https://orcid.org/0000-0002-3216-9025
Abstract
Although Coronavirus disease 2019 (COVID-19) is primarily a respiratory infectious disease, it has also been associated with a wide range of other clinical manifestations. It is widely accepted in the scientific community that many patients after recovery continue to experience COVID-19-related symptoms, including cognitive impairment. The aim of this systematic review was to investigate the cognitive profile of patients with long-COVID syndrome.
A systematic search of empirical studies was conducted through the PubMed/Medline and Scopus electronic databases. Cross-sectional and longitudinal studies published between 2020 and 2023 were included.
Of the 516 studies assessed for eligibility, 36 studies met the inclusion criteria. All included studies support the presence of persistent cognitive changes after COVID-19 disease. Executive function, memory, attention, and processing speed appear to be the cognitive domains that are predominantly associated with long-COVID syndrome, whereas language is an area that has not been sufficiently investigated.
In this review, the high frequency of cognitive impairment after COVID-19 is evident. If we consider that cognitive functioning affects our ability to live independently and is a key determinant of our quality of life, it is imperative to precisely define those factors that may induce cognitive impairment in COVID-19 survivors, with the ultimate goal of early diagnosis of cognitive changes and, consequently, the development of targeted rehabilitation interventions to address them.
INTRODUCTION
In December 2019, the first cases of the new severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) were reported in Wuhan, China. Coronavirus disease 2019 (COVID-19) is caused by coronavirus 2 causing SARS-CoV-2. COVID-19 has rapidly spread all over the world despite important efforts to contain it. The majority of individuals infected with COVID-19 experience mild to moderate illness, whereas approximately 10%–15% develop severe illness and 5% become critically ill. The virus is considered one of the most infectious viruses affecting human society. This pathogen results in a syndrome leading in numerous cases to a critical care respiratory condition that requires specialized management in intensive care units.
Common symptoms included fever, dry cough, fatigue, and dyspnea, whereas respiratory failure and subsequent pneumonia frequently led to hospitalization (Rodriguez-Morales et al., 2020). Neurological manifestations (such as headache, myalgias, or loss of smell) were increasingly recognized (Almeria et al., 2020). Patients with COVID-19 presented with a wide range of neurological manifestations that are due to the injury of the central nervous system (CNS) and peripheral nervous system (PNS) via cytokine storm, blood clots, direct damage by SARS-CoV-2, and/or molecular mimicry (Fotuhi et al., 2020).
At least one fifth of cases required supportive care in intensive care units, which was limited in most developing countries. Despite the implementation of optimal supportive interventions, the case fatality rate among hospitalized patients exceeded 10%. Similar to other viral respiratory pathogens, COVID-19 presented in the majority of cases with a rapidly progressive course of fever, cough, and dyspnea (Rodriguez-Morales et al., 2020). Although COVID-19 is primarily regarded as a respiratory disease, it has also been significantly associated with a wide range of neuropsychiatric manifestations because of its invasion of both the CNS and PNS.
Several studies have suggested that symptoms of COVID-19 persist even after the initial disease, resulting in the development and widespread use of a new term, “long COVID”. “Long COVID” is defined as a condition in which symptoms persist for more than 12 weeks after the onset of COVID-19 and cannot be explained by an alternative diagnosis. It is an umbrella term, characterized by symptom heterogeneity, possibly related to a variety of distinct pathological mechanisms.
One of the most common symptoms observed in long COVID is impaired cognitive function. An increasing number of studies are investigating the cognitive profile of patients suffering from long-COVID syndrome through the administration of both screening tests and more comprehensive neuropsychological test batteries. Although there is inconsistency in the results of the different research studies, impairment in memory, attention, executive function, and language appears to be closely associated with COVID-19 (Matias-Guiu et al., 2023).
However, which factors might induce these changes in cognitive functions in COVID-19 survivors is not clear. Significantly, the scientific community has turned its attention to investigating the possible association of long-term cognitive impairment with inflammatory factors. Indeed, a recent study has supported the long-term impairment of cognitive functions, in particular executive functions, and their association with inflammatory profiles (He et al., 2023). Moreover, the presence of post-COVID sleep disturbance may have a negative impact on cognitive functions, while sociodemographic as well as clinical parameters (i.e., high comorbidity, low physical exercise, more severe frailty pre-COVID-19) were found to be associated with poorer cognitive performance in long-COVID syndrome.
As people began to describe their protracted recovery from COVID-19 infection, expressions like “brain fog” were coined to capture a group of unseen difficulties. This expression covered an array of cognitive and linguistic problems such as memory loss, word-finding difficulties, poor concentration, and difficulty following and remembering a topic of conversation (Cummings, 2021). A primary focus of diagnostics is to document the presence or absence of impairment throughout the continuum of care, as well as to identify the level of impairment in each cognitive domain (e.g., executive function, visuospatial processing, memory, orientation) and how these affect the ability to communicate (Ramage, 2020). Whereas changes in memory and executive function have been reported in post-COVID-19 patients, much less is known about the nature of their language performance (language expression and comprehension abilities).
Several studies have explored cognitive function after COVID-19, shedding light on the enduring neurological effects of the virus. Scientists have been delving into the complexities of post-acute sequelae of SARS-CoV-2 infection (PASC) to understand the persistence of cognitive deficits in individuals who have recovered from COVID-19. Investigations have encompassed a spectrum of cognitive domains, including memory, attention, executive function, and processing speed. Researchers have explored potential mechanisms, such as viral neurotropism, inflammatory responses, and immune system involvement, to unravel the underlying causes of these enduring effects on the brain. Thus, the purpose of this systematic review was to retrieve relevant studies that assessed the long-term neurocognitive impact of the disease and address what has been reported on neurocognitive impairment and the resulting consequences in COVID-19 survivors. In addition, our focus was to report clinical data specifically concerning the neurocognitive deficits in the adult population with COVID.
