; increased fatigue, poorer subjective memory, and even impaired reasoning and processing speed might occur, to some extent, even in the general non-COVID + ve population during the pandemic. However, two observations suggest otherwise: first, the fact that the severity of these deficits was tightly coupled to the severity of COVID-19 symptoms, whereas no such relationship existed between WHO severity ratings and mental-health scores, and second, participants who had better than average physical health scores (i.e., were relatively unaffected by the infection in physical terms) performed similarly to pre-pandemic controls in all cognitive domains, whereas cognitive impairments were only seen in those participants who reported poor physical health and more severe illness. Taken together, these results suggest that the physical/cognitive profile identified here is one that is specifically related to COVID-19 infection itself.
We further explored whether the observed pattern of cognitive deficits was explained by the more severe cases of illness that required hospitalisation. Several preliminary studies have suggested that cognitive impairments following COVID-19 infection are dependent on the level of medical assistance received1,22 although at least one study has reported no correlation between hospitalisation and cognitive impairments24. This is an important issue because long-lasting cognitive deficits have been reported in non-COVID-19 patients following treatment in the ICU, suggesting that factors such as mechanical ventilation, sedation, drug therapy and disturbed sleep may all contribute to the emergent cognitive profile, independent of infection7 In the current study, direct comparisons of the hospitalised and non-hospitalised post COVID-19 subgroups against the controls showed that both groups had significant cognitive impairments. Moreover, the groups did not differ significantly on any measure of cognitive outcome, although numerically the hospitalised group performed worse than the non-hospitalised group in terms of reasoning, verbal ability, processing speed and overall cognitive performance (Fig. 5b). Given this trend, we included hospitalisation in our regression analyses to see if it explained away or reduced the observed relationship(s) between physical health and cognitive performance. The results showed that physical health predicted cognitive performance even when mental health measures and hospitalisation were accounted for, highlighting that the effects of COVID-19 on cognition are not entirely driven by hospitalisation. This is an important observation, given that most studies of the prolonged effects of COVID-19 have focused on hospitalised cohorts.
Although the current study provides clear evidence for a broad pattern of cognitive impairment following COVID-19 infection, the effect was, at least to some extent, domain specific. That is to say, notable impairments were found relative to controls in speed of processing and in the reasoning and verbal domains, but not in STM performance. It is important to point out that this relationship is not absolute, due to the nature of the three main factors that were extracted from performance across a diverse range of 12 cognitive tests. Specifically, we use the term STM descriptively to refer to a single factor derived from performance across the entire battery of tests, and this will include some contributions from tests that do not ostensibly measure aspects of STM. Nevertheless, as we have shown previously42, and convincingly replicate here (Fig. 1A, Table S3), this factor is driven primarily by tests such as ‘spatial span’, ‘monkey ladder’, ‘token search’ and ‘paired associates’, all of which are derived from standard measures of STM (for details, including references for the original forms of the tasks on which the computerized versions are based, see Hampshire et al.42 or Wild et al.44). This finding sheds some light on the nature and extent of the subjective experience of COVID-19 survivors, often called ‘long COVID’ or ‘brain fog’ - the expression now used widely to describe the subjective sense of cognitive impairment following COVID-19 infection. Specifically, that ‘brain fog’ in this context includes processing (or ‘thinking’) speed, reasoning and verbal abilities, but leaves short term memory relatively spared.
We are hesitant to draw conclusions about the underlying neural systems responsible for this pattern of impairments post COVID-19 infection, but an fMRI-based brain parcellation based on the same 12 tests used in this study may shed some preliminary light on this question. Hampshire et al. (2012) found that the STM factor, which was resilient to the effects of COVID-19 infection in the current study, had a brain network analogue that included the insula/frontal operculum, the superior frontal sulcus, and the ventral portion of the anterior cingulate cortex/ presupplementary motor area. The reasoning and verbal factors – which, in contrast, did show impaired performance in the COVID + sample – were associated with two non-overlapping networks, composed of: (reasoning) the inferior frontal sulcus, the inferior parietal cortex, and the dorsal portion of the anterior cingulate cortex/ presupplementary motor area, and (verbal) the left inferior frontal gyrus and bilateral superior temporal regions. Although there is no obvious explanation for why these two particular networks might be affected by COVID-19 infection while others are spared, future work could use neuroimaging to investigate the hypothesis that some brain regions or networks, and the cognitive systems they support, are more susceptible to damage from COVID-19 infection.
Regardless of the underlying neuropathological substrates, the functional dissociation observed in this study is important for understanding how COVID-19 related ‘brain fog’ differs from other conditions and circumstances that have been described using similar terms, such as post-chemotherapy cognitive dysfunction (‘chemo brain’), cognitive impairment following heart bypass surgery (‘pump head’) and disturbed cognition after sleep deprivation. Regarding the latter, in a recent study of more than 10,000 participants assessed using the same battery of 12 tests used in the current study (and from which the current control sample was drawn) we showed that typical sleep duration also had no bearing on short-term memory performance, unlike reasoning and verbal skills, which were impaired by both too little, or too much, sleep on a regular basis45.
A key unanswered question remains the longitudinal nature of these problems. In those patients who are most affected, are these cognitive disturbances temporary, permanent, or do they signal the onset of a neurodegenerative cascade that results in a deteriorating pattern of impairment over time? While it is too early to answer this question in full, some preliminary clues can be gleaned from our data analysis. For example, the approximate time since infection demonstrated negligible correlations with all measures (Figs. 1B, S2), suggesting that cognitive impairments do not get better or worse over the short term. The current cohort will be followed and retested regularly for at least a year, and possibly longer, and until that process is complete the longitudinal trajectory of post COVID-19 cognitive impairment will remain unclear.
Finally, this study illustrates that is possible to acquire comprehensive, high quality cognitive assessments remotely without any inter-personal contact, which is essential in a pandemic situation involving a highly contagious virus. Online data collection also allowed us to reach a broader population than would have been possible by approaching patients from associated health networks and will afford us the opportunity to easily follow our cohort longitudinally.
In conclusion, we have shown clear cognitive impairments following COVID-19 infection. These are likely not the result of a “global” impact on cognitive processing, as STM performance was relatively preserved. Crucially, in the domains that were affected, cognitive performance was related to measures of physical health, including COVID severity, but not mental health. This has implications from a clinical viewpoint, as survivors who exhibit increased anxiety or depression may or may not have cognitive deficits, whereas these are much more likely in patients who experience a greater physical toll from the illness. Our findings underscore the fact that the physical, emotional, mental, and cognitive sequelae of COVID-19 are not bound together as a single neurocognitive syndrome.