COVCOG 3--Trajectory of Long COVID: Longitudinal Changes in Symptoms and Cognitive Impairment. A Third Publication from the COVID and Cognition Study
Saved in:
| Title: | COVCOG 3--Trajectory of Long COVID: Longitudinal Changes in Symptoms and Cognitive Impairment. A Third Publication from the COVID and Cognition Study |
|---|---|
| Language: | English |
| Authors: | Sabine P. Yeung (ORCID |
| Source: | Applied Cognitive Psychology. 2025 39(2). |
| Availability: | Wiley. Available from: John Wiley & Sons, Inc. 111 River Street, Hoboken, NJ 07030. Tel: 800-835-6770; e-mail: cs-journals@wiley.com; Web site: https://www.wiley.com/en-us |
| Peer Reviewed: | Y |
| Page Count: | 22 |
| Publication Date: | 2025 |
| Document Type: | Journal Articles Reports - Research |
| Descriptors: | COVID-19, Pandemics, Chronic Illness, Neurological Impairments, Thinking Skills, Daily Living Skills, Employment Problems, Health, Symptoms (Individual Disorders), Improvement |
| DOI: | 10.1002/acp.70040 |
| ISSN: | 0888-4080 1099-0720 |
| Abstract: | Long COVID has widespread and long-lasting multisystemic impacts on patients' bodies, cognition, and daily functioning, including the ability to work. Longitudinal studies are important in investigating the expected timelines along the course of recovery. This mixed cross-sectional/longitudinal study examines how symptoms (cognitive and noncognitive) and objective cognitive function evolve in post-COVID-19 patients (n = 187) compared to noninfected controls (n = 207). Participants completed a questionnaire about their COVID-19 experience and cognitive tasks at baseline and again at 2-3 follow-ups during a 9-month period. While some noncognitive symptoms improved over time (ds = 0.34-0.87), cognitive symptoms and neurological symptoms, as well as memory function assessed with objective cognitive assessments, remained unimproved (nonsignificant change over time). Neurological symptoms predicted both cognitive symptoms and cognitive impairment across time. Our finding suggested that people with past COVID-19 infection did not experience improvement in cognitive function over time, at least for the duration of this 9-month longitudinal study. |
| Abstractor: | As Provided |
| Entry Date: | 2025 |
| Accession Number: | EJ1468390 |
| Database: | ERIC |
|
Full text is not displayed to guests.
Login for full access.
|
|
| FullText | Links: – Type: pdflink Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwH_0SGePqKk4C3IZcixzBw2AAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDCbYiHC_na7Jk-O3dQIBEICBm6oftBcFfNAbFOJegb0rJtltEMcYa5xg9Z2hhX7-cT9zat9zz8y91s4Js0xx-QVFtdlKItnlR4kD5fj5bRXM4-_rYX4wBblVzasf_pAATXW4yWgVExDCSpZ5hQSkbVU-p5lIL56acP6GSLg3EMCR9JZHDxuncRGn9lwc0mJo2bGOvBApHDwnSafjqUrtV2FrhuSmHxmnHyr03VEJ Text: Availability: 1 Value: <anid>AN0184623586;bu801mar.25;2025Apr23.01:41;v2.2.500</anid> <title id="AN0184623586-1">COVCOG 3—Trajectory of Long COVID: Longitudinal Changes in Symptoms and Cognitive Impairment. A Third Publication From the COVID and Cognition Study </title> <p>Long COVID has widespread and long‐lasting multisystemic impacts on patients' bodies, cognition, and daily functioning, including the ability to work. Longitudinal studies are important in investigating the expected timelines along the course of recovery. This mixed cross‐sectional/longitudinal study examines how symptoms (cognitive and noncognitive) and objective cognitive function evolve in post‐COVID‐19 patients (n = 187) compared to noninfected controls (n = 207). Participants completed a questionnaire about their COVID‐19 experience and cognitive tasks at baseline and again at 2–3 follow‐ups during a 9‐month period. While some noncognitive symptoms improved over time (ds = 0.34–0.87), cognitive symptoms and neurological symptoms, as well as memory function assessed with objective cognitive assessments, remained unimproved (nonsignificant change over time). Neurological symptoms predicted both cognitive symptoms and cognitive impairment across time. Our finding suggested that people with past COVID‐19 infection did not experience improvement in cognitive function over time, at least for the duration of this 9‐month longitudinal study.</p> <p>Keywords: cognition; COVID‐19; executive functions; language; long COVID; memory; neurological; symptoms</p> <hd id="AN0184623586-2">Introduction</hd> <p>Long COVID, or "Post‐COVID‐19 syndrome," is recognized as the persistence of COVID‐19 symptoms for more than 12 weeks following infection (UK National Institute for Health and Care Excellence [NICE] [<reflink idref="bib45" id="ref1">45</reflink>]). It is estimated that nearly 2 million people in the UK (UK Office for National Statistics [ONS] [<reflink idref="bib49" id="ref2">49</reflink>]) are now experiencing some version of this condition, in 69% of those at least 1 year, and 41% at least 2 years since infection. Crucially, Long COVID can occur regardless of the severity of initial infection (O'Mahoney et al. [<reflink idref="bib47" id="ref3">47</reflink>]): The majority of Long COVID cases are found in nonhospitalized patients having had a mild acute illness (FAIR Health [<reflink idref="bib16" id="ref4">16</reflink>]). Given this alarming prevalence, Long COVID has been referred to as "the pandemic after the pandemic" (Levine [<reflink idref="bib37" id="ref5">37</reflink>]). At barely 4 years since the emergence of the virus, research on the trajectory in long‐term sufferers is vital. This "COVCOG 3" study applied mixed cross‐sectional/longitudinal design to track how physical and cognitive symptoms and objective cognitive function change over time in post‐COVID‐19 patients, compared to noninfected controls, over a 9‐month follow‐up period.</p> <p>Long COVID patients experience symptoms affecting multiple organ systems (Davis et al. [<reflink idref="bib12" id="ref6">12</reflink>]; Raveendran et al. [<reflink idref="bib55" id="ref7">55</reflink>]). For example, Davis et al. ([<reflink idref="bib11" id="ref8">11</reflink>]) found that systemic symptoms such as fatigue were present in almost all of the 3762 participants surveyed, while 85% also experienced others, such as cardiovascular (e.g., palpitations) and neuropsychiatric symptoms (e.g., cognitive dysfunction, sensorimotor problems, issues with emotion, and mood). Systemic and cognitive/neurological symptoms were still present at 7 months following infection. These findings are consistent with our own ("COVCOG 1": Guo et al. [<reflink idref="bib21" id="ref9">21</reflink>]). Indeed, meta‐analysis (Ceban et al. [<reflink idref="bib9" id="ref10">9</reflink>]) confirms that fatigue and cognitive symptoms are among the most common and debilitating for Long COVID patients.</p> <p>There is considerable evidence to suggest that post‐COVID illness can be long‐lasting. For example, 42% of the 33,281 patients studied in Hastie et al.'s ([<reflink idref="bib26" id="ref11">26</reflink>]) longitudinal study reported incomplete recovery 6–18 months post‐infection, with the proportion reporting no recovery reaching 16% in those who had been hospitalized. Similarly, 17% of the 1106 patients surveyed by Ballouz et al. ([<reflink idref="bib5" id="ref12">5</reflink>]) reported not being recovered at 24 months. Recovery appears to follow different trajectories for different types of symptoms, with symptoms such as sore throat resolving quickly (Huang et al. [<reflink idref="bib30" id="ref13">30</reflink>]; Xiong et al. [<reflink idref="bib62" id="ref14">62</reflink>]), while fatigue and cognitive dysfunction linger as long as 1–2 years (Dennis et al. [<reflink idref="bib14" id="ref15">14</reflink>]; Huang et al. [<reflink idref="bib31" id="ref16">31</reflink>]; Petersen et al. [<reflink idref="bib51" id="ref17">51</reflink>]). Indeed, recent review papers highlight the persistence of cognitive problems in Long COVID patients—which were less likely to improve if the chronic symptoms lasted more than 12 months—and the frustration these patients experienced (Ladds et al. [<reflink idref="bib34" id="ref18">34</reflink>]; Li et al. [<reflink idref="bib38" id="ref19">38</reflink>]). This raises the possibility that COVID‐19 infection may increase vulnerability to long‐term or permanent cognitive decline, neurodegeneration, and dementia (Bougakov et al. [<reflink idref="bib7" id="ref20">7</reflink>]; Goldberg et al. [<reflink idref="bib19" id="ref21">19</reflink>]; Miners et al. [<reflink idref="bib41" id="ref22">41</reflink>]).</p> <p>Cognitive problems in Long COVID are diverse. In Davis et al. ([<reflink idref="bib11" id="ref23">11</reflink>]), 85.1% of participants with illness lasting longer than 28 days reported cognitive dysfunction or "brain fog," 72.8% reported memory impairments, both short‐ and long‐term, and 46.3% had "tip of the tongue" word‐finding difficulties. Among participants at month 7, over half still reported cognitive dysfunction and memory impairments. Consistent with this, the most frequently reported cognitive symptoms in our cohort at baseline assessment ("COVCOG 1") at week 3–31+ included difficulty concentrating (77.8%), brain fog (69%), forgetfulness (67.5%) and tip‐of‐the‐tongue problems (59.5%) (Guo et al. [<reflink idref="bib21" id="ref24">21</reflink>]). Other studies have also shown that these and language and executive function issues are commonly reported in Long COVID (Nalbandian et al. [<reflink idref="bib44" id="ref25">44</reflink>]; Ziauddeen et al. [<reflink idref="bib65" id="ref26">65</reflink>]).</p> <p>Studies employing objective screening tools such as the Montreal Cognitive Assessment (MoCA) detected even higher rates of cognitive impairment in Long COVID than subjective reports seemed to imply (Ceban et al. [<reflink idref="bib9" id="ref27">9</reflink>]; Davis et al. [<reflink idref="bib12" id="ref28">12</reflink>]). Hampshire et al. ([<reflink idref="bib25" id="ref29">25</reflink>]) studied cognitive test performance in 84,285 individuals from the online Great British Intelligence Test. Even those who were considered to have recovered from COVID‐19 (no longer reporting respiratory symptoms) showed significant deficits in semantic problem solving, spatial working memory, and selective attention tasks compared to uninfected controls. Consistent results were observed in a later study from this research group investigating over 112,000 participants: Those with unresolved persistent symptoms demonstrated large deficits in tasks assessing memory, reasoning, and executive function (Hampshire et al. [<reflink idref="bib23" id="ref30">23</reflink>]). Similarly, in our COVCOG sample, we found evidence for reduced memory performance, driven by impairment in word recognition memory in those with past COVID‐19 infection compared to uninfected individuals ("COVCOG 2": Guo et al. [<reflink idref="bib22" id="ref31">22</reflink>]). Other studies have found a similar pattern of findings, with memory, language, executive function, and response speed being particularly affected in post‐COVID‐19 patients (Becker et al. [<reflink idref="bib6" id="ref32">6</reflink>]; Graham et al. [<reflink idref="bib20" id="ref33">20</reflink>]; Herrera et al. [<reflink idref="bib28" id="ref34">28</reflink>]; Miskowiak et al. [<reflink idref="bib42" id="ref35">42</reflink>]; Zhao et al. [<reflink idref="bib64" id="ref36">64</reflink>]; see also Crivelli et al. [<reflink idref="bib10" id="ref37">10</reflink>] for a systematic review and meta‐analysis).</p> <p>Cognitive issues are probably underpinned by neurological factors. Neurological symptoms are reported by a large proportion of patients (Mao et al. [<reflink idref="bib40" id="ref38">40</reflink>]; Pryce‐Roberts et al. [<reflink idref="bib53" id="ref39">53</reflink>]; Romero‐Sánchez et al. [<reflink idref="bib57" id="ref40">57</reflink>]). For example, anosmia is one of the most recognized and widely experienced symptoms of COVID‐19 (e.g., Lechien et al. [<reflink idref="bib35" id="ref41">35</reflink>]; Mao et al. [<reflink idref="bib40" id="ref42">40</reflink>]; Romero‐Sánchez et al. [<reflink idref="bib57" id="ref43">57</reflink>]). These neurological symptoms can persist months after the initial illness (Taruffi et al. [<reflink idref="bib61" id="ref44">61</reflink>]). There is mounting evidence that many COVID‐19 patients display signs of neurological deficits and abnormal neural activity (Galanopoulou et al. [<reflink idref="bib17" id="ref45">17</reflink>]; Helms et al. [<reflink idref="bib27" id="ref46">27</reflink>]; Kandemirli et al. [<reflink idref="bib33" id="ref47">33</reflink>]). A longitudinal study of pre‐ and post‐infection anatomical magnetic resonance imaging (MRI) brain scans in 394 COVID‐19 patients showed abnormalities across multiple regions including the left parahippocampal gyrus, orbitofrontal cortex and cingulate cortex, and right hippocampus (Douaud et al. [<reflink idref="bib15" id="ref48">15</reflink>]). These accounts of neurological involvement in COVID‐19 also seem to translate into impaired performance on cognitive tests: A PET study of 29 older patients presenting with neurological symptoms found that the degree of frontoparietal hypometabolism was associated with lower neuropsychological scores, particularly in tests of verbal memory and executive functions (Hosp et al. [<reflink idref="bib29" id="ref49">29</reflink>]).</p> <p>Various mechanisms have been proposed to explain these COVID‐19 induced neurological conditions and cognitive impairments. These primarily involve abnormal inflammation or coagulation (Nalbandian et al. [<reflink idref="bib44" id="ref50">44</reflink>]). Long‐lasting hypoxia and disruption of the blood–brain barrier through systemic inflammation and cytokine release have been implicated in severe acute COVID‐19, which typically leads to neuroinflammation and long‐term cognitive deficits (Baker et al. [<reflink idref="bib4" id="ref51">4</reflink>]; Nalbandian et al. [<reflink idref="bib44" id="ref52">44</reflink>]). Cognitive impairment could arise from hippocampal and cortical impact associated with neuroinflammation (Steardo et al. [<reflink idref="bib59" id="ref53">59</reflink>]). Additionally, hypercoagulation in severe COVID‐19 is associated with delirium and cognitive deficits (Baker et al. [<reflink idref="bib4" id="ref54">4</reflink>]), while microvascular endothelial damage also correlates with hemorrhagic or ischemic brain lesions in patients with COVID‐19 (Østergaard [<reflink idref="bib50" id="ref55">50</reflink>]).</p> <p>The consequences of chronic Long COVID can be life‐altering. The ONS ([<reflink idref="bib49" id="ref56">49</reflink>]) reports 79% of people with Long COVID experience negative effects on daily activities, which were severe in 20% of cases. Our "COVCOG 1" study found that 63.9% of participants with ongoing symptoms reported difficulty coping with day‐to‐day activities, 54.6% had experienced long periods unable to work, and 34.5% had lost their job due to illness. These impacts scaled with symptom severity (Guo et al. [<reflink idref="bib21" id="ref57">21</reflink>]). Specifically, Davis et al. ([<reflink idref="bib11" id="ref58">11</reflink>]) reported that 86.2% of working respondents felt mildly to severely unable to work specifically because of cognitive dysfunction. In a study with 1683 individuals following COVID‐19 infection, those reporting more daily cognitive symptoms were more likely to experience at least moderate interference with daily functioning and less likely to have full‐time employment (Jaywant et al. [<reflink idref="bib32" id="ref59">32</reflink>]). As Long COVID becomes a prevalent social problem with far‐reaching implications, research is needed to ascertain patterns of development in Long COVID experience over time. Longitudinal studies could provide valuable information on expected timelines along the course of recovery, if any.