METHODS
The present systematic review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guidelines (Page et al., 2021).
Eligibility Criteria
We included observational studies (cross-sectional and longitudinal) quantitatively assessing adult patients with COVID-19 on different modalities and cognitive functions by means of standardized screening tests or comprehensive batteries. Inclusion criteria were studies in English, and COVID-19 infection confirmed by serology or PCR with new or deteriorated cognitive impairment confirmed by cognitive assessment. Studies were included when cognitive assessment time from onset of COVID-19 was defined at or more than 12 weeks. There was no restriction regarding gender or ethnicity. Studies were excluded which included participants with a history of cognitive decline, dementia, or mild cognitive disorders as well as those with pre-existing psychiatric, neurological, or neurodevelopmental conditions that could potentially affect cognitive functions. The presence of subjective cognitive complaints in patients following COVID-19 illness was not a criterion for exclusion. Additionally, studies were excluded if they included participants with severe, uncontrollable motor or sensory impairments that would interfere with neuropsychological assessments. Including such participants could introduce confusion and compromise the reliability of the findings. Furthermore, studies with participants under the age of 18 were not considered. Reviews, meta-analyses, opinion papers, abstracts, and research protocols were excluded.
Search Strategy
The online search strategy was set until May 31, 2023, through three of the major public scientific databases, PubMed/Medline, and Scopus. The following search terms were entered: (“CORONAVIRUS” OR “SARS-CoV-2” OR “COVID-19”) AND (“neurocognitive impairment” OR “neurocognitive deficit” OR “cognitive impairment” OR “cognition”). For all scientific databases, the fields of search were title, abstract, and keywords. Gray literature was not searched. Additional studies that were manually retrieved have been included.
Study Selection Process and Data Collection
Initially, three researchers independently conducted a database search and sent it to the Zotero reference manager to exclude duplicate documents. After duplicate exclusion, the title and abstract screening was performed independently by two independent investigators (EP and MCP). For any disagreement that arose, the study was re-evaluated by the same researchers and if they still disagreed, then a third researcher (ΕΑ) made the final decision. Full-text-review data extraction was performed by three independent authors (EP, MCP, and EA), whereas a fourth independent rater (LM) checked the extracted data and resolved disagreements. The resulting studies were evaluated according to eligibility criteria. During the synthesis process, the analysis of relationships within and between the described studies led to an overall assessment of the strength of the available evidence for drawing conclusions based on a narrative synthesis. A table form was used for data extraction once consensus was achieved on selected studies. Information extracted included author, year, title, study design, country, aim of study, median age, sample size and type, disease severity (or clinical characteristics), education, type of assessment, time of evaluation from COVID-19 onset, neurocognitive domains assessed, neurocognitive and functional assessment instruments. Heterogeneity was qualitatively explored in the structure of the analysis process, considering methodological similarities and differences among the included studies.
RESULTS
Characteristics of the Included Studies
A total of 406 studies were derived from the PubMed/Medline database and 110 studies from Scopus. Thirty-six studies were finally included in the systematic review. The PRISMA flowchart of the study is presented in Fig. 1. Study characteristics are included in Table 1.
PRISMA 2020 flow diagram for the systematic review, which included searches of Pubmed/Medline and Scopus electronic databases.
Studies Assessing Cognition 12 weeks post Covid-19 infection
| Author/year | Country | Study design | Time cognitive assessment from onset infection COVID-19 | Participants | Aim of study | Cognitive test | Results |
|---|---|---|---|---|---|---|---|
| Abramoff et al., 2023 | USA | Observational cohort study | >12–30+ weeks | Total 324 outpatients 74 patients were hospitalized, mean age (SD) =46.6(14.0) Education age = 37 patients (11.9%) has <12 formal education | To examine the factors associated with variation in their psychological and cognitive functioning and health-related quality of life | MoCA1, Hospital anxiety and depression scale, European Quality of Life Scale instrument | Scores below 18 on the education-adjusted MoCA scale, serve as the threshold for classifying patients with cognitive impairment. Alongside reporting group means, the analysis also includes the distribution of scores falling within low or very low ranges, accompanied by effect sizes and odds ratios. Among survivors seeking care for persistent COVID symptoms, 17% exhibited cognitive impairment. Individuals from poor and African American backgrounds experienced more severe psychological and cognitive effects following infection |
| Albu et al., 2021 | Spain | Cross-sectional | 89–124 days after | discharged 30 patients (16 post-ICU patients and 14 non-ICU patients | describes sequelae and persistent symptoms at more than 3 months after acute COVID-19 and their impact on activities of daily living, sleep, and quality of life in patients in an outpatient rehabilitation program | Barcelona Test (orientation), Digit Span forward subtest of the Wechsler Adult Intelligence Scale III (immediate attention), RAVLT2 (verbal memory), Digit Span backward subtest of the Wechsler Adult Intelligence Scale III (working memory), PMR task (executive function), HADS3 | Low scores were observed across various cognitive domains, including orientation, attention, verbal learning, long-term verbal memory, verbal recognition, working memory, and executive control. However, none of these differences were statistically significant, as indicated by X2 values ranging from 0.18 to 1.77 and p-values ranging from .18 to .89. Cognitive impairment was detected in 63.3% of patients, with a comparable profile observed in both subgroups |
| Amalakanti et al., 2021 | India | Case control study | Outpatients Not clear | 93 asymptomatic patients (mean age 36.2 ± 11.7); 102 controls (mean age 35.6 ± 9.8) | to detect MCI in asymptomatic patients subjects COVID-19 with MOCA | MoCA | For individuals aged 18–19, the average MoCA score in cases was 25.9 with a standard deviation of 2.1, compared to 27 with a standard deviation of 1.7 in controls. Among those aged 30–49, controls had an average MoCA score of 25.6 with a standard deviation of 4.3. For individuals aged 50 and above, cases had an average MoCA score of 24 with a standard deviation of 3.5, while controls had an average score of 24.5 with a standard deviation of 3.5 |
| Becker et al., 2021 | USA | Cross-sectional | 7.6 months after | Ambulatory or Discharged Total = 740 mean age = 49 (38–59) years 464 patients were women 516 individuals had other comorbidities (103) patients have educational level ≤ 12 years | to investigate rates of cognitive impairment in survivors of COVID-19 who were treated in outpatient, emergency department (ED), or inpatient hospital setting | Number Span Forward and Backward (attention and working memory), TMT A & B (processing speed and e.f4), phonemic and category Fluency (language) and the Hopkins Verbal Learning Test-revised (memory encoding, recall, and recognition) | Hospitalized patients demonstrated higher odds ratios of impairments in attention (OR: 2.8; 95% CI: 1.3–5.9), executive functioning (OR: 1.8; 95% CI: 1.0–3.4), category fluency (OR: 5.2; 95% CI: 3.0–9.0), memory encoding (OR: 2.3; 95% CI: 1.3–4.1), and memory recall (OR: 2.2; 95% CI: 1.3–3.8) compared to the outpatient group. Emergency department (ED) patients exhibited greater impairment in category fluency (OR: 1.8; 95% CI: 1.1–3.1) and memory encoding (OR: 1.7; 95% CI: 1.0–3.0) compared to outpatients |
| Birberg Thornberg et al., 2022 | Sweden | an ambidirectional population-based cohort study | 5 months after discharge | total 133 patients | to explore objective neurocognitive deficits in COVID-19 patients five months after discharge, and any associations with demographic factors and disease severity indicators | RBANS5, CWIT6, D-KEFS7, HADS, MFI8 | In the RBANS global cognition index, 45 patients (36.9%) scored below 1.5 SD, and 28 (23.0%) scored below 2 SD. The most prevalent deficits were observed in the Coding subtest, assessing psychomotor speed, with 47 patients (36.2%) affected, followed by the Semantic fluency subtest, evaluating verbal ability, with 34 patients (25.8%), and the Digit span subtest, assessing short-term memory and auditory attention, with 31 patients (23.7%). Additionally, performance on the CWIT was below the cutoff for 10.6% of the patients |
| Boesl et al., 2021 | Germany | Retrospective cohort study | Meantime 184.5 days after | 100 ambulatory patients Range age (20–79) Mean age 45.8 | assessed clinical data of their first 100 ambulatory patients regarding clinical presentations and a screening assessment for detecting cognitive impairment ≥12 weeks after the acute infection with SARS-CoV-2. | MoCA, BDI9, FSS10,ESS11 | A total of 30% of the patients with cognitive deficits showed pathologic MoCA scores (≤25/30 points) |
| Brown et al., 2022 | USA | Observational cohort study | Mean 22.4 weeks (SD 12.8) Range 1–51 | total 297 patients (66 hospitalized; 20 ICU) Middle age 46.9 9(SD) = 14.1 | studied specific associations between cognitive impairment, self-reported disruptions in memory, psychiatric symptoms (depression, anxiety and pos-ttraumatic depression) and sleep disorders | HADS, IES-612(post-traumatic stress reactions), MoCA -Blind, Self-reported memory disturbance | Depression symptom severity was significantly associated with MoCA total score. MoCA Digit Span-Backward, MoCA Fluency, MoCA Delay recall were all significantly associated with depression |
| Chen et al., 2022 | Georgia | Prospective Cohort Study (primarily findings) | on average 125 days after | 200 patients (81 patients with comorbid condition) Mean age 44.6 (range 19–82) | iwas established to investigate the severity and chronicity of these neurologic findings over the 5-year period following infection | MoCA, NIH-TB13 (List Sorting Working Memory Test and Picture Vocabulary Test), PHQ-914,GAD-715 | Over time, there was a notable improvement in performance (d = 0.97, p = .03). However, scores remained below the cutoff for detecting cognitive impairment (<26 out of 30). NIH Toolbox cognitive scores, adjusted for demographic factors, showed that scores below the 25th national percentile indicated impairment. 47% of participants who completed the MoCA met the criteria for mild cognitive impairment. 30% of participants demonstrated impairment on vocabulary and 32% demonstrated impairment on working memory |
| Crivelli et al., 2022 | Argentina | Prospective cohort study | 142 days from disease onset | 45 outpatients; 45 controls Median age = 50 years Median age education = 17 years | to investigate the impact of COVID- 19 on cognitive functions in forty-five post COVID patients and forty -five healthy controls. Patients were evaluated for an average of 142 days after illness | MoCA, TMT A & B16,Digit Span Forwards, Digit- Symbol Coding, Craft Story, RAVLT, Benson Figure, Wisconsin Card Sorting Test, Stroop Test, Phonological Fluency, Clock Drawing Test, Multilingual Naming Test, Semantic Fluency, HADS and WHODAS 2.0 Functional Scale | No significant differences were found in the screening measures (MoCA p = .15). Significant differences between groups were found in cognitive composites of memory (p = .016, Cohen’s d = 0.73) attention (p < .001, Cohen’s d = 1.2), executive functions p < .001, Cohen’s d = 1.4), and language (p < .002, Cohen’s d = 87) |