</p> <p>There is—to date—little test‐based longitudinal research on cognitive function over time in Long COVID. Among these very few studies, Del Brutto et al. ([<reflink idref="bib13" id="ref60">13</reflink>]) evaluated cognitive function in mild COVID‐19 patients before and at 6 and 18 months after infection using the MoCA. A significant decline in MoCA scores was observed at 6 months relative to pre‐infection, but these improved again at 18 months, suggesting at least a partial recovery. Similarly, Afzali et al. ([<reflink idref="bib2" id="ref61">2</reflink>]) found that across three 45‐day intervals, performance on the Paced Auditory Serial Addition Test (PASAT) improved across all patients, but that response speed slowed over time in those who had suffered severe acute illness. This evidence of reduced recovery in those with more severe initial illness is reflected in evidence in hospitalized patients: Using the Cognitive Impairment in Psychiatry (SCIP) assessment, Miskowiak et al. ([<reflink idref="bib43" id="ref62">43</reflink>]) demonstrated that degraded performance on memory and processing speed did not show significant change between 3 months and 1 year post infection.</p> <p>The two previous papers from the COVID and Cognition study (COVCOG) used the baseline data from this mixed cross‐sectional/longitudinal study ("COVCOG 1": Guo et al. [<reflink idref="bib21" id="ref63">21</reflink>]; "COVCOG 2": Guo et al. [<reflink idref="bib22" id="ref64">22</reflink>]). This third longitudinal paper ("COVCOG 3") documents how symptoms (both cognitive and noncognitive) and objectively measured cognitive function change over time post‐COVID. Note that our samples were recruited from the community (such as through social media and word of mouth), with the majority being nonhospitalized patients. Given previous evidence, we form the following hypotheses:</p> <hd id="AN0184623586-3">1 Hypothesis</hd> <p> <emph>Changes in COVID‐19 symptoms over time. We expect that different symptom areas will follow divergent recovery trajectories, depending on general severity of ongoing illness at baseline. Specifically, we predict that cognitive and neurological symptoms will be persistent, while others may recede more rapidly</emph>.</p> <hd id="AN0184623586-4">2 Hypothesis</hd> <p> <emph>Changes in cognitive performance over time. We expect the slower recovery in cognitive and neurological symptoms to be reflected in objective cognitive measures of learning and memory, with the deficits in memory evident at baseline (Guo et al.</emph> [<reflink idref="bib22" id="ref65">22</reflink>]<emph>) reducing but not disappearing during the follow‐up period</emph>.</p> <hd id="AN0184623586-5">3 Hypothesis</hd> <p> <emph>Predictors of cognitive symptoms and cognitive performance. We investigate the relationship between noncognitive symptoms experienced at different stages of the illness and cognitive symptoms and performance. While we expect to see numerous relationships, we specifically predict that neurological symptoms during the early illness will continue to predict individual differences in cognitive performance across multiple time‐points</emph>.</p> <hd id="AN0184623586-6">Method</hd> <p></p> <hd id="AN0184623586-7">Participants</hd> <p>Over the data collection period from October 2020 to July 2022, we recruited 476 adults (321 [67.4%] female) from communities in majority English‐speaking countries. The baseline data for 421 participants (181 in the COVID group, 185 in the No‐COVID group, 55 with an unknown infection status) and reasons for noninclusion and detailed recruitment procedure were reported previously (Guo et al. [<reflink idref="bib21" id="ref66">21</reflink>], [<reflink idref="bib22" id="ref67">22</reflink>]). This paper uses cases with sufficiently complete data on the final baseline dataset from 187 participants in the COVID group and 207 in the No‐COVID group.</p> <p>The COVID baseline group was further divided into three subgroups according to their severity of ongoing illness experienced up to the time of baseline assessment. At the baseline, participants who had COVID were asked to describe their experience of the infection after the first 3 weeks of acute illness. Those who reported having been "never ill" or "feeling better" were assigned to the "Recovered" subgroup (<emph>R</emph>; <emph>n</emph> = 48). Those who reported having "ongoing mild/moderate symptoms" were assigned to the "Ongoing–Mild/Moderate" subgroup (<emph>C+</emph>; <emph>n</emph> = 61). Those who reported having "ongoing severe symptoms" were placed in the "Ongoing–Severe" subgroup (<emph>C++</emph>; <emph>n</emph> = 78). Figure 1 shows full details of participant grouping (see also Guo et al. [<reflink idref="bib21" id="ref68">21</reflink>], [<reflink idref="bib22" id="ref69">22</reflink>] and Figure S1). Attrition statistics are reported with particular analyses in Section 3. Note that the baseline sample is slightly larger in the current study than in the publications on that data (COVCOG1 and COVCOG2) due to the inclusion of participants who completed the session after publication.</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01mar25/acp70040-fig-0001.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70040-fig-0001.jpg" title="1 Study procedural flow and participant grouping at baseline." /> </p> <p></p> <hd id="AN0184623586-9">Procedures</hd> <p>The COVCOG study was conducted through Gorilla (<ulink href="http://www.gorilla.sc">www.gorilla.sc</ulink>). Informed consent was obtained from participants prior to testing at each session. The baseline measures included a questionnaire and cognitive tests, which took approximately 1 h to complete. At the end of the baseline session, participants were asked to indicate whether they were willing to be contacted for a follow‐up. Those who fully completed the baseline session (the questionnaire and all cognitive tests) and indicated their wish to continue participation were sent an invitation via email inviting them to complete a follow‐up 6 weeks later. Only those within the COVID and No‐COVID groups were followed up.</p> <p>In our original design, the No‐COVID group was only asked a short follow‐up question "have you had COVID yet?" at each follow‐up in order to check whether they developed COVID; if yes, they would join the COVID group for the follow‐ups. However, at the time of starting our second follow‐up for the COVID group, we changed the recruitment strategy and invited both the COVID and No‐COVID groups to complete full follow‐up measures. This was due to a change in focus, as understanding of the condition grew between 2020 and 2021, from solely attempting a before‐after infection comparison, to longitudinal comparison of performance and experiences between infected and uninfected groups. The No‐COVID group thus experienced one fewer follow‐up session. At the follow‐up, participants completed questionnaire and cognitive tests similar to, but shorter than, those in the baseline session—lasting approximately 35 min. At the end of the follow‐up session, participants were asked whether they were willing to be contacted again. This process gave four sessions for the COVID group (baseline plus three follow‐ups) and three sessions for the No‐COVID group (baseline plus two follow‐ups) across a 9‐month period, with approximate inter‐session intervals of 10–20 weeks.</p> <hd id="AN0184623586-10">Measures</hd> <p></p> <hd id="AN0184623586-11">Questionnaire</hd> <p>At baseline, the questionnaire covered demographics (e.g., sex, age, education level, country of residence, ethnicity, profession, and employment status) and health. The health information included height and weight, medical history and health‐related behaviors, such as usual diet intake, use of tobacco and alcohol and physical activity level as well as experience of COVID‐19. COVID experience was scored as infection status, severity of illness and experience of individual symptoms during the first 3 weeks of infection ("initial illness"; 36 symptoms on a 3‐point scale from 1 = <emph>Not at all</emph> to 3 = <emph>Very severe</emph>), since the first 3 weeks ("ongoing illness"; 52 symptoms on a 5‐point scale from 1 = <emph>Not at all</emph> to 5 = <emph>Very severe and often</emph>) and in the past 1–2 days ("current symptoms"; 52 symptoms; dichotomously: yes/no). Participants who had not been infected with COVID‐19 were not asked about the severity and symptoms of COVID‐19, because we were not at first comparing symptom prevalence between the groups. When vaccination became available from December 2020, the questionnaire also asked about vaccination status (including the number of dose(s) received, timing of the most recent dose and the type of vaccine). For full details of the baseline questionnaire, see "COVCOG 1": Guo et al. ([<reflink idref="bib21" id="ref70">21</reflink>]).</p> <p>The follow‐up questionnaires were similar to the baseline questionnaire, but shorter, omitting demographics, medical history and reports of acute‐period of COVID‐19 symptoms (i.e., "initial illness"). At each follow‐up, the questionnaire invited participants to provide updates of their vaccination status and any new COVID‐19 infections. The COVID group was asked to report on their ongoing ("since the last session") and current ("in the past 1–2 days") symptoms and illness experience. If a participant in the No‐COVID group reported a new infection, they entered the same flow of questions as the COVID group during the baseline. If they remained uninfected, they were also asked to report on their experience of ongoing and current symptoms as the COVID group did at this time—to provide an uninfected control level for each symptom.</p> <hd id="AN0184623586-12">Cognitive Tests</hd> <p>After filling the questionnaire, participants completed a set of cognitive tests assessing memory, language, and executive function. At baseline, these tests included a Word List Recognition Memory Test (Figure 2a), a Pictorial Associative Memory Test (Figure 2b) to assess memory, a Category Fluency Test (Figure 2c) to assess language ability, the Wisconsin Card Sorting Test [WCST] (Figure 2d) and a 2D Mental Rotation Test to assess executive function. The tasks chosen reflected a combination of well‐established tasks that covered memory, language, and executive functions but that were also at that time available on the Gorilla platform. Additionally, a Number Counting Test was included as an attention/"bot" check for data quality, whilst a Relational Reasoning Test was given only to the No‐COVID group as a means by which to IQ‐match control participants for potential pre‐post infection longitudinal explorations. Data from these latter two tasks are not within the scope of this paper. The presentation order of the cognitive tests was counterbalanced for each participant. More detailed descriptions of these tasks were given previously (Guo et al. [<reflink idref="bib22" id="ref71">22</reflink>]). At the first follow‐up session, participants had completed these tests again. At subsequent follow‐up session(s), to reduce participant burden, the same set of cognitive tests was implemented, but without the 2D Mental Rotation Test (precluding examination of further change in performance on that task). The stimuli and their presentation order in the cognitive tests differed between sessions (see Supporting Information for information on cognitive tests at baseline and follow‐ups).</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01mar25/acp70040-fig-0002.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70040-fig-0002.jpg" title="2 Cognitive tests assessing memory (Word List Recognition Memory Test [a], Pictorial Associative Memory Test [b]), language (Category Fluency Test [c]), and executive function (Wisconsin Card Sorting Test [d])." /> </p> <p></p> <hd id="AN0184623586-14">Data Processing and Analysis</hd> <p>Here we present an overview of our analysis plan; full details are documented in the Supporting Information. Statistical analyses were conducted using IBM SPSS Statistics. As a preliminary, we performed attrition analysis to understand how attrition could have shaped our samples (Sections 3.1.1–3.1.2). Numerous tests were used to compare sample characteristics, including chi‐square tests, Mann–Whitney <emph>U</emph> tests, and ANOVAs.</p> <p>To examine changes in COVID‐19 symptoms (Hypothesis 1) and cognitive performance (Hypothesis 2) over time, we grouped individual symptom/task variables into factors by exploratory factor analyses (EFA) (see Sections 3.2 and 3.3 and Tables S10–S12 for factor definitions as previously reported). Mixed between‐subject (groups) and within‐subject (time) ANOVAs controlling for sex, age, education, and assessment interval(s) were performed on these factors to compare within‐subject changes between groups, that is, whether and how the COVID and No‐COVID groups may differ in symptoms and cognitive performance over time (Time X Group interaction).</p> <p>Finally, in participants with past COVID‐19 infection, we applied multiple linear regressions with forward stepwise variable introduction to explore the relationship between noncognitive symptoms, cognitive symptoms, and cognitive impairment at various stages of illness. This analysis was conducted both at individual assessment time points and across time (Hypothesis 3). Sex, age, and education were controlled for in all regression models; thus, adjusted <emph>R</emph> square change (∆<emph>R</emph><sups>2</sups><subs>adj</subs>) is reported to reflect the additional variance explained by the target predictor(s).</p> <p>Across all analyses, Sidak correction for multiple comparisons was employed, and the Sidak α is quoted where appropriate. Only results surviving correction are reported as significant in the main text. All statistics are recorded in the Supporting Information.