| Damiano et al., 2022 | Brazil | Cohort study (preliminary findings) | 6–9 months after (mean 207 days, SD 20.4) | 425 patients Mean age = 55.7 55.5% participants <12 years of education | to investigate the occurrence of psychiatric and cognitive impairments in a cohort of survivors of moderate or severe forms of COVID-19 | CIS-R17,SCID-5-RV18,HADS,ASQ19,PCL-C20, AUDIT21,MCS22,Temporal and Spatial Orientation23,TMT Part A, VFT24,CFS25,IPAQ-SV26 | The study found deficits on executive and attentional domains. Memory Decline was subjectively reported by 51.1% of the patients. All patients performed worse in TMT- A across all ages. Older age and disorientation were associated with a worse performance on the TMT-A. Older age, higher Frailty Scores (prior COVID) and temporal- spatial disorientation were associated with worse performance in the Verbal Flency Test |
| Delgado-Alonso et al., 2022 | Spain | Cross-sectional study | 9.12 ± 3.46 months after the onset of disease | 50 patients; 50 controls Mean age = 51.06 ± 11.65 Mean years education = 13.58 ± 4.01 | to determine the characteristics of cognitive dysfunction in patients reporting cognitive complaints after COVID-19 and to evaluate the correlation between cognitive function and anxiety, depression, sleep, and olfactory function | Corsi Block-Tapping Test, SDMT27.BNT28,JLO29, ROCF30, FCSRT31,Stroop Color-Word Interference Test, VOSP32,Computerized Neuropsychological Battery Vienna Test System33,BSIT34, STAI35, BDI-II36, PSQI37, Modified Fatigue Impact Scale,3.0 T MRI38 | Cognitive impairment based on performance scores on various neuropsychological assessments compared to normative data -scaled scores (this score is equivalent to a percentile of ≤5 or a z-score of < −1.65, which categorizes a performance as cognitively impaired), normative data which provide representative scores for a cognitively healthy population in Spain and for tests administered via a computerized system, cognitive impairment is assessed by calculating z-scores.COVID-19 group showed worst scores in the recall and recognition trials of the FGT, Inhibition Test, NBV, TMT Part A & B and in several visual tasks of the WAF Battery. Cognitive performance was correlated with olfactory dysfunction and sleep quality and anxiety to a lesser extent, but not depression |
| Del Brutto et al., 2021 | Ecuador | Prospective cohort study | 6 months from disease onset | 52 patients; 41 controls Mean age = 62.6 ± 11 years Comorbid condition | design, the occurrence of cognitive decline among individuals with a history of mild symptomatic SARS-CoV-2 infection was assessed | MoCA | Post- COVID MoCA scores were notably lower than the mean scores before the pandemic in seropositive individuals (21.7 ± 4 vs. 19.6 ± 4.2; p = 0.010), but not in those who tested negative for antibodies (21.5 ± 5 vs. 21 ± 4; p = 0.618 |
| Duindam et al., 2022 | Netherlands | Prospective observational cohort | 6 months after hospital discharge | 96 patients IQR age = 61 (55–69) years 38% patients = low level education 90.6% had comorbid condition | was to study long term cognitive outcomes following severe COVID-19 and the relation to anti-inflammatory therapies | MOCA, TMT Part A & B, Letter Digit Substitution Test, Digit Span, NART-IQ39,, HADS, CFS,BSI40, CFQ41 | After 6.5 ± 1.3 months, 27% of patients exhibited cognitive impairment. There was no correlation between objective cognitive dysfunction and subjective cognitive dysfunction. Patients who received both dexamethasone and tocilizumab showed poorer executive functioning scores (Trail Making Test). Specifically, 21% of patients scored impaired on both TMT-A and TMT-B, while 18% scored impaired on the Digit Span Test |
| Ferrando et al., 2022 | USA | Cross-sectional study | 6–8 months after | 60 patients (clinical group = 32 and nonclinical group = 28) Mean age = 41 years 39% prior psychiatric history On average 1.5 comorbid medical comorbidities | to describe neuropsychological, medical, psychiatric, and functional correlates of cognitive complaints experienced after recovery from acute COVID-19 infection | IADL42,Chalder Fatigue Scale, PHQ-943, QLESQ44, Test Of Premorbid Function, PAOF45,RBANS TMT Part A & B, Verbal Fluency, Stroop Color- Word Test, Posttraumatic Stress Disorder Checklist, Serological samples, | Subjective cognitive function suggests mild to moderate perceived cognitive difficulties related to memory, language, and cognition. The overall sample scored significantly lower than normative values on the RBANS total score, as well as on subtests measuring immediate and delayed memory and language |
| Ferrucci et al., 2021 | Italy | Cross-sectional | 4–5 months after (mean_ SD = 4.43–1.22 months) | 38 patients Years range = 22–74 Mean age education = 12.39 | To study the occurrence of cognitive abnormalities in the months following hospital discharge | MoCA, BRB-NT46, BDI-II, SSD questionnaire47 | Cognitive impairment was present in 60.5% of the sample, indicated by scores falling below cutoffs on at least one task of the BRB-NT. Among them, 42.1% exhibited deficits in processing speed, while 26.3% displayed delayed verbal recall deficits, and 10.5% showed deficits in immediate verbal recall |
| Frontera et al., 2021 | USA | Prospective cohort study | 6 months after | 196 patients with new neurological and 186 patients non neurological | conducted a prospective study of 6-month outcomes of hospitalized COVID-19 patients. Patients with new neurological complications during hospitalization who survived were propensity score-matched to COVID-19 survivors without neurological complications hospitalized during the same period | Barthel Index for activities of daily living, Telephone MoCA, modified Rankin Scale, Neuro-Qol48 Via tele assistance | Half of the individuals exhibited impaired cognition, indicated by a telephone MoCA score below 18 |
| García-Sánchez et al., 2022 | Spain | Prospective cohort study | 187 days after | 63 patients (33 previous hospitalized) Mean age (SD) =51.5 (12.5) Mean level education = 14.4 years | explore the frequency of impaired performance across cognitive domains in post-COVID patients with subjective complaints and uncover whether impairment existed within a single domain or across multiple. | MoCA, CPT-II49,RAVLT, ROCFT50, Digit Span Forward and Backward, BNT, Block Design Test, Symbol Search, TMT Part A & B, Stroop Task, Phonetic and Semantic Fluency Tasks, 15-OT51, HADS | In patients with single-domain impairment, attention deficits were the most common, observed in 19.0% of cases, significantly surpassing deficits in executive functioning (p = 0.01), short-term/working memory (p = 0.001), and language (p < 0.001). Additionally, among patients with multiple-domain impairment, attention was the cognitive domain most frequently affected, particularly in combination with deficits in learning, long-term memory, and executive functioning |
| Gautam et al., 2022 | United Kingdom | Retrospective cases series | Mean days after COVID onset (SD) = 143.4 (42.4) | 200 Discharged patients. Mean age(SD) = 56.5(13.2) Half of patients with comorbid condition | assessed the medium- term effects of COVID-19 on survivors of severe disease | MOCA | Some degree of cognitive impairment, particularly affecting concentration, and short-term recall, was observed in 12.5% of patients |
| Graham et al., 2021 | USA | Prospective cohort study | On average at 4.72 months + group compared to 5.82 month | 100 ambulatory patients; 36 with cognitive evaluation | to characterize the spectrum of neurologic manifestations in non-hospitalized Covid-19 “long haulers | PROMIS52,NIH Toolbox53 48 patients through tele visits and 52 in-person | The cognitive performance of COVID patients on tasks related to attention and working memory was inferior when compared to a demographically similar population from the USA |
| Hampshire et al., 2022 | United Kingdom | Cohort Study | On average 179 days after illness onset (SD) = 62 | 46 patients Mean age = 51 years Age range = 28–33 | to study a possible association between severe COVID-19 and persistent cognitive deficits | Custom Computerized cognitive assessment Battery, GAD-754,PHQ-9, PCL-555 | Deficits exhibited a notable effect size among individuals who had undergone hospitalization |
| Hartung et al., 2022 | Germany | Prospective multicenter study | ≥6 months after | 969 patients; matched control group | To investigate frequency of post-COVID fatigue and cognitive impairment | MoCA, FACIT Fatigue Scale, PHQ-8, GAD-7, PSQI | 26% of patients experienced mild cognitive impairment, while 1% exhibited moderate impairment. Older age, male gender, shorter education, and a history of neuropsychiatric disease were associated with cognitive impairment. There was no significant correlation found between fatigue and cognitive impairment |
| He et al., 2023 | China | Observational cross-sectional study | At 15 months after discharge | 66 patients-majority mild (median age = 35.50) and 79 healthy controls Range age total = 17–61 years | To investigate cognitive function and neuropsychiatric symptoms of 66 COVID-19 survivors and 79 healthy controls (HCs) using a digital, gamified cognitive function evaluation tool and questionnaires at 15 months after discharge. Venous blood samples were collected to measure cytokine levels | THINC-it tool56 | Criteria for determining cognitive impairment from the cognitive tests involve comparisons to a healthy control group matched by sex, age, and education level. COVID-19 survivors showed long-term cognitive impairment in executive function even at 15 months after illness onset |
| Krishnan et al., 2022 | USA | Observational Study | 5.5 after infection On average 168 days after illness onset | Total 20 patients—average education of 15 years 85% of patients with previous mental- medical history | to document cognitive function among individuals recovering from COVID-19 and to identify factors that were associated with outcomes, including preexisting health conditions and severity of COVID-19 illness | WMS-IV57,Logical memory, RAVLT, Brief Visuospatial Memory Test—Revised, Reading subtest from the wide Range Achievement Test-IV, BNT- Second Edition, Lexical and semantic verbal fluencies, Line orientation, Digit Span, Matrix Reasoning, Similarities subtests58,DKEFS59,TMT, Wisconsin Card Sorting Test, Conners Continuous Performance Test-3, SDMT, Coding and Symbol Search60 | Defines cognitive impairment in post-COVID-19 patients as performance on neuropsychological tests that falls 1.5 SD or more below the mean based on published normative data. Mild cognitive deficits were seen on test involving attention and processing or executive function. |
| Lamontagne et al., 2021 | Canada and USA | Observational study | Mean (SD) 123.63 (94.71) days after infection | 50 patients (3 subgroups; acute Covid-19,PASC and post PASC) and 50 healthy control Patients age range = 20–53 years Mean age education (SD) = 16.2(2.95) | investigated mood and cognitive functioning in individuals who have recovered from COVID-19 infection | Self- reported measures of stress, depression and anhedonia, Attention Network Test and cognitive abilities (Attentional Control Scale) | The criteria for identifying impairment appear to rely on substantial statistical distinctions in task performance between the COVID-19 group and controls, rather than employing a set numerical threshold. The most notable effects were observed in individuals diagnosed 1–4 months before evaluation. The COVID-19 recovery group exhibited significantly higher scores on perceived stress. Within the COVID-19 group, there was a specific deficit in attention, characterized by impaired executive functioning while alertness and orienting abilities remained unaffected |
| Lauria et al., 2022 | Italy | Single-center study | On average 96.5 days after the onset of COVID-19 symptoms | 100 patients Mean age(SD) = 73.4(6.2) Mean years education (SD) = 12.7(8.7) | To investigate the neurologic and cognitive features of a sample of elderly patients with confirmed diagnosis of COVID-19 evaluated in the post-acute phase through a direct neuropsychological evaluation | MMSE, FAB61,RAVLT, Multiple Features Target Cancellation Test, TMT, Digit Span forward and backward | For the MMSE, the adjusted score averaged 28.2 ± 1.7, as anticipated for a study population comprising relatively educated individuals with no history of cognitive impairment. For the FAB, 20% of patients exhibited either pathologic or borderline performances. On the Rey Auditory Verbal Learning Test, 11% of patients showed either pathologic or borderline performances on immediate recall, and 17% on delayed recall. In the Multiple Features Target Cancellation Test, 4% of patients demonstrated either pathologic or borderline performances. Regarding the Trail Making Test, 33% of patients displayed either pathologic or borderline performances. Lastly, in the Digit Span forward and backward tests, 12% and 23% of patients, respectively, exhibited either pathologic or borderline performances |