</p> <hd id="AN0184623586-15">Results</hd> <p></p> <hd id="AN0184623586-16">Sample Characteristics at Baseline and Follow‐Ups</hd> <p></p> <hd id="AN0184623586-17">Attrition</hd> <p>Figure 3 shows attrition statistics at the baseline and follow‐up sessions (for full details of the characteristics, see Tables S1–S4). Unsurprisingly, completion of baseline was associated with COVID‐19 infection status (<emph>χ</emph><sups>2</sups>[<reflink idref="bib2" id="ref72">2</reflink>] = 14.93, <emph>p</emph> &lt; 0.001, <emph>V</emph> = 0.18), with the No‐COVID group having a higher tendency to drop out before completing the whole baseline session. It was, however, not associated with any other demographic variable (severity of ongoing illness, sex, age, education level, country of residence or ethnicity). Continued participation into the first follow‐up (FU1) was associated with COVID‐19 infection status (<emph>χ</emph><sups>2</sups>[<reflink idref="bib1" id="ref73">1</reflink>] = 15.48, <emph>p</emph> &lt; 0.001, <emph>V</emph> = 0.21), severity of ongoing illness at baseline (<emph>χ</emph><sups>2</sups>[<reflink idref="bib2" id="ref74">2</reflink>] = 12.66, <emph>p</emph> = 0.002, <emph>V</emph> = 0.27), sex (<emph>χ</emph><sups>2</sups>[<reflink idref="bib2" id="ref75">2</reflink>] = 9.36, <emph>p</emph> = 0.009, <emph>V</emph> = 0.17), age (<emph>χ</emph><sups>2</sups>[<reflink idref="bib5" id="ref76">5</reflink>] = 32.06, <emph>p &lt;</emph> 0.001, <emph>V</emph> = 0.31) and education level (<emph>χ</emph><sups>2</sups>[<reflink idref="bib5" id="ref77">5</reflink>] = 17.38, <emph>p</emph> = 0.004, <emph>V</emph> = 0.23). The No‐COVID group and the <emph>R</emph> subgroup were more likely to drop out before FU1. The rate of dropout was also higher in men, those aged 18–30, and those below college‐level education. Similarly, participation in the second follow‐up (FU2) was associated with COVID‐19 infection status (<emph>χ</emph><sups>2</sups>[<reflink idref="bib1" id="ref78">1</reflink>] = 14.51, <emph>p</emph> &lt; 0.001, <emph>V</emph> = 0.31), ongoing illness severity (<emph>χ</emph><sups>2</sups>[<reflink idref="bib2" id="ref79">2</reflink>] = 9.61, <emph>p</emph> = 0.008, <emph>V</emph> = 0.32) and education level (<emph>χ</emph><sups>2</sups>[<reflink idref="bib5" id="ref80">5</reflink>] = 25.3, <emph>p</emph> &lt; 0.001, <emph>V</emph> = 0.41), with the No‐COVID group, the <emph>R</emph> subgroup, and those whose education was below bachelor's degree level showing a higher tendency to drop out between FU1 and FU2. Only the COVID group was invited to the third follow‐up (FU3). Completion of FU3 was associated with education level (<emph>χ</emph><sups>2</sups>[<reflink idref="bib5" id="ref81">5</reflink>] = 13.35, <emph>p</emph> = 0.02, <emph>V</emph> = 0.43), with more dropouts observed in those at GCSE level or below but also in those with a master's or doctorate degree. Figure 4 shows the characteristics that significantly differed between the samples at the baseline and follow‐up sessions (see Table S5 for full details).</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01mar25/acp70040-fig-0003.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70040-fig-0003.jpg" title="3 Attrition statistics at the baseline and follow‐up sessions." /> </p> <p></p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01mar25/acp70040-fig-0004.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70040-fig-0004.jpg" title="4 Characteristics significantly differed between the samples at the baseline and follow‐up sessions: COVID‐19 infection status (FU1: χ2(2) = 32.84, p &lt; 0.001, V = 0.23; FU2: χ2(2) = 38.56, p &lt; 0.001, V = 0.27; FU3: χ2(2) = 66.72, p &lt; 0.001, V = 0.36); ongoing illness severity (FU2: χ2(2) = 10.18, p = 0.006, V = 0.2; FU3: χ2(2) = 12.37, p = 0.002, V = 0.23); age (FU1: χ2(5) = 23.5, p &lt; 0.001, V = 0.2; FU2: χ2(5) = 27.71, p &lt; 0.001, V = 0.23; FU3: χ2(5) = 31.91, p &lt; 0.001, V = 0.25); education level (FU2: χ2(5) = 23.8, p &lt; 0.001, V = 0.21; FU3: χ2(5) = 22.6, p &lt; 0.001, V = 0.21)." /> </p> <p></p> <hd id="AN0184623586-20">The COVID and the No‐COVID Groups</hd> <p>We compared the characteristics of the COVID group and the No‐COVID group at the baseline (to confirm findings reported in COVCOG1 in the slightly larger sample) and follow‐up sessions. Of the No‐COVID group from baseline, four participants had developed COVID‐19, and two reported being unsure of their infection status at the subsequent follow‐up. Due to the small number, these participants (<emph>n</emph> = 6) were not included in the analyses of follow‐up data. ANOVAs showed that, as expected due to the design change, the two groups differed in the intervals (in weeks) between the baseline and FU1 (COVID group: <emph>M</emph> = 11.72, SD = 9.63; No‐COVID group: <emph>M</emph> = 20.69, SD = 7.2) (<emph>F</emph>[<reflink idref="bib1" id="ref82">1</reflink>,<reflink idref="bib147" id="ref83">147</reflink>] = 34.73, <emph>p</emph> &lt; 0.001, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.19) but not between FU1 and FU2 (COVID group: <emph>M</emph> = 10.88, SD = 7.49; No‐COVID group: <emph>M</emph> = 13.29, SD = 2.74) (Assessment interval between FU2 and FU3 for the COVID group: <emph>M</emph> = 14.56 weeks, SD = 5.61).</p> <p>Patterns of differences in demographics between the two groups at baseline were consistent with those previously (Guo et al. [<reflink idref="bib21" id="ref84">21</reflink>]; see Figure 5 and Table S6). At FU1, the differences in age, education, profession, and employment remained significant; while at FU2, the two groups did not differ in demographics (see Tables S7 and S8). The COVID and the No‐COVID groups did not differ in medical history at the baseline or any follow‐up session (Sidak <emph>α</emph> = 0.003). However, ANOVAs showed that, at baseline, the COVID group had a higher BMI (<emph>M</emph> = 27.08, SD = 7.11) than the No‐COVID group (<emph>M</emph> = 25.01, SD = 5.7), even when controlling for sex, age, and education level (<emph>F</emph>[<reflink idref="bib1" id="ref85">1</reflink>,<reflink idref="bib385" id="ref86">385</reflink>] = 4.07, <emph>p</emph> = 0.044, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.01). Such tendencies persisted at FU1 but were not significant after controlling for demographics. At FU2, there were no differences in BMI between the two groups.</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01mar25/acp70040-fig-0005.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70040-fig-0005.jpg" title="5 Demographic characteristics significantly differed between the COVID and the No‐COVID groups at baseline. Differences were found in age (top panel), education level (middle panel), ethnicity (bottom left), profession (bottom center), and employment type (bottom right). The COVID group was older, higher educated, more likely to be of Northern European ethnicity (than of Asian descent), and in full‐time employment." /> </p> <p></p> <p>Within the COVID group, participants ranged between 3 and 31+ weeks since infection (positive test/symptom onset) at the time of baseline. Based on their report at baseline and the interval between assessments, we calculated weeks since infection at each follow‐up session. At FU1, participants ranged between 10 and 69+ weeks since infection. At FU2, they ranged between 14 and 74+ weeks. At FU3, the range was between 35 and 78+ weeks since infection. A total of 90.9% (<emph>n</emph> = 170/187) of baseline participants had at least one symptom lasting beyond the first 3 weeks of infection. As before (Guo et al. [<reflink idref="bib21" id="ref87">21</reflink>]), these individuals predominantly self‐identified as experiencing or having experienced Long COVID at baseline (<emph>n</emph> = 125/170, 73.5%), as well as at FU1 (<emph>n</emph> = 74/83, 89.2%), FU2 (<emph>n</emph> = 59/65, 90.8%), and FU3 (<emph>n</emph> = 47/49, 95.9%).</p> <p>The impacts of Long COVID on individuals' lives were related to ongoing illness severity. In particular, the majority of those who had ongoing severe illness (the <emph>C++</emph> subgroup) reported an inability to work for a long period due to illness (<emph>n</emph> = 60/78, 76.9%) and having difficulty coping with day‐to‐day activities (<emph>n</emph> = 56/78, 71.8%) at baseline. At all follow‐ups, over half of the <emph>C++</emph> subgroup were still unable to work (FU1: <emph>n</emph> = 28/49, 57.1%; FU2: <emph>n</emph> = 24/41, 58.5%; FU3: <emph>n</emph> = 17/32, 53.1%) or cope with daily activities (FU1: <emph>n</emph> = 31/49, 63.3%; FU2: <emph>n</emph> = 27/41, 65.9%; FU3: <emph>n</emph> = 21/32, 65.6%). Other impacts reported by the <emph>C++</emph> subgroup at baseline, which continued over the follow‐ups, included having difficulty getting medical professionals to take their symptoms seriously (38.8%–56.4%), feelings of having experienced a trauma (43.9%–53.1%) and facing financial difficulty as a result of illness (21.9%–24.5%). Those who had ongoing mild/moderate illness (the <emph>C+</emph> subgroup) also continued to experience similar difficulties: A material proportion found it hard to cope with day‐to‐day activities at baseline and across follow‐ups (Baseline: <emph>n</emph> = 32/61, 52.5%; FU1: <emph>n</emph> = 13/33, 39.4%; FU2: <emph>n</emph> = 10/23, 43.5%; FU3: <emph>n</emph> = 7/17, 41.2%). Table S9 shows full details of the impacts of Long COVID on the subgroups.</p> <hd id="AN0184623586-22">Hypothesis 1: Changes in COVID‐19 Symptoms Over Time</hd> <p>Following "COVCOG 1" (Guo et al. [<reflink idref="bib21" id="ref88">21</reflink>]), we derived symptom factors from individual symptoms using EFA with the updated baseline dataset (see Tables S10–S12 for full procedure and factor loadings). The five initial symptom factors remained "F1: Neurological/Psychiatric," "F2: Fatigue/Mixed," "F3: Gastrointestinal," "F4: Respiratory/Infectious," and "F5: Dermatological." In addition to the "Cognitive" symptom factor (derived separately from brain fog, forgetfulness, tip‐of‐the‐tongue problems, semantic disfluency and difficulty concentrating), the six ongoing/current noncognitive symptom factors were "F1: Neurological," "F2: Gastrointestinal/Autoimmune/Fatigue," "F3: Cardiopulmonary," "F4: Dermatological/Fever," "F5: Mood," and "F6: Appetite Loss." Higher scores indicated higher severity in the individual (for ongoing "since infection/last session" symptoms assessed in ratings) or higher occurrence in the population (for currently experienced symptoms assessed dichotomously). Results on group differences (COVID vs. No‐COVID groups; <emph>C++</emph> vs. <emph>C+</emph> vs. <emph>R</emph> subgroups) in symptoms (factors and individual; cognitive, and noncognitive) are provided for all stages in Tables S13–S26.</p> <hd id="AN0184623586-23">COVID Versus No‐COVID (FU1–FU2)</hd> <p>Mixed ANOVAs were conducted to examine changes in the ongoing and current Cognitive symptom factors and the 6 noncognitive symptom factor scores over time across FU1 and FU2 (within‐subject) between the COVID and the No‐COVID groups (between‐subject; ongoing factors from FU1 through FU2: shared <emph>n</emph> = 85; current factors: shared <emph>n</emph> = 64) (Time X Infection interaction). For ongoing symptom factors, there was no main effect of time (across FU1 and FU2), but there were main effects of infection group (COVID &gt; No‐COVID) in the ongoing Cognitive symptom factor (<emph>F</emph>[<reflink idref="bib1" id="ref89">1</reflink>,<reflink idref="bib79" id="ref90">79</reflink>] = 40.76, <emph>p</emph> &lt; 0.001, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.34) and all noncognitive symptom factors (<emph>p</emph>s ≤ 0.001–0.005, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.09–0.45), except Appetite Loss. Significant Time X Infection interaction effects were found for ongoing Gastrointestinal/Autoimmune/Fatigue (<emph>F</emph>[<reflink idref="bib1" id="ref91">1</reflink>,<reflink idref="bib79" id="ref92">79</reflink>] = 5.45, <emph>p</emph> = 0.022, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.07), Cardiopulmonary (<emph>F</emph>[<reflink idref="bib1" id="ref93">1</reflink>,<reflink idref="bib79" id="ref94">79</reflink>] = 6.62, <emph>p</emph> = 0.012, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.08) and Mood (<emph>F</emph>[<reflink idref="bib1" id="ref95">1</reflink>,<reflink idref="bib79" id="ref96">79</reflink>] = 7.72, <emph>p</emph> = 0.007, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.09) symptom factors, but not for the Cognitive symptom factor or Neurological, Dermatological/Fever, or Appetite Loss symptom factors (see Figure 6 and Table S27). At FU1, the COVID group had higher severity scores on all three symptom factors than the No‐COVID group (<emph>p</emph>s &lt; 0.001, <emph>d</emph>s = 0.95–2.25). The group differences in the ongoing Gastrointestinal/Autoimmune/Fatigue and Cardiopulmonary (<emph>p</emph>s &lt; 0.001, <emph>d</emph>s = 1.51–1.69), but not in Mood, symptom factors persisted at FU2. The COVID group showed improvements in ongoing Gastrointestinal/Autoimmune/Fatigue (<emph>p</emph> = 0.018, <emph>d</emph> = 0.22) and Mood (<emph>p</emph> = 0.013, <emph>d</emph> = 0.23) symptom factors from FU1 to FU2, while no significant changes were found for the No‐COVID group. Results of individual symptoms are in Table S29.</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01mar25/acp70040-fig-0006.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70040-fig-0006.jpg" title="6 Time X infection interaction effects in ongoing symptom factors from FU1 to FU2 in the COVID and the No‐COVID groups. Significant results were found for gastrointestinal/autoimmune/fatigue, cardiopulmonary, and mood, but not cognitive or neurological symptom factors. The Y‐axes show the mean scores." /> </p> <p></p> <p>There was no Time X Infection interaction for any current symptom factor. However, an overall main effect of time was observed for the current Gastrointestinal/Autoimmune/Fatigue symptom factor (<emph>F</emph>[<reflink idref="bib1" id="ref97">1</reflink>,<reflink idref="bib58" id="ref98">58</reflink>] = 4.23, <emph>p</emph> = 0.044, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.07). Main effects of infection group (COVID &gt; No‐COVID) were also found for the current Cognitive symptom factor (<emph>F</emph>[<reflink idref="bib1" id="ref99">1</reflink>,<reflink idref="bib58" id="ref100">58</reflink>] = 9.55, <emph>p</emph> = 0.003, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.14), and Neurological, Gastrointestinal/Autoimmune/Fatigue and Cardiopulmonary symptom factors (<emph>p</emph>s = 0.001–0.027, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.08–0.16) (see Table S28).</p> <hd id="AN0184623586-25">COVID Ongoing Subgroups (Baseline–FU1–FU2–FU3)</hd> <p>When comparing the <emph>C+</emph> and the <emph>C++</emph> subgroups from baseline through FU3 (given a lack of data from the <emph>R</emph> subgroup at FUs) (see Figure 7 and Tables S30–S31), mixed ANOVAs showed that there was no main effect of time across all sessions for ongoing symptom factors. However, from baseline to FU1 (shared <emph>n</emph> = 82), significant Time X Subgroup interactions were found for ongoing Cardiopulmonary (<emph>F</emph>[<reflink idref="bib1" id="ref101">1</reflink>,<reflink idref="bib76" id="ref102">76</reflink>] = 5.95, <emph>p</emph> = 0.017, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.07) and Dermatological/Fever (<emph>F</emph>[<reflink idref="bib1" id="ref103">1</reflink>,<reflink idref="bib76" id="ref104">76</reflink>] = 6.33, <emph>p</emph> = 0.014, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.08) symptom factors, but not for the Cognitive symptom factor or Neurological, Gastrointestinal/Autoimmune/Fatigue, Mood or Appetite Loss symptom factors. Improvements in the Cardiopulmonary symptom factor were observed in both the <emph>C++</emph> (<emph>p</emph> &lt; 0.001, <emph>d</emph> = 0.82) and the <emph>C+</emph> subgroups (<emph>p</emph> = 0.019, <emph>d</emph> = 0.34) from baseline to FU1. The <emph>C++</emph> subgroup, but not the <emph>C+</emph> subgroup, also showed improvement in the Dermatological/Fever (<emph>p</emph> &lt; 0.001, <emph>d</emph> = 0.42) symptom factor.</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01mar25/acp70040-fig-0007.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70040-fig-0007.jpg" title="7 Time X subgroups interaction effects in ongoing noncognitive and cognitive symptom factors over time in the C+ and C++ subgroups." /> </p> <p></p> <p>From baseline through FU2 (shared <emph>n</emph> = 64), the interaction effects for ongoing Cardiopulmonary (<emph>F</emph>[<reflink idref="bib2" id="ref105">2</reflink>,<reflink idref="bib114" id="ref106">114</reflink>] = 3.53, <emph>p</emph> = 0.033, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.06) and Dermatological/Fever (<emph>F</emph>[<reflink idref="bib2" id="ref107">2</reflink>,<reflink idref="bib114" id="ref108">114</reflink>] = 3.79, <emph>p</emph> = 0.026, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.06) symptom factors remained significant. The <emph>C++</emph> subgroup showed improvements in Cardiopulmonary and Dermatological/Fever symptom factors from baseline to FU1 and FU2 (<emph>p</emph>s &lt; 0.001, <emph>d</emph>s = 0.77–0.8), while the <emph>C+</emph> subgroup improved across the time points (from baseline to FU2: <emph>p</emph> &lt; 0.001, <emph>d</emph> = 0.7); from FU1 to FU2: <emph>p</emph> = 0.014, <emph>d</emph> = 0.4 in Cardiopulmonary, but not Dermatological/Fever, symptom factor. From baseline through FU3 (shared <emph>n</emph> = 48), there was a significant interaction for ongoing Cardiopulmonary (<emph>F</emph>[<reflink idref="bib3" id="ref109">3</reflink>,<reflink idref="bib120" id="ref110">120</reflink>] = 2.72, <emph>p</emph> = 0.047, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.06) symptom factor only. Improvements were observed in the <emph>C++</emph> subgroup, but not the <emph>C+</emph> subgroup, from baseline to FU1, FU2, and FU3 (<emph>p</emph>s &lt; 0.001, <emph>d</emph>s = 0.79–87). There was no interaction or main effect of time for any <emph>current</emph> symptom factor. Results of individual symptoms are in Table S32.</p> <hd id="AN0184623586-27">Hypothesis 2: Changes in Cognitive Performance Over Time</hd> <p>Previously (Guo et al. [<reflink idref="bib22" id="ref111">22</reflink>]), we grouped data of individual cognitive task variables into cognitive task factors, reflecting performance in memory, language and executive function. From the baseline data of this updated dataset, we generated four slightly modified factors. The factor structures of follow‐up data followed then that of baseline data. The factors were: "F1: Memory RT (reaction time)," which included overall RTs, RTs for correct responses and RTs for incorrect responses of the Word List Recognition Memory Test and the Pictorial Associative Memory Test; "F2: Memory Performance," which included correct percentage and d′ of the Word List Recognition Memory Test and correct percentage of the Pictorial Associative Memory Test; "F3: Category Fluency," which included correct words, related words, incorrect words and correct percentage of the Category Fluency Test; "F4: Executive Function RT," which included RT for correct responses and for nonperseverative errors of the WCST and "F5: Executive Function Performance," which included correct responses, number of perseverative and nonperseverative errors of the WCST. Extreme outliers were removed. For full details of factor loadings and procedure, see Table S33.</p> <hd id="AN0184623586-28">COVID Versus No‐COVID (Baseline–FU1–FU2)</hd> <p>Cognitive performance was first compared between the COVID group and the No‐COVID group with the largest possible sample size (including participants without complete longitudinal data) separately at baseline, FU1, and FU2 using one‐way ANOVAs (see Figure 8 and Tables S34 and S35 for detailed results). Then, we performed mixed ANOVAs to examine within‐subject changes in cognitive performance across baseline, FU1, and FU2 between the two groups (Time X Infection interaction). In terms of the 5 cognitive task factors, there was no significant interaction from baseline to FU1 (shared <emph>n</emph>s = 145–147 depending on the task factor). From baseline through FU2 (shared <emph>n</emph>s = 88–90), a significant interaction was found for the Memory RT factor (<emph>F</emph>[<reflink idref="bib2" id="ref112">2</reflink>,<reflink idref="bib166" id="ref113">166</reflink>] = 4.43, <emph>p</emph> = 0.013, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.05). Pairwise comparisons, although suggesting no difference in RTs between the two groups in this longitudinal sample at any time point, showed that the RT of the No‐COVID group at FU2 was shorter than their RTs at baseline (<emph>p</emph> = 0.005, <emph>d</emph> = 0.81) and FU1 (<emph>p</emph> = 0.002, <emph>d</emph> = 0.77), while RTs of the COVID group remained unchanged across time (Figure 9; see also Tables S36 and S37 for full details, including the overall main effects of time found for the Executive Function RT and Performance factors, and results of individual variables).</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01mar25/acp70040-fig-0008.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70040-fig-0008.jpg" title="8 The COVID and the No‐COVID groups' scores on the Memory Performance factor and the Memory RT factor. Results on baseline data were largely consistent with &quot;COVCOG 2&quot; (Guo et al. [22]). The COVID group had significantly lower scores in the Memory Performance factor than the No‐COVID group at baseline (F[1,347] = 6.01, p = 0.015, ήp2 = 0.02) and at FU2 (F[1,85] = 5.14, p = 0.026, ήp2 = 0.06). The difference at FU1 did not reach significance (Figure 8a). The COVID group also had longer RTs in the Memory RT factor than the No‐COVID group at FU2 (F[1,85] = 4.03, p = 0.048, ήp2 = 0.05) (Figure 8b). No significant group differences were found for other factors." /> </p> <p></p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01mar25/acp70040-fig-0009.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70040-fig-0009.jpg" title="9 Time X infection interaction effects in cognitive task factors from baseline to FU2 in the COVID and the No‐COVID groups." /> </p> <p></p> <hd id="AN0184623586-31">Recovered (R) Versus Mild/Moderate (C+) Versus Severe (C++) Versus No‐COVID</hd> <p>Figure 10 and Tables S38 and S39 show the one‐way ANOVA results of the comparison analyses on the cognitive performance between the No‐COVID group and the COVID subgroups (<emph>R</emph>, <emph>C+</emph>, <emph>C++</emph>). Mixed ANOVAs were performed to examine within‐subject changes in cognitive performance over time between the four groups (Time X Group interaction). In terms of the 5 cognitive task factors, from baseline to FU1 (shared <emph>n</emph>s = 145–147 depending on the task factor), there was a significant interaction for the Executive Function RT factor (<emph>F</emph>[<reflink idref="bib3" id="ref114">3</reflink>,<reflink idref="bib139" id="ref115">139</reflink>] = 3, <emph>p</emph> = 0.033, <emph>ή</emph><subs><emph>p</emph></subs><sups>2</sups> = 0.06). Pairwise comparisons showed that the RT of the <emph>R</emph> subgroup at FU1 was shorter than at baseline (<emph>p</emph> = 0.024, <emph>d</emph> = 0.57). However, there was no main effect of group on RTs at any time point. From baseline through FU2 (shared <emph>n</emph>s = 88–90), there was no significant interaction for any factor. Comparing the COVID subgroups from baseline through FU3 (shared <emph>n</emph>s = 54–56), no significant interaction was observed (see Tables S40 and S41 for full details, including the overall main effects of time found for the Executive Function RT and Performance factors, and results of individual variables).</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01mar25/acp70040-fig-0010.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70040-fig-0010.jpg" title="10 Scores of the Memory Performance factor in the No‐COVID group and the R, C+, and C++ subgroups. One‐way ANOVAs suggested that the results of baseline data are largely consistent with &quot;COVCOG 2&quot; as expected (Guo et al. [22]). Comparison analyses for FU3 were performed only between the COVID subgroups as the No‐COVID group did not have data for FU3. Significant group differences were found in the Memory Performance factor at baseline (F[3,345] = 3.94, p = 0.009, ήp2 = 0.03), FU1 (F[3,139] = 3.71, p = 0.013, ήp2 = 0.07), and FU2 (F[3,83] = 3.05, p = 0.033, ήp2 = 0.1). At baseline (p = 0.004, d = 0.49) and FU2 (p = 0.035, d = 0.8), the C++ subgroup had lower scores than the No‐COVID group, while at FU1 (p = 0.033, d = 0.87), the C++ subgroup also had lower scores than the R subgroup. At FU3, there was no significant group difference between the COVID subgroups in any factor (see Figure 10 and Table S38)." /> </p> <p></p> <hd id="AN0184623586-33">Hypothesis 3: Predictors of Cognitive Symptoms and Cognitive Performance</hd> <p></p> <hd id="AN0184623586-34">Predicting Cognitive Symptoms From Noncognitive Symptoms</hd> <p></p> <hd id="AN0184623586-35">Within Each Assessment Time Point</hd> <p>Multiple linear regressions were applied to test whether noncognitive symptom factors would predict the Cognitive symptom factor; this was done at different phases of illness in those infected with COVID‐19 (see Table S42 for full details). At baseline, Fatigue/Mixed symptoms experienced during the initial illness (the first 3 weeks of infection) (<emph>β</emph> = 0.48, <emph>p</emph> &lt; 0.001) predicted Cognitive symptoms experienced since the first 3 weeks (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.217, <emph>F</emph>[<reflink idref="bib4" id="ref116">4</reflink>,<reflink idref="bib165" id="ref117">165</reflink>] = 21.67, <emph>p</emph> &lt; 0.001), while Neurological/Psychiatric symptoms (<emph>β</emph> = 0.29, <emph>p</emph> &lt; 0.001) predicted currently experienced Cognitive symptoms (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.079, <emph>F</emph>[<reflink idref="bib4" id="ref118">4</reflink>,<reflink idref="bib137" id="ref119">137</reflink>] = 9.84, <emph>p</emph> &lt; 0.001). Neurological (<emph>β</emph> = 0.27, <emph>p</emph> = 0.018), Gastrointestinal/Autoimmune/Fatigue (<emph>β</emph> = 0.26, <emph>p</emph> = 0.048), and Cardiopulmonary (<emph>β</emph> = 0.23, <emph>p</emph> = 0.018) symptoms experienced between the 3 weeks and baseline jointly predicted cognitive symptoms during the same period (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.44, <emph>F</emph>[<reflink idref="bib6" id="ref120">6</reflink>,<reflink idref="bib163" id="ref121">163</reflink>] = 35.6, <emph>p</emph> &lt; 0.001). Neurological symptoms (<emph>β</emph> = 0.34, <emph>p</emph> &lt; 0.001) during that period also predicted currently experienced Cognitive symptoms (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.103, <emph>F</emph>[<reflink idref="bib4" id="ref122">4</reflink>,<reflink idref="bib137" id="ref123">137</reflink>] = 11.24, <emph>p</emph> &lt; 0.001). Current Gastrointestinal/Autoimmune/Fatigue symptoms at baseline (<emph>β</emph> = 0.43, <emph>p</emph> &lt; 0.001) predicted concurrent Cognitive symptoms (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.181, <emph>F</emph>[<reflink idref="bib4" id="ref124">4</reflink>,<reflink idref="bib137" id="ref125">137</reflink>] = 16.28, <emph>p</emph> &lt; 0.001). Similar patterns were found for the follow‐ups (see Table S42).</p> <hd id="AN0184623586-36">Prediction Across Time</hd> <p>Symptoms experienced during the initial 3 weeks of illness (reported at baseline) were able to predict cognitive symptoms across subsequent follow‐ups (see full details in Table S42). Neurological/Psychiatric symptoms during the initial illness predicted ongoing cognitive symptoms at FU1 and FU3 (<emph>β</emph> = 0.29–0.52, <emph>p</emph>s &lt; 0.001), and currently experienced cognitive symptoms at FU1 and FU2 (<emph>β</emph> = 0.32–0.4, <emph>p</emph>s = 0.003). Initial Fatigue/Mixed symptoms predicted ongoing cognitive symptoms at FU2 (<emph>β</emph> = 0.52, <emph>p</emph> &lt; 0.001).</p> <p>Early ongoing symptoms (those experienced between the initial illness and baseline) were also able to predict later cognitive symptoms. Neurological symptoms during this period predicted current Cognitive symptoms at FU1 (β = 0.3, <emph>p</emph> = 0.009) and ongoing Cognitive symptoms experienced between FU1 and FU2 (β = 0.28, <emph>p</emph> = 0.029). Baseline ongoing Cardiopulmonary symptoms predicted ongoing Cognitive symptoms across all three follow‐ups (β = 0.39–0.72, <emph>p</emph>s &lt; 0.001). Baseline ongoing Mood‐related symptoms predicted Cognitive symptoms experienced between baseline and FU1 (β = 0.22, <emph>p</emph>s = 0.024).</p> <p>Finally, symptoms experienced at the time of the baseline (baseline current symptoms) also predicted later cognitive symptoms. Current neurological symptoms experienced at baseline predicted currently experienced cognitive symptoms at FU1 and ongoing Cognitive symptoms between FU1 and FU2 (<emph>β</emph> = 0.38–0.44, <emph>p</emph>s ≤ 0.001–0.004). Baseline current Gastrointestinal/Autoimmune/Fatigue symptoms negatively predicted current cognitive symptoms at FU1 (<emph>β</emph> = −0.36, <emph>p</emph> = 0.036). Baseline current cardiopulmonary symptoms predicted ongoing cognitive symptoms at all three follow‐ups and currently experienced cognitive symptoms at FU1 and FU3 (<emph>β</emph> = 0.24–0.54, <emph>p</emph>s ≤ 0.001–0.047).</p> <hd id="AN0184623586-37">Predicting Cognitive Performance From Noncognitive and Cognitive Symptoms</hd> <p></p> <hd id="AN0184623586-38">Noncognitive Symptoms as Predictors of Cognitive Performance</hd> <p>We tested the effects of noncognitive symptoms on concurrent cognitive performance (five task factors: Memory RT, Memory Performance, Category Fluency, Executive Function RT, Executive Function Performance) at different phases of illness (see Figure 11 and Table S43 for full details). Because the natural scale directions of "better" symptoms and performance are inverse, all correlations and regression coefficients are expected to be negative and were so found. As reported previously (Guo et al. [<reflink idref="bib22" id="ref126">22</reflink>]), baseline Fatigue/Mixed symptoms experienced during the initial illness (<emph>β</emph> = −0.26, <emph>p</emph> = 0.001) predicted the baseline Memory Performance factor (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.054, <emph>F</emph>[<reflink idref="bib4" id="ref127">4</reflink>,<reflink idref="bib170" id="ref128">170</reflink>] = 3.69, <emph>p</emph> = 0.007). Additionally, ongoing Neurological symptoms between FU2 and FU3 (<emph>β</emph> = −0.45, <emph>p</emph> = 0.001) predicted the FU3 Memory Performance factor (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.187, <emph>F</emph>[<reflink idref="bib4" id="ref129">4</reflink>,<reflink idref="bib44" id="ref130">44</reflink>] = 4.37, <emph>p</emph> = 0.005). Across time, early ongoing Neurological symptoms experienced between the initial illness and baseline (<emph>β</emph> = −0.38, <emph>p</emph> = 0.005) predicted the Memory Performance factor at FU3 (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.116, <emph>F</emph>[<reflink idref="bib4" id="ref131">4</reflink>,<reflink idref="bib51" id="ref132">51</reflink>] = 4.45, <emph>p</emph> = 0.004). No significant association between noncognitive symptoms and other cognitive task factors was found.