| Matias-Guiu et al., 2023 | Spain | Mean 484.60 ± 174.46 days after the acute infection | 404 Patients; mean age = 48.6 ± 9.2 145 controls Clinical comorbidities in the post COVID group | To develop objective criteria for cognitive dysfunction associated with the post-COVID syndrome | SDMT, Stroop Test, FCRST62,ROCF63, Digit Span Backwards, VOSP64, BNT, Semantic Fluency, WAIS-IV 65Matrix reasoning, JLO66,TAVEC67,Brief Test of Attention, Object Naming | Using cutoff points of −1 and − 1.5 SD below the normative mean for the cognitive tests administered. Processing speed and attention were the cognitive domains most affected, with 41.2% and 17.3% of patients classified as having impairment in at least one cognitive domain. Younger age and lower levels of education were linked to cognitive impairment, whereas hospitalization did not show any association | |
| Miskowiak et al., 2021 | Denmark | Prospective cohort study | 3–4 months after | 29 Discharged patients; matched 100 healthy controls. Mean age post-COVID = 56.2 (10.6) Mean age education = 14.3(3.9) Charlson comorbidity score = 2.9(2.4) | To investigate the frequency, pattern and severity of cognitive impairments after COVID-19 hospital discharge, their relation to subjective cognitive complaints, quality of life and illness variables | SCIP-D68,TMT Part B | Using a 0.5 standard deviation cut-off for SCIP total scores, a total of 19 patients, constituting 65% of the sample, were categorized as cognitively impaired |
| Miskowiak et al., 2023 | Denmark | Prospective cohort study | an average of 7 (±4) months | 194 patients; mean age = 50.8(15.4); mean age education = 15.3(3.9) 150 healthy controls Charlson comorbidity score = 34 (19) | To investigate the frequency, pattern, and functional implications of cognitive impairments in patients at a long-COVID clinic who were referred after hospitalization with COVID-19 | SCIP-D, TMT Part B | SCIP total score: statistically significant lower performance in patients than healthy controls (p < .001). Verbal learning and memory: statistically significant lower performance in patients than healthy controls (p = .001 and p = .005, respectively). Working memory: statistically significant lower performance in patients than healthy controls (p < .001). Verbal fluency: statistically significant lower performance in patients than healthy controls (p = .004). Processing speed: no statistically significant differences between patients and healthy controls. MT-B: statistically significant lower performance in patients according to expected scores based on age, sex, and education |
| Ortelli et al., 2021 | Italy | Cross-sectional | 9–13 weeks after | 12 Discharged patients; age 67 ± 9.6 12 healthy controls;age 64.3 ± 10.5 Total patients develop neurological complications after COVID | to provide a comprehensive clinical, neurophysiological, and neuropsychological profile of fatigued patients suffering from neurological manifestations related to SARS-CoV-2, who recovered from the acute phase of COVID-19 | MoCA, FAB, three computerized attentive tasks69 | MoCA & FAB: statistically significant lower scores in patients compared to healthy controls (p < .001) Computerized tasks: Reaction times were significantly longer in patients than healthy controls in both SIT (p = .015) and NT (p = .046). Percentage of errors was significantly larger in patients than healthy controls in all three tasks (p < .001) |
| Pihlaja et al., 2023 | Finland | Cohort study | At 3 and 6 months after infection | 184 patients (mean age(SD) =53.4 (12.8) Reported comorbidities. Non-COVID control group = 53 individuals (mean age(SD) = 54.9(12.3) | To describe the prevalence of subjective and objective cognitive dysfunction at 3 and 6 months after COVID-19 and the associations of subjective cognitive symptoms and psychological and disease-related factors | MoCA | At 6 months, there was a statistically significant difference between ICU, WARD, and HOME groups (p < .001). However, there was no statistically significant difference between all COVID-19 patients and control group |
| Pilotto et al., 2021 | Italy | Prospective cohort study | 6 months after | 105 Discharged patients (no premorbid neurological conditions) Three subgroups (mild, moderate, severe) Average age = 64.8 (±12.6) | to evaluate prevalence and predictors of long-term neurological manifestations. | MoCA | 17.5% of participants had scores indicative of cognitive impairment |
| Rass et al., 2021 | Austria | Multicenter observational cohort study | 3 months after COVID-19 (102 days [IQR] = [91–110] days) | 135 patients Median age 56 (range = 19–87) 33 patients pre-diagnosed neurological disorders before COVID | to assess neurological manifestations and health-related quality of life (QoL) | MoCA | Cognitive impairment, as measured by the MoCA, were identified in 23% of the patient population. Among those with severe COVID-19, the prevalence was 29%, while it was 30% in moderate cases and 3% in mild cases |
| Rubega et al., 2022 | Italy | Observational study | 12 months after discharge | 33 total patients (2 subgroups) n = 16 ICU patients age interval = 50–80 n = 17 non-ICU patients age interval = 53–78 12 healthy controls matched | to study long-term cognitive and affective sequelae and sleep high-density electroencephalography (EEG) in people with a previous hospital admission for acute COVID-19. | MoCA, FAB, Stroop task, Digit Span forward and backward, RAVLT, TMT PART A & B, SDMT | Only on the TMT-B there was a trend towards worse performance in COVID-19 patients, specifically in non-ICU patients (CTRL vs. non-ICU: p = .04). On the remaining tests, no statistically significant differences between the groups were found |
| Vannorsdall et al., 2022 | USA | Prospective cohort study | Mean time diagnosis (SD) = 126.5 (70.1) | 82 Discharged patients (n = 48 post -ICU; n = 34 post non ICU) 95% participants = ≤12 years education Mean age 54.5 years | to prospectively characterize cognition, mental health symptoms, and functioning approximately 4 months after an initial diagnosis of COVID-19 in a racially and ethnically diverse group of patients. | RAVLT, Oral Trail Making Test70,Digit Span forward and backward, Semantic and phonemic fluency via tele assistance | RAVLT acquisition and delayed recall: 26.8% of patients were mildly/moderately impaired and 14.6% were severely impaired. TMT-A: 35.4% of patients were mildly/moderately impaired and 25.6% were severely impaired. TMT-B: 15.9% of patients were mildly/moderately impaired and 13.4% were severely impaired. Digit Span forward: 1.2% of patients were mildly/moderately impaired and 1.2% were severely impaired. Digit Span backward: 13.4% of patients were mildly/moderately impaired and 4.9% were severely impaired. Phonemic fluency: 25.6% of patients were mildly/moderately impaired and 8.5% were severely impaired. Semantic fluency: 31.7% of patients were mildly/moderately impaired and 9.8% were severely impaired |
| Walle-Hansen et al., 2021 | Norway | Prospective cohort study | 6 months after | 106 Discharged patients (subdivided by age into 2 subgroups) (ambulatory evaluation) Mean age = 74.3 (range 60–96) Comorbid condition | To study age-related change in health-related quality of life, functional decline and mortality among older patients following hospitalization due to COVID-19 | MoCA | A negative change in cognitive functions was reported by 43% of patients 6 months after hospitalization. MoCA scores were lower in the oldest age group (mean: 21.7) compared to younger participants (mean: 25.3) |
| Woo et al., 2020 | Germany | Cross-sectional study | Median: 85 days from COVID-19 recovery [range: 20–105 days] | 18 patients (mean age [SD] = 42.2 [14.3]) 10 healthy controls (mean age [SD] = 38.4 [14.4]) | Study focused on neurocognitive deficits in mild COVID-19 young patients | TICS-M (by phone or directly) | The performance of COVID-19 patients was significantly lower (mean: 38.83; range: 31–46) than the healthy controls (mean: 45.8; range: 43–50), especially on short-term memory, attention, and concentration/language tasks |
| Author/year | Country | Study design | Time cognitive assessment from onset infection COVID-19 | Participants | Aim of study | Cognitive test | Results |
|---|---|---|---|---|---|---|---|
| Abramoff et al., 2023 | USA | Observational cohort study | >12–30+ weeks | Total 324 outpatients 74 patients were hospitalized, mean age (SD) =46.6(14.0) Education age = 37 patients (11.9%) has <12 formal education | To examine the factors associated with variation in their psychological and cognitive functioning and health-related quality of life | MoCA1, Hospital anxiety and depression scale, European Quality of Life Scale instrument | Scores below 18 on the education-adjusted MoCA scale, serve as the threshold for classifying patients with cognitive impairment. Alongside reporting group means, the analysis also includes the distribution of scores falling within low or very low ranges, accompanied by effect sizes and odds ratios. Among survivors seeking care for persistent COVID symptoms, 17% exhibited cognitive impairment. Individuals from poor and African American backgrounds experienced more severe psychological and cognitive effects following infection |
| Albu et al., 2021 | Spain | Cross-sectional | 89–124 days after | discharged 30 patients (16 post-ICU patients and 14 non-ICU patients | describes sequelae and persistent symptoms at more than 3 months after acute COVID-19 and their impact on activities of daily living, sleep, and quality of life in patients in an outpatient rehabilitation program | Barcelona Test (orientation), Digit Span forward subtest of the Wechsler Adult Intelligence Scale III (immediate attention), RAVLT2 (verbal memory), Digit Span backward subtest of the Wechsler Adult Intelligence Scale III (working memory), PMR task (executive function), HADS3 | Low scores were observed across various cognitive domains, including orientation, attention, verbal learning, long-term verbal memory, verbal recognition, working memory, and executive control. However, none of these differences were statistically significant, as indicated by X2 values ranging from 0.18 to 1.77 and p-values ranging from .18 to .89. Cognitive impairment was detected in 63.3% of patients, with a comparable profile observed in both subgroups |
| Amalakanti et al., 2021 | India | Case control study | Outpatients Not clear | 93 asymptomatic patients (mean age 36.2 ± 11.7); 102 controls (mean age 35.6 ± 9.8) | to detect MCI in asymptomatic patients subjects COVID-19 with MOCA | MoCA | For individuals aged 18–19, the average MoCA score in cases was 25.9 with a standard deviation of 2.1, compared to 27 with a standard deviation of 1.7 in controls. Among those aged 30–49, controls had an average MoCA score of 25.6 with a standard deviation of 4.3. For individuals aged 50 and above, cases had an average MoCA score of 24 with a standard deviation of 3.5, while controls had an average score of 24.5 with a standard deviation of 3.5 |
| Becker et al., 2021 | USA | Cross-sectional | 7.6 months after | Ambulatory or Discharged Total = 740 mean age = 49 (38–59) years 464 patients were women 516 individuals had other comorbidities (103) patients have educational level ≤ 12 years | to investigate rates of cognitive impairment in survivors of COVID-19 who were treated in outpatient, emergency department (ED), or inpatient hospital setting | Number Span Forward and Backward (attention and working memory), TMT A & B (processing speed and e.f4), phonemic and category Fluency (language) and the Hopkins Verbal Learning Test-revised (memory encoding, recall, and recognition) | Hospitalized patients demonstrated higher odds ratios of impairments in attention (OR: 2.8; 95% CI: 1.3–5.9), executive functioning (OR: 1.8; 95% CI: 1.0–3.4), category fluency (OR: 5.2; 95% CI: 3.0–9.0), memory encoding (OR: 2.3; 95% CI: 1.3–4.1), and memory recall (OR: 2.2; 95% CI: 1.3–3.8) compared to the outpatient group. Emergency department (ED) patients exhibited greater impairment in category fluency (OR: 1.8; 95% CI: 1.1–3.1) and memory encoding (OR: 1.7; 95% CI: 1.0–3.0) compared to outpatients |