</p> <p> <img src="https://imageserver.ebscohost.com/img/embimages/rdk/BU8/01mar25/acp70040-fig-0011.jpg?ephost1=dGJyMNXb4kSepq84yOvqOLCmsE6epq5Srqa4SK6WxWXS" alt="acp70040-fig-0011.jpg" title="11 Plot of significant linear regressions with the forward stepwise method controlling for demographics predicting cognitive performance factors from noncognitive symptom factors." /> </p> <p></p> <p>Further analyses were carried out on the individual task variables of the Memory Performance factor, which included correct percentage and d′ from the Word List Recognition Memory Test, and correct percentage from the Pictorial Associative Memory Test (Sidak <emph>α</emph> = 0.017) (see full details in Table S44). Consistent with our previous results (Guo et al. [<reflink idref="bib22" id="ref133">22</reflink>]), at baseline, Fatigue/Mixed symptoms experienced during the initial illness (<emph>β</emph> = −0.26 to −0.27, <emph>p</emph>s &lt; 0.001) predicted the percentage of items correct and d′ of the Word List Recognition Memory Test. At FU1 and FU2, current Cardiopulmonary symptoms (<emph>β</emph> = −0.27 to −0.32, <emph>p</emph>s = 0.01–0.017) predicted the correct percentage of the Pictorial Associative Memory Test at each time point. At FU3, ongoing Neurological symptoms between FU2 and FU3 (<emph>β</emph> = −0.45, <emph>p</emph> = 0.001) predicted d′ of the Word List Recognition Memory Test. Across time, Neurological symptoms at baseline were a significant predictor of performance on the Word List Recognition Memory Test at FU3. The initial, ongoing, and current Neurological factors at baseline (<emph>β</emph> = −0.35 to −0.37, <emph>p</emph>s = 0.007–0.009) all predicted the correct percentage of the Word List Test at FU3. The ongoing and current Neurological factors at baseline (<emph>β</emph> = −0.33 to −0.39, <emph>p</emph>s = 0.004–0.013) also predicted <emph>d</emph>′ of the Word List Test at FU3.</p> <hd id="AN0184623586-40">Cognitive Symptoms as Predictors of Cognitive Performance</hd> <p>Finally, we tested the effects of cognitive symptoms on cognitive performance. There was no significant association between the ongoing/current Cognitive symptom factors and the cognitive task factors at any time point (see Table S45). Given that some specific cognitive symptoms may be related more directly to specific areas of cognitive performance, for exploratory purposes, we further examined the role of five individual ongoing/current cognitive symptoms (i.e., brain fog, forgetfulness, tip‐of‐the‐tongue problems, semantic disfluency and difficulty concentrating) as predictors (see full details in Table S46). In line with our previous results (Guo et al. [<reflink idref="bib22" id="ref134">22</reflink>]), at baseline, current difficulty concentrating (<emph>β</emph> = −0.24, <emph>p</emph> = 0.005) predicted the Category Fluency factor (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.093, <emph>F</emph>[<reflink idref="bib4" id="ref135">4</reflink>,<reflink idref="bib129" id="ref136">129</reflink>] = 4.41, <emph>p</emph> = 0.002). At FU1, ongoing semantic disfluency (<emph>β</emph> = −0.32, <emph>p</emph> = 0.003) predicted the Category Fluency factor (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.299, <emph>F</emph>[<reflink idref="bib4" id="ref137">4</reflink>,<reflink idref="bib78" id="ref138">78</reflink>] = 9.73, <emph>p</emph> &lt; 0.001). At FU2, ongoing tip‐of‐the‐tongue problems (<emph>β</emph> = −0.6, <emph>p</emph> = 0.001) and semantic disfluency (<emph>β</emph> = 0.63, <emph>p</emph> = 0.002), together with difficulty concentrating (although this did not individually survive Sidak correction (<emph>β</emph> = −0.31, <emph>p</emph> = 0.04)), predicted the Category Fluency factor, although in different directions (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.264, <emph>F</emph>[<reflink idref="bib6" id="ref139">6</reflink>,<reflink idref="bib58" id="ref140">58</reflink>] = 4.82, <emph>p</emph> &lt; 0.001). At FU3, current brain fog (<emph>β</emph> = −0.47, <emph>p</emph> &lt; 0.001) predicted the Memory Performance factor (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.24, <emph>F</emph>[<reflink idref="bib4" id="ref141">4</reflink>,<reflink idref="bib43" id="ref142">43</reflink>] = 4.72, <emph>p</emph> = 0.003). Across time, there was no significant association between individual ongoing/current cognitive symptoms at baseline and any of the cognitive task factors at the follow‐ups.</p> <p>We conducted further analyses on the individual task variables of the Category Fluency factor (including the repeated words variable that did not load into the factor; Sidak <emph>α</emph> = 0.01) and the Memory Performance factor (Sidak <emph>α</emph> = 0.017). Results from the baseline were largely consistent with our COVCOG 2 report (Guo et al. [<reflink idref="bib22" id="ref143">22</reflink>]). In terms of the Category Fluency test variables (see Table S47), at baseline, ongoing tip‐of‐the‐tongue problems (<emph>β</emph> = −0.47, <emph>p</emph> &lt; 0.001), together with semantic disfluency, which did not survive Sidak correction (<emph>β</emph> = 0.31, <emph>p</emph> = 0.024), predicted the number of correct words (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.058, <emph>F</emph>[<reflink idref="bib5" id="ref144">5</reflink>,<reflink idref="bib152" id="ref145">152</reflink>] = 5.19, <emph>p</emph> &lt; 0.001). Current difficulty concentrating (<emph>β</emph> = −0.24, <emph>p</emph> = 0.005) also predicted the number of correct words (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.049, <emph>F</emph>[<reflink idref="bib4" id="ref146">4</reflink>,<reflink idref="bib130" id="ref147">130</reflink>] = 5.09, <emph>p</emph> &lt; 0.001). At FU1, ongoing semantic disfluency (<emph>β</emph> = −0.33, <emph>p</emph> = 0.002) predicted the number of correct words (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.081, <emph>F</emph>[<reflink idref="bib4" id="ref148">4</reflink>,<reflink idref="bib78" id="ref149">78</reflink>] = 10.15, <emph>p</emph> &lt; 0.001). At FU2, the pattern was similar at baseline, with ongoing tip‐of‐the‐tongue problems (<emph>β</emph> = −0.52, <emph>p</emph> = 0.004), together with semantic disfluency (although this did not survive Sidak correction (<emph>β</emph> = 0.37, <emph>p</emph> = 0.038)), predicted the number of correct words (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.08, <emph>F</emph>[<reflink idref="bib5" id="ref150">5</reflink>,<reflink idref="bib59" id="ref151">59</reflink>] = 6.94, <emph>p</emph> &lt; 0.001). There were no significant associations at FU3. Across time, ongoing tip‐of‐the‐tongue problems at baseline (<emph>β</emph> = −0.41, <emph>p</emph> = 0.004) predicted the number of correct words at FU3 (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.133, <emph>F</emph>[<reflink idref="bib4" id="ref152">4</reflink>,<reflink idref="bib51" id="ref153">51</reflink>] = 3.69, <emph>p</emph> = 0.01). Ongoing difficulty concentrating at baseline (<emph>β</emph> = −0.78, <emph>p</emph> &lt; 0.001), together with brain fog (which did not survive Sidak correction (<emph>β</emph> = 0.53, <emph>p</emph> = 0.014)), predicted the number of related words at FU3 (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.175, <emph>F</emph>[<reflink idref="bib5" id="ref154">5</reflink>,<reflink idref="bib50" id="ref155">50</reflink>] = 4.68, <emph>p</emph> = 0.001).</p> <p>In terms of variables of the Memory Performance factor (see Table S48), at baseline, ongoing tip‐of‐the‐tongue problems (<emph>β</emph> = −0.3, <emph>p</emph> &lt; 0.001; ∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.071, <emph>F</emph>[<reflink idref="bib4" id="ref156">4</reflink>,<reflink idref="bib156" id="ref157">156</reflink>] = 4.56, <emph>p</emph> = 0.002) and current forgetfulness (<emph>β</emph> = −0.21, <emph>p</emph> = 0.014; ∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.035, <emph>F</emph>[<reflink idref="bib4" id="ref158">4</reflink>,<reflink idref="bib133" id="ref159">133</reflink>] = 3.99, <emph>p</emph> = 0.004) predicted the correct percentage of the Word List Recognition Memory Test. At FU2, current brain fog (<emph>β</emph> = −0.32, <emph>p</emph> = 0.011) predicted the correct percentage of the Pictorial Associative Memory Test (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.089, <emph>F</emph>[<reflink idref="bib4" id="ref160">4</reflink>,<reflink idref="bib52" id="ref161">52</reflink>] = 4.81, <emph>p</emph> = 0.002). At FU3, current brain fog (<emph>β</emph> = −0.47, <emph>p</emph> &lt; 0.001) predicted <emph>d</emph>′ of the Word List Test (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.205, <emph>F</emph>[<reflink idref="bib4" id="ref162">4</reflink>,<reflink idref="bib43" id="ref163">43</reflink>] = 4.65, <emph>p</emph> = 0.003). Across time, the extent of ongoing tip‐of‐the‐tongue problems at baseline (<emph>β</emph> = −0.28, <emph>p</emph> = 0.017) predicted the correct percentage of the Word List Test at FU1 (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.053, <emph>F</emph>[<reflink idref="bib4" id="ref164">4</reflink>,<reflink idref="bib82" id="ref165">82</reflink>] = 3.48, <emph>p</emph> = 0.011). Ongoing forgetfulness at baseline (<emph>β</emph> = −0.35, <emph>p</emph> = 0.016) predicted <emph>d</emph>′ of the Word List Test at FU3 (∆<emph>R</emph><sups>2</sups><subs>adj</subs> = 0.084, <emph>F</emph>[<reflink idref="bib4" id="ref166">4</reflink>,<reflink idref="bib51" id="ref167">51</reflink>] = 3.66, <emph>p</emph> = 0.011).</p> <hd id="AN0184623586-41">Discussion</hd> <p>The present "COVCOG 3" paper aimed at understanding the trajectory of cognition in post‐acute COVID‐19 (Long COVID). A longitudinal analysis examined how symptoms (both cognitive and noncognitive) and objective cognitive function changed over time in people with past COVID‐19 infection and varying levels of Long COVID severity (the COVID group and subgroups), compared to uninfected controls (the No‐COVID group). We also investigated how diverse symptom patterns may predict cognitive problems. Results were largely consistent with our expectations given previous evidence.</p> <hd id="AN0184623586-42">Symptoms</hd> <p>With regard to changes in Long COVID symptoms, we found varying profiles for some symptoms, but cognitive and neurological symptoms proved persistent over time. Some noncognitive symptoms improved in the COVID group relative to the No‐COVID group (i.e., ongoing gastrointestinal/autoimmune/fatigue and mood symptoms appeared to reduce), but cardiopulmonary symptoms were persistently higher in the COVID group across time. The lack of improvement in cognitive and neurological symptoms was particularly notable in those who had reported moderate to severe ongoing Long COVID at baseline (<emph>C++</emph> and/or <emph>C+</emph> subgroups). This finding is in line with other studies showing that cognitive/neurological symptoms could persist for long periods after COVID‐19 infection (e.g., Ceban et al. [<reflink idref="bib9" id="ref168">9</reflink>]; Davis et al. [<reflink idref="bib11" id="ref169">11</reflink>]; Dennis et al. [<reflink idref="bib14" id="ref170">14</reflink>]), potentially due to neurological damage and abnormal neural activity in areas of the brain pivotal to cognitive functioning, such as the hippocampus and frontal lobes (Douaud et al. [<reflink idref="bib15" id="ref171">15</reflink>]; Hosp et al. [<reflink idref="bib29" id="ref172">29</reflink>]).</p> <hd id="AN0184623586-43">Cognitive Performance</hd> <p>With regard to objective cognitive function, consistent with "COVCOG 2" (Guo et al. [<reflink idref="bib22" id="ref173">22</reflink>]), memory function—both response accuracy and speed—is particularly impacted in the COVID group (most noticeably in the <emph>C++</emph> subgroup). Most of the group differences found at baseline remained at subsequent follow‐ups. Analysis of change in performance over time showed that while the No‐COVID group demonstrated practice‐related improvement (shorter RTs and some improvement in performance), such progress was not seen in the COVID group.</p> <p>Our findings suggest that people with past COVID‐19 infection overall, particularly those who suffered mild to severe ongoing illness, did not experience improvement in cognitive function over time—or at least not over the 10–15 months post infection covered by this study. Previous studies showed differing patterns of change in cognitive function in people with varying initial illness severity (Afzali et al. [<reflink idref="bib2" id="ref174">2</reflink>]; Del Brutto et al. [<reflink idref="bib13" id="ref175">13</reflink>]; Miskowiak et al. [<reflink idref="bib43" id="ref176">43</reflink>]). However, given that almost all our participants showed a relatively mild initial illness (nonhospitalized), our study suggests that the lack of cognitive improvement may be associated with the presence of <emph>ongoing</emph> illness. Those who had experienced infection but considered themselves recovered (the <emph>R</emph> subgroup) showed a trajectory roughly similar to those without infection history; however, those who reported some symptoms of Long COVID at 1–6 months post infection, regardless of severity (i.e., the <emph>C+</emph> and <emph>C++</emph> subgroups) tended to still be experiencing some level of cognitive impairment as long as 9 months post infection.</p> <hd id="AN0184623586-44">Predictors of Cognitive Problems</hd> <p>As expected, neurological symptoms consistently predicted both cognitive symptoms and cognitive performance at/across different phases of illness following COVID‐19 infection. Neurological/psychiatric symptoms during the initial 3 weeks of illness predicted both concurrent memory performance and performance at the final follow‐up, 9 months later. Indeed, neurological symptoms across the entire follow‐up period predicted ongoing/concurrent cognitive symptoms and performance (particularly of the Word List Test) at the final follow‐up. Additionally, cardiopulmonary symptoms experienced since the initial illness consistently predicted concurrent cognitive symptoms, with currently experienced cardiopulmonary symptoms at FU1 and FU2 also predicting lower accuracy in the Pictorial Associative Memory Test during the same sessions. Fatigue/mixed symptoms during the initial illness predicted baseline and follow‐up ongoing cognitive symptoms and baseline memory performance (specifically Word List Test), but gastrointestinal/autoimmune/fatigue symptoms experienced since the initial illness predicted ongoing/concurrent cognitive symptoms at baseline and follow‐ups. There were also some associations between mood‐related symptoms and follow‐up cognitive symptoms. Overall, more severe neurological, cardiopulmonary, and fatigue‐related symptoms were associated with more severe cognitive symptoms and worse memory; neurological symptoms were mostly predictive of future cognition, while cardiopulmonary symptoms were more predictive of concurrent cognition. These support other findings, for example, that headache during acute COVID predicts cognitive symptoms one year later in older adults (Cavaco et al. [<reflink idref="bib8" id="ref177">8</reflink>]).</p> <p>In the absence of neurophysiological data, these results cannot speak directly to the mechanism. However, they do suggest that cognitive problems may be driven both by neurological events during the acute and post‐acute periods, and ongoing effects from cardiovascular and breathing pathology. This may suggest that acute neurological events during the initial illness may lead to long‐term issues, particularly with memory. Were such events to result in structural damage such as ischemic lesions, it might be predicted both that these would be associated with long‐term cognitive change, and that this change might be relatively nonspecific in nature (as the location of the lesion would not be prescribed). On the other hand, ongoing cardiopulmonary issues may result in reduced oxygen supply to the brain (either via respiratory issues or through inefficient oxygen transfer), which would be expected to lead to concurrent cognitive issues. Notably, it is predominantly ongoing cardiopulmonary, rather than systemic inflammation‐related (e.g., autoimmune/fatigue/fever) symptoms that appear to best predict variance in cognitive symptoms and performance. However, given that inflammation can be a risk factor for cardiovascular problems, particularly abnormal clotting (e.g., Levi [<reflink idref="bib36" id="ref178">36</reflink>]; Obeagu and Obeagu [<reflink idref="bib48" id="ref179">48</reflink>]), this does not rule out inflammation as a mechanistic factor. Indeed, previous work has demonstrated suggestive evidence for multiple mechanistic pathways, often citing systemic‐ and neuro‐inflammation as underlying causal factors (e.g., Adingupu et al. [<reflink idref="bib1" id="ref180">1</reflink>]; Quan et al. [<reflink idref="bib54" id="ref181">54</reflink>]; Sarkar et al. [<reflink idref="bib58" id="ref182">58</reflink>]).</p> <p>Subjective cognitive symptoms were also predictive of objective cognitive function, particularly in the memory and language areas. Higher levels of ongoing tip‐of‐the‐tongue problems, ongoing/current forgetfulness, and current brain fog since the initial illness predicted lower accuracy in the verbal (Word List) memory test at baseline and/or subsequent follow‐ups. Currently experienced brain fog at the final two follow‐ups also predicted lower accuracy in the nonverbal (Pictorial) memory test and overall memory performance during the same period, respectively. Moreover, ongoing tip‐of‐the‐tongue problems, ongoing semantic disfluency, and current difficulty concentrating since the initial illness predicted fewer correct responses in the language (Category Fluency) test at baseline and/or follow‐ups. Some previous studies found weak associations between subjective reports and objective measures of cognition in COVID‐19 patients (Godoy‐González et al. [<reflink idref="bib18" id="ref183">18</reflink>]; Pihlaja et al. [<reflink idref="bib52" id="ref184">52</reflink>]). These studies suggested pandemic‐related psychological factors, such as post‐traumatic stress disorder or depression, as underlying causes of cognitive impairments, but neglected a potential issue of low statistical power. However, our results from the community sample suggest that self‐reported cognitive symptoms are broadly consistent with objective cognitive performance in this population.</p> <hd id="AN0184623586-45">Impacts Associated With "Long COVID"</hd> <p>The majority of our participants (over 70% at baseline and rising through follow‐ups) self‐identified as experiencing "Long COVID," and many reported severe and persistent impacts on quality of life. In particular, those with severe ongoing illness at baseline continued to report being unable to work or cope with daily activities, and suffering resultant financial distress through to the final follow‐ups, that is, approximately 10–15 months from infection. A further persistent challenge was the difficulty in getting medical professionals to take their symptoms seriously, and many continued to feel that they had experienced a trauma. These findings align with the UK national report (ONS [<reflink idref="bib49" id="ref185">49</reflink>]) and other previous studies on Long COVID (Davis et al. [<reflink idref="bib11" id="ref186">11</reflink>]; Ziauddeen et al. [<reflink idref="bib65" id="ref187">65</reflink>]). Notably, over 80% of participants in the Davis et al. ([<reflink idref="bib11" id="ref188">11</reflink>]) study who reported a reduced ability to work attributed this to cognitive dysfunction. These long‐lasting impacts are striking, particularly the combination of degree and severity of impact with the dismissal by medical professionals (Guo et al. [<reflink idref="bib21" id="ref189">21</reflink>]; ONS [<reflink idref="bib49" id="ref190">49</reflink>]). Cognitive sequelae of COVID‐19 thus have the potential for long‐term consequences not just for individuals but also—given the prevalence of Long COVID—for the economy and wider society, for example, workforce morbidity (e.g., Ayoubkhani et al. [<reflink idref="bib3" id="ref191">3</reflink>]; Reuschke et al. [<reflink idref="bib56" id="ref192">56</reflink>]). It is of great importance to prevent, predict, identify, and treat issues associated with Long COVID (Guo et al. [<reflink idref="bib21" id="ref193">21</reflink>]).</p> <p>Long COVID might lead to long‐term or even permanent cognitive decline, with increased vulnerability to neurodegeneration and dementia (Bougakov et al. [<reflink idref="bib7" id="ref194">7</reflink>]; Goldberg et al. [<reflink idref="bib19" id="ref195">19</reflink>]; Miners et al. [<reflink idref="bib41" id="ref196">41</reflink>]). Such effects have been suggested in some older adults (e.g., Liu et al. [<reflink idref="bib39" id="ref197">39</reflink>]; Taquet et al. [<reflink idref="bib60" id="ref198">60</reflink>]) but a definite answer for younger sufferers is not yet available. Severe and long‐lasting cognitive deficits have been reported in participants who experienced very severe COVID (i.e., including ICU care and mechanical ventilation; e.g., Hampshire et al. [<reflink idref="bib24" id="ref199">24</reflink>]). This and the evidence of some long‐lasting cognitive deficits in the relatively young, and in those who suffered mild initial illness, must leave continuing concerns. Further longitudinal research is needed to monitor COVID‐related cognitive dysfunction and its impacts over the longer term.</p> <hd id="AN0184623586-46">Limitations and Future Research</hd> <p>Limitations of the baseline study design were discussed in our previous reports (Guo et al. [<reflink idref="bib21" id="ref200">21</reflink>], [<reflink idref="bib22" id="ref201">22</reflink>]). There are also limitations specific to the longitudinal follow‐ups. The first major issue arises from the change in the follow‐up procedure for the No‐COVID group. Originally, we regularly asked the No‐COVID group to indicate whether they had been infected with COVID‐19; they would join the COVID group for the follow‐ups only if they answered "Yes," and otherwise did not complete any measures. However, we changed our design to also invite the No‐COVID group to complete the same follow‐up measures for between‐group comparisons. This change resulted from both the changing narrative and knowledge about Long COVID (including widespread denial that the condition existed (e.g., Yong [<reflink idref="bib63" id="ref202">63</reflink>])), and from fewer than expected individuals developing new COVID infections after baseline (indeed only 6 did so across the entire length of the study). By the time that the COVID group was at their final, third follow‐up, the No‐COVID group was at their second follow‐up, after which the study ended. Because of this design change, there was no data from the No‐COVID group to compare with the COVID group at FU3. The two groups also differed in the intervals (in weeks) between the baseline and the first follow‐up, with the No‐COVID group falling approximately 9 weeks apart. To offset a possible ambiguity of interpretation, the assessment intervals were precisely controlled for in analyses examining changes over time. A related issue (see also Guo et al. [<reflink idref="bib21" id="ref203">21</reflink>]) is that due to the original design, symptom information was not collected for the No‐COVID group at baseline, or (in terms of ongoing/current symptoms) from those who had COVID but reported never having felt ill or having recovered (the <emph>R</emph> subgroup). This has limited our ability to compare changes in COVID‐19 symptoms across the full time course of the study in all groups. In the current study, we were only able to look at group differences in symptom data between the COVID and the No‐COVID group at FU1 and FU2 (but not the baseline or the final, third follow‐up), and between the <emph>C++</emph> and <emph>C+</emph> subgroups (rather than the <emph>R</emph> subgroup) from baseline through the final follow‐up.</p> <p>The third issue concerns the composition of the samples; our participants with past COVID‐19 infection tended to be older and more educated than the uninfected controls. Such demographic differences were present at baseline and remained at the first follow‐up. Attrition across time was higher in men, younger participants, and those with lower education levels. Thus, the proportions of the older age groups and more educated participants were even higher in the follow‐ups than at baseline. Where possible, we controlled for sex, age, and education in our analyses to mitigate some of these biases. Importantly, the No‐COVID group and the <emph>R</emph> subgroup were more likely to drop out from follow‐ups than participants who had ongoing Long COVID symptoms. This may be because people who had not been infected with COVID‐19 or who had recovered from it perceived less relevance in the study and were thus less motivated to continue participation. This common form of bias in longitudinal research may have biased our results toward the more severe end of the post‐COVID spectrum, and this may limit generalizability.</p> <p>Finally, the specific cognitive tasks chosen for the baseline assessment were done so within the limitations of which well‐established tasks were available and ready in the limited time scale during which the baseline study was arranged. Due to the urgent nature of capturing data from early in the pandemic, there was insufficient time to create and curate an ideal battery of tasks. As such, the use of tasks such as the Wisconsin card sorting task, which has recently attracted criticism as to its informational value (e.g., Nyhus and Barceló [<reflink idref="bib46" id="ref204">46</reflink>]), may limit the ability of the study to detect individual differences in executive function. Similarly, while both verbal and nonverbal item and associative memory tasks were employed, memory and stimulus type were confounded across these, raising the question of whether differential deficits may be the result of the type of recall required or the stimulus type. As a longitudinal follow‐up, the current study inherits the shortcomings of the original study in this respect. Future research should address these shortcomings by more carefully selecting cognitive assessments.</p> <hd id="AN0184623586-47">Summary and Conclusion</hd> <p>Notwithstanding some limitations in this study, the findings are consistent in message and notable in implications. Cognitive dysfunction, both self‐reported and objectively measured, appears to be a persistent feature of Long COVID even many months post‐infection. The performance element was particularly notable in memory, which was the most impaired at baseline and throughout follow‐ups. The failure to observe notable long‐term improvement in these conditions—highlighted as particularly relevant to quality of life—is highly concerning and warrants continued monitoring and evaluation. In particular, neurological symptoms experienced early in the illness continued to be predictive of cognitive issues many months later, while cardiopulmonary symptoms were more likely to track concurrent deficits. These findings offer the possibility to develop triage tools for follow‐up, as well as to test rehabilitative and therapeutic regimens.</p> <hd id="AN0184623586-48">Author Contributions</hd> <p> <bold>Sabine P. Yeung:</bold> writing – original draft, writing – review and editing, formal analysis, project administration. <bold>Panyuan Guo:</bold> writing – original draft, writing – review and editing, formal analysis, project administration. <bold>Francess L. Adlard:</bold> writing – original draft, writing – review and editing, formal analysis. <bold>Seraphina R. Zhang:</bold> writing – original draft, writing – review and editing, formal analysis, project administration. <bold>Vidita Bhagat:</bold> writing – original draft, writing – review and editing, formal analysis, project administration. <bold>Josiah Cho:</bold> writing – original draft, writing – review and editing, formal analysis, project administration. <bold>Lyn Curtis:</bold> writing – original draft, writing – review and editing, formal analysis, project administration. <bold>Muzaffer Kaser:</bold> supervision, formal analysis, writing – review and editing. <bold>Mark P. Haggard:</bold> writing – review and editing, formal analysis, supervision. <bold>Lucy G. Cheke:</bold> conceptualization, methodology, investigation, writing – original draft, writing – review and editing, formal analysis, supervision, project administration.</p> <hd id="AN0184623586-49">Disclosure</hd> <p>No material from other sources was reproduced in this paper. The study was not a clinical trial and thus did not require clinical trial registration. It was not pre‐registered.</p> <hd id="AN0184623586-50">Ethics Statement</hd> <p>The study was reviewed by the Cambridge Psychology Research Ethics Committee at the University of Cambridge (PRE.2020.106, 8/9/2020).</p> <hd id="AN0184623586-51">Consent</hd> <p>Informed consent was obtained from participants prior to testing.</p> <hd id="AN0184623586-52">Conflicts of Interest</hd> <p>The authors declare no conflicts of interest.</p> <hd id="AN0184623586-53">Data Availability Statement</hd> <p>The data that support the findings of this study are available from the corresponding author upon reasonable request.</p> <p>GRAPH: Data S1.</p> <ref id="AN0184623586-54"> <title> Footnotes </title> <blist> <bibl id="bib1" idref="ref73" type="bt">1</bibl> <bibtext> Funding: The authors received no specific funding for this work.</bibtext> </blist> </ref> <ref id="AN0184623586-55"> <title> References </title> <blist> <bibtext> Adingupu, D. D., A. Soroush, A. Hansen, R. Twomey, and J. F. Dunn. 2023. " Brain Hypoxia, Neurocognitive Impairment, and Quality of Life in People Post‐COVID‐19." Journal of Neurology 270, no. 7 : 3303 – 3314. https://doi.org/10.1007/s00415‐023‐11767‐2.</bibtext> </blist> <blist> <bibl id="bib2" idref="ref61" type="bt">2</bibl> <bibtext> Afzali, A., B. Hatef, H. Sahraei, G. H. Meftahi, A. Khaleghi, and G. P. Jahromi. 2023. " Changes in Psychological and Cognitive Variables as Well as Cortisol Levels in Recovered Covid‐19 Patients: A Longitudinal Study." Current Psychology 43, no. 13 : 1 – 10. https://doi.org/10.1007/s12144‐022‐04211‐7.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref109" type="bt">3</bibl> <bibtext> Ayoubkhani, D., F. Zaccardi, K. B. Pouwels, et al. 2024. " Employment Outcomes of People With Long Covid Symptoms: Community‐Based Cohort Study." European Journal of Public Health 34, no. 3 : 489 – 496. https://doi.org/10.1093/eurpub/ckae034.</bibtext> </blist> <blist> <bibl id="bib4" idref="ref51" type="bt">4</bibl> <bibtext> Baker, H. A., S. A. Safavynia, and L. A. Evered. 2021. " The 'Third Wave': Impending Cognitive and Functional Decline in COVID‐19 Survivors." British Journal of Anaesthesia 126, no. 1 : 44 – 47. https://doi.org/10.1016/j.bja.2020.09.045.</bibtext> </blist> <blist> <bibl id="bib5" idref="ref12" type="bt">5</bibl> <bibtext> Ballouz, T., D. Menges, A. Anagnostopoulos, et al. 2023. " Recovery and Symptom Trajectories up to Two Years After SARS‐CoV‐2 Infection: Population Based, Longitudinal Cohort Study." BMJ 381 : e074425. https://doi.org/10.1136/bmj‐2022‐074425.</bibtext> </blist> <blist> <bibl id="bib6" idref="ref32" type="bt">6</bibl> <bibtext> Becker, J. H., J. J. Lin, M. Doernberg, et al. 2021. " Assessment of Cognitive Function in Patients After COVID‐19 Infection." JAMA Network Open 4, no. 10 : e2130645. https://doi.org/10.1001/jamanetworkopen.2021.30645.</bibtext> </blist> <blist> <bibl id="bib7" idref="ref20" type="bt">7</bibl> <bibtext> Bougakov, D., K. Podell, and E. Goldberg. 2021. " Multiple Neuroinvasive Pathways in COVID‐19." Molecular Neurobiology 58, no. 2 : 564 – 575. https://doi.org/10.1007/s12035‐020‐02152‐5.</bibtext> </blist> <blist> <bibl id="bib8" idref="ref177" type="bt">8</bibl> <bibtext> Cavaco, S., G. Sousa, A. Gonçalves, et al. 2023. " Predictors of Cognitive Dysfunction One‐Year Post COVID‐19." Neuropsychology 37, no. 5 : 557 – 567. https://doi.org/10.1037/neu0000876.</bibtext> </blist> <blist> <bibl id="bib9" idref="ref10" type="bt">9</bibl> <bibtext> Ceban, F., S. Ling, L. M. W. Lui, et al. 2022. " Fatigue and Cognitive Impairment in Post‐COVID‐19 Syndrome: A Systematic Review and Meta‐Analysis." Brain, Behavior, and Immunity 101 : 93 – 135. https://doi.org/10.1016/j.bbi.2021.12.020.</bibtext> </blist> <blist> <bibtext> Crivelli, L., K. Palmer, I. Calandri, et al. 2022. " Changes in Cognitive Functioning After COVID‐19: A Systematic Review and Meta‐Analysis." Alzheimer's &amp; Dementia 18, no. 5 : 1047 – 1066. https://doi.org/10.1002/alz.12644.</bibtext> </blist> <blist> <bibtext> Davis, H. E., G. S. Assaf, L. McCorkell, et al. 2021. " Characterizing Long COVID in an International Cohort: 7 Months of Symptoms and Their Impact." eClinicalMedicine 38 : 101019. https://doi.org/10.1016/j.eclinm.2021.101019.</bibtext> </blist> <blist> <bibtext> Davis, H. E., L. McCorkell, J. M. Vogel, and E. J. Topol. 2023. " Long COVID: Major Findings, Mechanisms and Recommendations." Nature Reviews Microbiology 21, no. 3 : 133 – 146. https://doi.org/10.1038/s41579‐022‐00846‐2.</bibtext> </blist> <blist> <bibtext> Del Brutto, O. H., D. A. Rumbea, B. Y. Recalde, and R. M. Mera. 2022. " Cognitive Sequelae of Long COVID May Not Be Permanent: A Prospective Study." European Journal of Neurology 29, no. 4 : 1218 – 1221. https://doi.org/10.1111/ene.15215.</bibtext> </blist> <blist> <bibtext> Dennis, A., D. J. Cuthbertson, D. Wootton, et al. 2023. " Multi‐Organ Impairment and Long COVID: A 1‐Year Prospective, Longitudinal Cohort Study." Journal of the Royal Society of Medicine 116, no. 3 : 97 – 112. https://doi.org/10.1177/01410768231154703.</bibtext> </blist> <blist> <bibtext> Douaud, G., S. Lee, F. Alfaro‐Almagro, et al. 2022. " SARS‐CoV‐2 Is Associated With Changes in Brain Structure in UK Biobank." Nature 604, no. 7907 : 697 – 707. https://doi.org/10.1038/s41586‐022‐04569‐5.</bibtext> </blist> <blist> <bibtext> FAIR Health. 2022. "Patients Diagnosed With Post‐COVID Conditions: An Analysis of Private Healthcare Claims Using the Official ICD‐10 Diagnostic Code." https://s3.amazonaws.com/media2.fairhealth.org/whitepaper/asset/Patients%20Diagnosed%20with%20Post‐COVID%20Conditions%20‐%20A%20FAIR%20Health%20White%20Paper.pdf.</bibtext> </blist> <blist> <bibtext> Galanopoulou, A. S., V. Ferastraoaru, D. J. Correa, et al. 2020. " EEG Findings in Acutely Ill Patients Investigated for SARS‐CoV‐2/COVID‐19: A Small Case Series Preliminary Report." Epilepsia Open 5, no. 2 : 314 – 324. https://doi.org/10.1002/epi4.12399.</bibtext> </blist> <blist> <bibtext> Godoy‐González, M., G. Navarra‐Ventura, G. Gomà, et al. 2023. " Objective and Subjective Cognition in Survivors of COVID‐19 One Year After ICU Discharge: The Role of Demographic, Clinical, and Emotional Factors." Critical Care 27 : 188. https://doi.org/10.1186/s13054‐023‐04478‐7.</bibtext> </blist> <blist> <bibtext> Goldberg, E., K. Podell, D. K. Sodickson, and E. Fieremans. 2020. " The Brain After COVID‐19: Compensatory Neurogenesis or Persistent Neuroinflammation? " eClinicalMedicine 31 : 100684. https://doi.org/10.1016/j.eclinm.2020.100684.</bibtext> </blist> <blist> <bibtext> Graham, E. L., J. R. Clark, Z. S. Orban, et al. 2021. " Persistent Neurologic Symptoms and Cognitive Dysfunction in Non‐Hospitalized Covid‐19 'Long Haulers'." Annals of Clinical and Translational Neurology 8, no. 5 : 1073 – 1085. https://doi.org/10.1002/acn3.51350.</bibtext> </blist> <blist> <bibtext> Guo, P., A. Benito Ballesteros, S. P. Yeung, et al. 2022a. " COVCOG 1: Factors Predicting Physical, Neurological and Cognitive Symptoms in Long COVID in a Community Sample. A First Publication From the COVID and Cognition Study." Frontiers in Aging Neuroscience 14 : 804922. https://doi.org/10.3389/fnagi.2022.804922.</bibtext> </blist> <blist> <bibtext> Guo, P., A. Benito Ballesteros, S. P. Yeung, et al. 2022b. " COVCOG 2: Cognitive and Memory Deficits in Long COVID: A Second Publication From the COVID and Cognition Study." Frontiers in Aging Neuroscience 14 : 804937. https://doi.org/10.3389/fnagi.2022.804937.</bibtext> </blist> <blist> <bibtext> Hampshire, A., A. Azor, C. Atchison, et al. 2024. " Cognition and Memory After Covid‐19 in a Large Community Sample." New England Journal of Medicine 390, no. 9 : 806 – 818. https://doi.org/10.1056/NEJMoa2311330.</bibtext> </blist> <blist> <bibtext> Hampshire, A., D. A. Chatfield, A. M. MPhil, et al. 2022. " Multivariate Profile and Acute‐Phase Correlates of Cognitive Deficits in a COVID‐19 Hospitalised Cohort." eClinicalMedicine 47 : 101417. https://doi.org/10.1016/j.eclinm.2022.101417.</bibtext> </blist> <blist> <bibtext> Hampshire, A., W. Trender, S. R. Chamberlain, et al. 2021. " Cognitive Deficits in People Who Have Recovered From COVID‐19." eClinicalMedicine 39 : 101044. https://doi.org/10.1016/j.eclinm.2021.101044.</bibtext> </blist> <blist> <bibtext> Hastie, C. E., D. J. Lowe, A. McAuley, et al. 2022. " Outcomes Among Confirmed Cases and a Matched Comparison Group in the Long‐COVID in Scotland Study." Nature Communications 13, no. 1 : 5663. https://doi.org/10.1038/s41467‐022‐33415‐5.</bibtext> </blist> <blist> <bibtext> Helms, J., S. Kremer, H. Merdji, et al. 2020. " Neurologic Features in Severe SARS‐CoV‐2 Infection." New England Journal of Medicine 382, no. 23 : 2268 – 2270. https://doi.org/10.1056/NEJMc2008597.</bibtext> </blist> <blist> <bibtext> Herrera, E., M. D. C. Pérez‐Sánchez, R. San Miguel‐Abella, et al. 2023. " Cognitive Impairment in Young Adults With Post COVID‐19 Syndrome." Scientific Reports 13, no. 1 : 6378. https://doi.org/10.1038/s41598‐023‐32939‐0.</bibtext> </blist> <blist> <bibtext> Hosp, J. A., A. Dressing, G. Blazhenets, et al. 2021. " Cognitive Impairment and Altered Cerebral Glucose Metabolism in the Subacute Stage of COVID‐19." Brain 144, no. 4 : 1263 – 1276. https://doi.org/10.1093/brain/awab009.</bibtext> </blist> <blist> <bibtext> Huang, L., Q. Yao, X. Gu, et al. 2021. " 1‐Year Outcomes in Hospital Survivors With COVID‐19: A Longitudinal Cohort Study." Lancet 398, no. 10302 : 747 – 758. https://doi.org/10.1016/S0140‐6736(21)01755‐4.</bibtext> </blist> <blist> <bibtext> Huang, L., X. Li, X. Gu, et al. 2022. " Health Outcomes in People 2 Years After Surviving Hospitalisation With COVID‐19: A Longitudinal Cohort Study." Lancet Respiratory Medicine 10, no. 9 : 863 – 876. https://doi.org/10.1016/S2213‐2600(22)00126‐6.</bibtext> </blist> <blist> <bibtext> Jaywant, A., F. M. Gunning, L. E. Oberlin, et al. 2024. " Cognitive Symptoms of Post‐COVID‐19 Condition and Daily Functioning." JAMA Network Open 7, no. 2 : e2356098. https://doi.org/10.1001/jamanetworkopen.2023.56098.</bibtext> </blist> <blist> <bibtext> Kandemirli, S. G., L. Dogan, Z. T. Sarikaya, et al. 2020. " Brain MRI Findings in Patients in the Intensive Care Unit With COVID‐19 Infection." Radiology 297, no. 1 : E232 – E235. https://doi.org/10.1148/radiol.2020201697.</bibtext> </blist> <blist> <bibtext> Ladds, E., J. L. Darbyshire, N. D. Bakerly, Z. Falope, and I. Tucker‐Bell. 2024. " Cognitive Dysfunction After Covid‐19." BMJ 384 : e075387. https://doi.org/10.1136/bmj‐2023‐075387.</bibtext> </blist> <blist> <bibtext> Lechien, J. R., C. M. Chiesa‐Estomba, D. R. De Siati, et al. 2020. " Olfactory and Gustatory Dysfunctions as a Clinical Presentation of Mild‐To‐Moderate Forms of the Coronavirus Disease (COVID‐19): A Multicenter European Study." European Archives of Oto‐Rhino‐Laryngology 277, no. 8 : 2251 – 2261. https://doi.org/10.1007/s00405‐020‐05965‐1.</bibtext> </blist> <blist> <bibtext> Levi, M. 2017. " Inflammation and Coagulation." In Inflammation: From Molecular and Cellular Mechanisms to the Clinic, 833 – 860. https://onlinelibrary.wiley.com/doi/10.1002/9783527692156.ch32.</bibtext> </blist> <blist> <bibtext> Levine, D. 2022. "Long COVID has Become the 'Pandemic After the Pandemic'." U.S. News &amp; World Report, March 10, 2022. https://<ulink href="http://www.usnews.com/news/health&amp;#8208;news/articles/2022&amp;#8208;03&amp;#8208;10/long&amp;#8208;covid&amp;#8208;has&amp;#8208;become&amp;#8208;the&amp;#8208;pandemic&amp;#8208;after&amp;#8208;the&amp;#8208;pandemic">www.usnews.com/news/health&amp;#8208;news/articles/2022&amp;#8208;03&amp;#8208;10/long&amp;#8208;covid&amp;#8208;has&amp;#8208;become&amp;#8208;the&amp;#8208;pandemic&amp;#8208;after&amp;#8208;the&amp;#8208;pandemic</ulink>.</bibtext> </blist> <blist> <bibtext> Li, Z., Z. Zhang, Z. Zhang, Z. Wang, and H. Li. 2023. " Cognitive Impairment After Long COVID‐19: Current Evidence and Perspectives." Frontiers in Neurology 14 : 1239182. https://doi.org/10.3389/fneur.2023.1239182.</bibtext> </blist> <blist> <bibtext> Liu, Y. H., Y. Chen, Q. H. Wang, et al. 2022. " One‐Year Trajectory of Cognitive Changes in Older Survivors of COVID‐19 in Wuhan, China: A Longitudinal Cohort Study." JAMA Neurology 79, no. 5 : 509 – 517. https://doi.org/10.1001/jamaneurol.2022.0461.</bibtext> </blist> <blist> <bibtext> Mao, L., H. Jin, M. Wang, et al. 2020. " Neurologic Manifestations of Hospitalized Patients With Coronavirus Disease 2019 in Wuhan, China." JAMA Neurology 77, no. 6 : 683 – 690. https://doi.org/10.1001/jamaneurol.2020.1127.</bibtext> </blist> <blist> <bibtext> Miners, S., P. G. Kehoe, and S. Love. 2020. " Cognitive Impact of COVID‐19: Looking Beyond the Short Term." Alzheimer's Research &amp; Therapy 12, no. 1 : 170. https://doi.org/10.1186/s13195‐020‐00744‐w.</bibtext> </blist> <blist> <bibtext> Miskowiak, K. W., J. K. Pedersen, D. V. Gunnarsson, et al. 2023. " Cognitive Impairments Among Patients in a Long‐COVID Clinic: Prevalence, Pattern and Relation to Illness Severity, Work Function and Quality of Life." Journal of Affective Disorders 324 : 162 – 169. https://doi.org/10.1016/j.jad.2022.12.122.</bibtext> </blist> <blist> <bibtext> Miskowiak, K. W., L. Fugledalen, A. E. Jespersen, et al. 2022. " Trajectory of Cognitive Impairments Over 1 Year After COVID‐19 Hospitalisation: Pattern, Severity, and Functional Implications." European Neuropsychopharmacology 59 : 82 – 92. https://doi.org/10.1016/j.euroneuro.2022.04.004.</bibtext> </blist> <blist> <bibtext> Nalbandian, A., K. Sehgal, A. Gupta, et al. 2021. " Post‐Acute COVID‐19 Syndrome." Nature Medicine 27, no. 4 : 601 – 615. https://doi.org/10.1038/s41591‐021‐01283‐z.</bibtext> </blist> <blist> <bibtext> National Institute for Health and Care Excellence (NICE). 2022. "COVID‐19 Rapid Guideline: Managing the Long‐Term Effects of COVID‐19." https://<ulink href="http://www.nice.org.uk/guidance/ng188/resources/covid19&amp;#8208;rapid&amp;#8208;guideline&amp;#8208;managing&amp;#8208;the&amp;#8208;longterm&amp;#8208;effects&amp;#8208;of&amp;#8208;covid19&amp;#8208;pdf&amp;#8208;51035515742">www.nice.org.uk/guidance/ng188/resources/covid19&amp;#8208;rapid&amp;#8208;guideline&amp;#8208;managing&amp;#8208;the&amp;#8208;longterm&amp;#8208;effects&amp;#8208;of&amp;#8208;covid19&amp;#8208;pdf&amp;#8208;51035515742</ulink>.</bibtext> </blist> <blist> <bibtext> Nyhus, E., and F. Barceló. 2009. " The Wisconsin Card Sorting Test and the Cognitive Assessment of Prefrontal Executive Functions: A Critical Update." Brain and Cognition 71, no. 3 : 437 – 451. https://doi.org/10.1016/j.bandc.2009.03.005.</bibtext> </blist> <blist> <bibtext> O'Mahoney, L. L., A. Routen, C. Gillies, et al. 2023. " The Prevalence and Long‐Term Health Effects of Long Covid Among Hospitalised and Non‐Hospitalised Populations: A Systematic Review and Meta‐Analysis." eClinicalMedicine 55 : 101762. https://doi.org/10.1016/j.eclinm.2022.101762.</bibtext> </blist> <blist> <bibtext> Obeagu, E. I., and G. U. Obeagu. 2024. " Thromboinflammation in COVID‐19: Unraveling the Interplay of Coagulation and Inflammation." Medicine 103, no. 28 : e38922. https://doi.org/10.1097/MD.0000000000038922.</bibtext> </blist> <blist> <bibtext> Office for National Statistics (ONS). 2023. "Prevalence of Ongoing Symptoms Following Coronavirus (COVID‐19) Infection in the UK." March 30, 2023. https://<ulink href="http://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/prevalenceofongoingsymptomsfollowingcoronaviruscovid19infectionintheuk/30march2023">www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/prevalenceofongoingsymptomsfollowingcoronaviruscovid19infectionintheuk/30march2023</ulink>.</bibtext> </blist> <blist> <bibtext> Østergaard, L. 2021. " SARS CoV‐2 Related Microvascular Damage and Symptoms During and After COVID‐19: Consequences of Capillary Transit‐Time Changes, Tissue Hypoxia and Inflammation." Physiological Reports 9, no. 3 : e14726. https://doi.org/10.14814/phy2.14726.</bibtext> </blist> <blist> <bibtext> Petersen, M. S., M. F. Kristiansen, K. D. Hanusson, et al. 2021. " Long COVID in The Faroe Islands: A Longitudinal Study Among Nonhospitalized Patients." Clinical Infectious Diseases 73, no. 11 : e4058 – e4063. https://doi.org/10.1093/cid/ciaa1792.</bibtext> </blist> <blist> <bibtext> Pihlaja, R. E., L. S. Kauhanen, H. S. Ollila, et al. 2023. " Associations of Subjective and Objective Cognitive Functioning After COVID‐19: A Six‐Month Follow‐Up of ICU, Ward, and Home‐Isolated Patients." Brain, Behavior, &amp; Immunity ‐ Health 27 : 100587. https://doi.org/10.1016/j.bbih.2023.100587.</bibtext> </blist> <blist> <bibtext> Pryce‐Roberts, A., M. Talaei, and N. P. Robertson. 2020. " Neurological Complications of COVID‐19: A Preliminary Review." Journal of Neurology 267 : 1870 – 1873. https://doi.org/10.1007/s00415‐020‐09941‐x.</bibtext> </blist> <blist> <bibtext> Quan, M., X. Wang, M. Gong, Q. Wang, Y. Li, and J. Jia. 2023. " Post‐COVID Cognitive Dysfunction: Current Status and Research Recommendations for High Risk Population." Lancet Regional Health. Western Pacific 38 : 100836. https://doi.org/10.1016/j.lanwpc.2023.100836.</bibtext> </blist> <blist> <bibtext> Raveendran, A. V., R. Jayadevan, and S. Sashidharan. 2021. " Long COVID: An Overview." Diabetes &amp; Metabolic Syndrome 15, no. 3 : 869 – 875. https://doi.org/10.1016/j.dsx.2021.04.007.</bibtext> </blist> <blist> <bibtext> Reuschke, D., D. Houston, and P. Sissons. 2024. " Impacts of Long COVID on Workers: A Longitudinal Study of Employment Exit, Work Hours and Mental Health in the UK." PLOS ONE 19, no. 6 : e0306122. https://doi.org/10.1371/journal.pone.0306122.</bibtext> </blist> <blist> <bibtext> Romero‐Sánchez, C. M., I. Díaz‐Maroto, E. Fernández‐Díaz, et al. 2020. " Neurologic Manifestations in Hospitalized Patients With COVID‐19: The ALBACOVID Registry." Neurology 95, no. 8 : e1060 – e1070. https://doi.org/10.1212/WNL.0000000000009937.</bibtext> </blist> <blist> <bibtext> Sarkar, S., S. Karmakar, M. Basu, P. Ghosh, and M. K. Ghosh. 2023. " Neurological Damages in COVID‐19 Patients: Mechanisms and Preventive Interventions." MedComm 4, no. 2 : e247. https://doi.org/10.1002/mco2.247.</bibtext> </blist> <blist> <bibtext> Steardo, L., L. Steardo Jr., R. Zorec, and A. Verkhratsky. 2020. " Neuroinfection May Contribute to Pathophysiology and Clinical Manifestations of COVID‐19." Acta Physiologica 229, no. 3 : e13473. https://doi.org/10.1111/apha.13473.</bibtext> </blist> <blist> <bibtext> Taquet, M., R. Sillett, L. Zhu, et al. 2022. " Neurological and Psychiatric Risk Trajectories After SARS‐CoV‐2 Infection: An Analysis of 2‐Year Retrospective Cohort Studies Including 1 284 437 Patients." Lancet Psychiatry 9, no. 10 : 815 – 827. https://doi.org/10.1016/S2215‐0366(22)00260‐7.</bibtext> </blist> <blist> <bibtext> Taruffi, L., L. Muccioli, M. Mitolo, et al. 2023. " Neurological Manifestations of Long COVID: A Single‐Center One‐Year Experience." Neuropsychiatric Disease and Treatment 19 : 311 – 319. https://doi.org/10.2147/NDT.S387501.</bibtext> </blist> <blist> <bibtext> Xiong, Q., M. Xu, J. Li, et al. 2021. " Clinical Sequelae of COVID‐19 Survivors in Wuhan, China: A Single‐Centre Longitudinal Study." Clinical Microbiology and Infection 27, no. 1 : 89 – 95. https://doi.org/10.1016/j.cmi.2020.09.023.</bibtext> </blist> <blist> <bibtext> Yong, E. 2023. "Long COVID Is Being Erased‐Again." The Atlantic, April 21, 2023. https://<ulink href="http://www.theatlantic.com/health/archive/2023/04/long&amp;#8208;covid&amp;#8208;symptoms&amp;#8208;invisible&amp;#8208;disability&amp;#8208;chronic&amp;#8208;illness/673773/">www.theatlantic.com/health/archive/2023/04/long&amp;#8208;covid&amp;#8208;symptoms&amp;#8208;invisible&amp;#8208;disability&amp;#8208;chronic&amp;#8208;illness/673773/</ulink>.</bibtext> </blist> <blist> <bibtext> Zhao, S., E. M. Martin, P. A. Reuken, et al. 2024. " Long COVID Is Associated With Severe Cognitive Slowing: A Multicentre Cross‐Sectional Study." eClinicalMedicine 68 : 102434. https://doi.org/10.1016/j.eclinm.2024.102434.</bibtext> </blist> <blist> <bibtext> Ziauddeen, N., D. Gurdasani, M. E. O'Hara, et al. 2022. " Characteristics and Impact of Long Covid: Findings From an Online Survey." PLoS One 17, no. 3 : e0264331. https://doi.org/10.1371/journal.pone.0264331.</bibtext> </blist> </ref> <aug> <p>By Sabine P. Yeung; Panyuan Guo; Francess L. Adlard; Seraphina R. Zhang; Vidita Bhagat; Josiah Cho; Lyn Curtis; Muzaffer Kaser; Mark P. Haggard and Lucy G. Cheke</p> <p>Reported by Author; Author; Author; Author; Author; Author; Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib45" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib49" firstref="ref2"></nolink> <nolink nlid="nl3" bibid="bib47" firstref="ref3"></nolink> <nolink nlid="nl4" bibid="bib16" firstref="ref4"></nolink> <nolink nlid="nl5" bibid="bib37" firstref="ref5"></nolink> <nolink nlid="nl6" bibid="bib12" firstref="ref6"></nolink> <nolink nlid="nl7" bibid="bib55" firstref="ref7"></nolink> <nolink nlid="nl8" bibid="bib11" firstref="ref8"></nolink> <nolink nlid="nl9" bibid="bib21" firstref="ref9"></nolink> <nolink nlid="nl10" bibid="bib26" firstref="ref11"></nolink> <nolink nlid="nl11" bibid="bib30" firstref="ref13"></nolink> <nolink nlid="nl12" bibid="bib62" firstref="ref14"></nolink> <nolink nlid="nl13" bibid="bib14" firstref="ref15"></nolink> <nolink nlid="nl14" bibid="bib31" firstref="ref16"></nolink> <nolink nlid="nl15" bibid="bib51" firstref="ref17"></nolink> <nolink nlid="nl16" bibid="bib34" firstref="ref18"></nolink> <nolink nlid="nl17" bibid="bib38" firstref="ref19"></nolink> <nolink nlid="nl18" bibid="bib19" firstref="ref21"></nolink> <nolink nlid="nl19" bibid="bib41" firstref="ref22"></nolink> <nolink nlid="nl20" bibid="bib44" firstref="ref25"></nolink> <nolink nlid="nl21" bibid="bib65" firstref="ref26"></nolink> <nolink nlid="nl22" bibid="bib25" firstref="ref29"></nolink> <nolink nlid="nl23" bibid="bib23" firstref="ref30"></nolink> <nolink nlid="nl24" bibid="bib22" firstref="ref31"></nolink> <nolink nlid="nl25" bibid="bib20" firstref="ref33"></nolink> <nolink nlid="nl26" bibid="bib28" firstref="ref34"></nolink> <nolink nlid="nl27" bibid="bib42" firstref="ref35"></nolink> <nolink nlid="nl28" bibid="bib64" firstref="ref36"></nolink> <nolink nlid="nl29" bibid="bib10" firstref="ref37"></nolink> <nolink nlid="nl30" bibid="bib40" firstref="ref38"></nolink> <nolink nlid="nl31" bibid="bib53" firstref="ref39"></nolink> <nolink nlid="nl32" bibid="bib57" firstref="ref40"></nolink> <nolink nlid="nl33" bibid="bib35" firstref="ref41"></nolink> <nolink nlid="nl34" bibid="bib61" firstref="ref44"></nolink> <nolink nlid="nl35" bibid="bib17" firstref="ref45"></nolink> <nolink nlid="nl36" bibid="bib27" firstref="ref46"></nolink> <nolink nlid="nl37" bibid="bib33" firstref="ref47"></nolink> <nolink nlid="nl38" bibid="bib15" firstref="ref48"></nolink> <nolink nlid="nl39" bibid="bib29" firstref="ref49"></nolink> <nolink nlid="nl40" bibid="bib59" firstref="ref53"></nolink> <nolink nlid="nl41" bibid="bib50" firstref="ref55"></nolink> <nolink nlid="nl42" bibid="bib32" firstref="ref59"></nolink> <nolink nlid="nl43" bibid="bib13" firstref="ref60"></nolink> <nolink nlid="nl44" bibid="bib43" firstref="ref62"></nolink> <nolink nlid="nl45" bibid="bib147" firstref="ref83"></nolink> <nolink nlid="nl46" bibid="bib385" firstref="ref86"></nolink> <nolink nlid="nl47" bibid="bib79" firstref="ref90"></nolink> <nolink nlid="nl48" bibid="bib58" firstref="ref98"></nolink> <nolink nlid="nl49" bibid="bib76" firstref="ref102"></nolink> <nolink nlid="nl50" bibid="bib114" firstref="ref106"></nolink> <nolink nlid="nl51" bibid="bib120" firstref="ref110"></nolink> <nolink nlid="nl52" bibid="bib166" firstref="ref113"></nolink> <nolink nlid="nl53" bibid="bib139" firstref="ref115"></nolink> <nolink nlid="nl54" bibid="bib165" firstref="ref117"></nolink> <nolink nlid="nl55" bibid="bib137" firstref="ref119"></nolink> <nolink nlid="nl56" bibid="bib163" firstref="ref121"></nolink> <nolink nlid="nl57" bibid="bib170" firstref="ref128"></nolink> <nolink nlid="nl58" bibid="bib129" firstref="ref136"></nolink> <nolink nlid="nl59" bibid="bib78" firstref="ref138"></nolink> <nolink nlid="nl60" bibid="bib152" firstref="ref145"></nolink> <nolink nlid="nl61" bibid="bib130" firstref="ref147"></nolink> <nolink nlid="nl62" bibid="bib156" firstref="ref157"></nolink> <nolink nlid="nl63" bibid="bib133" firstref="ref159"></nolink> <nolink nlid="nl64" bibid="bib52" firstref="ref161"></nolink> <nolink nlid="nl65" bibid="bib82" firstref="ref165"></nolink> <nolink nlid="nl66" bibid="bib36" firstref="ref178"></nolink> <nolink nlid="nl67" bibid="bib48" firstref="ref179"></nolink> <nolink nlid="nl68" bibid="bib54" firstref="ref181"></nolink> <nolink nlid="nl69" bibid="bib18" firstref="ref183"></nolink> <nolink nlid="nl70" bibid="bib56" firstref="ref192"></nolink> <nolink nlid="nl71" bibid="bib39" firstref="ref197"></nolink> <nolink nlid="nl72" bibid="bib60" firstref="ref198"></nolink> <nolink nlid="nl73" bibid="bib24" firstref="ref199"></nolink> <nolink nlid="nl74" bibid="bib63" firstref="ref202"></nolink> <nolink nlid="nl75" bibid="bib46" firstref="ref204"></nolink> |
|---|---|
| Header | DbId: eric DbLabel: ERIC An: EJ1468390 AccessLevel: 3 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: COVCOG 3--Trajectory of Long COVID: Longitudinal Changes in Symptoms and Cognitive Impairment. A Third Publication from the COVID and Cognition Study – Name: Language Label: Language Group: Lang Data: English – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Sabine+P%2E+Yeung%22">Sabine P. Yeung</searchLink> (ORCID <externalLink term="https://orcid.org/0000-0002-5229-0279">0000-0002-5229-0279</externalLink>)<br /><searchLink fieldCode="AR" term="%22Panyuan+Guo%22">Panyuan Guo</searchLink><br /><searchLink fieldCode="AR" term="%22Francess+L%2E+Adlard%22">Francess L. Adlard</searchLink><br /><searchLink fieldCode="AR" term="%22Seraphina+R%2E+Zhang%22">Seraphina R. Zhang</searchLink> (ORCID <externalLink term="https://orcid.org/0009-0009-3587-385X">0009-0009-3587-385X</externalLink>)<br /><searchLink fieldCode="AR" term="%22Vidita+Bhagat%22">Vidita Bhagat</searchLink><br /><searchLink fieldCode="AR" term="%22Josiah+Cho%22">Josiah Cho</searchLink><br /><searchLink fieldCode="AR" term="%22Lyn+Curtis%22">Lyn Curtis</searchLink><br /><searchLink fieldCode="AR" term="%22Muzaffer+Kaser%22">Muzaffer Kaser</searchLink><br /><searchLink fieldCode="AR" term="%22Mark+P%2E+Haggard%22">Mark P. Haggard</searchLink><br /><searchLink fieldCode="AR" term="%22Lucy+G%2E+Cheke%22">Lucy G. Cheke</searchLink> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="SO" term="%22Applied+Cognitive+Psychology%22"><i>Applied Cognitive Psychology</i></searchLink>. 2025 39(2). – Name: Avail Label: Availability Group: Avail Data: Wiley. Available from: John Wiley & Sons, Inc. 111 River Street, Hoboken, NJ 07030. Tel: 800-835-6770; e-mail: cs-journals@wiley.com; Web site: https://www.wiley.com/en-us – Name: PeerReviewed Label: Peer Reviewed Group: SrcInfo Data: Y – Name: Pages Label: Page Count Group: Src Data: 22 – Name: DatePubCY Label: Publication Date Group: Date Data: 2025 – Name: TypeDocument Label: Document Type Group: TypDoc Data: Journal Articles<br />Reports - Research – Name: Subject Label: Descriptors Group: Su Data: <searchLink fieldCode="DE" term="%22COVID-19%22">COVID-19</searchLink><br /><searchLink fieldCode="DE" term="%22Pandemics%22">Pandemics</searchLink><br /><searchLink fieldCode="DE" term="%22Chronic+Illness%22">Chronic Illness</searchLink><br /><searchLink fieldCode="DE" term="%22Neurological+Impairments%22">Neurological Impairments</searchLink><br /><searchLink fieldCode="DE" term="%22Thinking+Skills%22">Thinking Skills</searchLink><br /><searchLink fieldCode="DE" term="%22Daily+Living+Skills%22">Daily Living Skills</searchLink><br /><searchLink fieldCode="DE" term="%22Employment+Problems%22">Employment Problems</searchLink><br /><searchLink fieldCode="DE" term="%22Health%22">Health</searchLink><br /><searchLink fieldCode="DE" term="%22Symptoms+%28Individual+Disorders%29%22">Symptoms (Individual Disorders)</searchLink><br /><searchLink fieldCode="DE" term="%22Improvement%22">Improvement</searchLink> – Name: DOI Label: DOI Group: ID Data: 10.1002/acp.70040 – Name: ISSN Label: ISSN Group: ISSN Data: 0888-4080<br />1099-0720 – Name: Abstract Label: Abstract Group: Ab Data: Long COVID has widespread and long-lasting multisystemic impacts on patients' bodies, cognition, and daily functioning, including the ability to work. Longitudinal studies are important in investigating the expected timelines along the course of recovery. This mixed cross-sectional/longitudinal study examines how symptoms (cognitive and noncognitive) and objective cognitive function evolve in post-COVID-19 patients (n = 187) compared to noninfected controls (n = 207). Participants completed a questionnaire about their COVID-19 experience and cognitive tasks at baseline and again at 2-3 follow-ups during a 9-month period. While some noncognitive symptoms improved over time (ds = 0.34-0.87), cognitive symptoms and neurological symptoms, as well as memory function assessed with objective cognitive assessments, remained unimproved (nonsignificant change over time). Neurological symptoms predicted both cognitive symptoms and cognitive impairment across time. Our finding suggested that people with past COVID-19 infection did not experience improvement in cognitive function over time, at least for the duration of this 9-month longitudinal study. – Name: AbstractInfo Label: Abstractor Group: Ab Data: As Provided – Name: DateEntry Label: Entry Date Group: Date Data: 2025 – Name: AN Label: Accession Number Group: ID Data: EJ1468390 |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1468390 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1002/acp.70040 Languages: – Text: English PhysicalDescription: Pagination: PageCount: 22 Subjects: – SubjectFull: COVID-19 Type: general – SubjectFull: Pandemics Type: general – SubjectFull: Chronic Illness Type: general – SubjectFull: Neurological Impairments Type: general – SubjectFull: Thinking Skills Type: general – SubjectFull: Daily Living Skills Type: general – SubjectFull: Employment Problems Type: general – SubjectFull: Health Type: general – SubjectFull: Symptoms (Individual Disorders) Type: general – SubjectFull: Improvement Type: general Titles: – TitleFull: COVCOG 3--Trajectory of Long COVID: Longitudinal Changes in Symptoms and Cognitive Impairment. A Third Publication from the COVID and Cognition Study Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Sabine P. Yeung – PersonEntity: Name: NameFull: Panyuan Guo – PersonEntity: Name: NameFull: Francess L. Adlard – PersonEntity: Name: NameFull: Seraphina R. Zhang – PersonEntity: Name: NameFull: Vidita Bhagat – PersonEntity: Name: NameFull: Josiah Cho – PersonEntity: Name: NameFull: Lyn Curtis – PersonEntity: Name: NameFull: Muzaffer Kaser – PersonEntity: Name: NameFull: Mark P. Haggard – PersonEntity: Name: NameFull: Lucy G. Cheke IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 03 Type: published Y: 2025 Identifiers: – Type: issn-print Value: 0888-4080 – Type: issn-electronic Value: 1099-0720 Numbering: – Type: volume Value: 39 – Type: issue Value: 2 Titles: – TitleFull: Applied Cognitive Psychology Type: main |
| ResultId | 1 |