| Birberg Thornberg et al., 2022 | Sweden | an ambidirectional population-based cohort study | 5 months after discharge | total 133 patients | to explore objective neurocognitive deficits in COVID-19 patients five months after discharge, and any associations with demographic factors and disease severity indicators | RBANS5, CWIT6, D-KEFS7, HADS, MFI8 | In the RBANS global cognition index, 45 patients (36.9%) scored below 1.5 SD, and 28 (23.0%) scored below 2 SD. The most prevalent deficits were observed in the Coding subtest, assessing psychomotor speed, with 47 patients (36.2%) affected, followed by the Semantic fluency subtest, evaluating verbal ability, with 34 patients (25.8%), and the Digit span subtest, assessing short-term memory and auditory attention, with 31 patients (23.7%). Additionally, performance on the CWIT was below the cutoff for 10.6% of the patients |
| Boesl et al., 2021 | Germany | Retrospective cohort study | Meantime 184.5 days after | 100 ambulatory patients Range age (20–79) Mean age 45.8 | assessed clinical data of their first 100 ambulatory patients regarding clinical presentations and a screening assessment for detecting cognitive impairment ≥12 weeks after the acute infection with SARS-CoV-2. | MoCA, BDI9, FSS10,ESS11 | A total of 30% of the patients with cognitive deficits showed pathologic MoCA scores (≤25/30 points) |
| Brown et al., 2022 | USA | Observational cohort study | Mean 22.4 weeks (SD 12.8) Range 1–51 | total 297 patients (66 hospitalized; 20 ICU) Middle age 46.9 9(SD) = 14.1 | studied specific associations between cognitive impairment, self-reported disruptions in memory, psychiatric symptoms (depression, anxiety and pos-ttraumatic depression) and sleep disorders | HADS, IES-612(post-traumatic stress reactions), MoCA -Blind, Self-reported memory disturbance | Depression symptom severity was significantly associated with MoCA total score. MoCA Digit Span-Backward, MoCA Fluency, MoCA Delay recall were all significantly associated with depression |
| Chen et al., 2022 | Georgia | Prospective Cohort Study (primarily findings) | on average 125 days after | 200 patients (81 patients with comorbid condition) Mean age 44.6 (range 19–82) | iwas established to investigate the severity and chronicity of these neurologic findings over the 5-year period following infection | MoCA, NIH-TB13 (List Sorting Working Memory Test and Picture Vocabulary Test), PHQ-914,GAD-715 | Over time, there was a notable improvement in performance (d = 0.97, p = .03). However, scores remained below the cutoff for detecting cognitive impairment (<26 out of 30). NIH Toolbox cognitive scores, adjusted for demographic factors, showed that scores below the 25th national percentile indicated impairment. 47% of participants who completed the MoCA met the criteria for mild cognitive impairment. 30% of participants demonstrated impairment on vocabulary and 32% demonstrated impairment on working memory |
| Crivelli et al., 2022 | Argentina | Prospective cohort study | 142 days from disease onset | 45 outpatients; 45 controls Median age = 50 years Median age education = 17 years | to investigate the impact of COVID- 19 on cognitive functions in forty-five post COVID patients and forty -five healthy controls. Patients were evaluated for an average of 142 days after illness | MoCA, TMT A & B16,Digit Span Forwards, Digit- Symbol Coding, Craft Story, RAVLT, Benson Figure, Wisconsin Card Sorting Test, Stroop Test, Phonological Fluency, Clock Drawing Test, Multilingual Naming Test, Semantic Fluency, HADS and WHODAS 2.0 Functional Scale | No significant differences were found in the screening measures (MoCA p = .15). Significant differences between groups were found in cognitive composites of memory (p = .016, Cohen’s d = 0.73) attention (p < .001, Cohen’s d = 1.2), executive functions p < .001, Cohen’s d = 1.4), and language (p < .002, Cohen’s d = 87) |
| Damiano et al., 2022 | Brazil | Cohort study (preliminary findings) | 6–9 months after (mean 207 days, SD 20.4) | 425 patients Mean age = 55.7 55.5% participants <12 years of education | to investigate the occurrence of psychiatric and cognitive impairments in a cohort of survivors of moderate or severe forms of COVID-19 | CIS-R17,SCID-5-RV18,HADS,ASQ19,PCL-C20, AUDIT21,MCS22,Temporal and Spatial Orientation23,TMT Part A, VFT24,CFS25,IPAQ-SV26 | The study found deficits on executive and attentional domains. Memory Decline was subjectively reported by 51.1% of the patients. All patients performed worse in TMT- A across all ages. Older age and disorientation were associated with a worse performance on the TMT-A. Older age, higher Frailty Scores (prior COVID) and temporal- spatial disorientation were associated with worse performance in the Verbal Flency Test |
| Delgado-Alonso et al., 2022 | Spain | Cross-sectional study | 9.12 ± 3.46 months after the onset of disease | 50 patients; 50 controls Mean age = 51.06 ± 11.65 Mean years education = 13.58 ± 4.01 | to determine the characteristics of cognitive dysfunction in patients reporting cognitive complaints after COVID-19 and to evaluate the correlation between cognitive function and anxiety, depression, sleep, and olfactory function | Corsi Block-Tapping Test, SDMT27.BNT28,JLO29, ROCF30, FCSRT31,Stroop Color-Word Interference Test, VOSP32,Computerized Neuropsychological Battery Vienna Test System33,BSIT34, STAI35, BDI-II36, PSQI37, Modified Fatigue Impact Scale,3.0 T MRI38 | Cognitive impairment based on performance scores on various neuropsychological assessments compared to normative data -scaled scores (this score is equivalent to a percentile of ≤5 or a z-score of < −1.65, which categorizes a performance as cognitively impaired), normative data which provide representative scores for a cognitively healthy population in Spain and for tests administered via a computerized system, cognitive impairment is assessed by calculating z-scores.COVID-19 group showed worst scores in the recall and recognition trials of the FGT, Inhibition Test, NBV, TMT Part A & B and in several visual tasks of the WAF Battery. Cognitive performance was correlated with olfactory dysfunction and sleep quality and anxiety to a lesser extent, but not depression |
| Del Brutto et al., 2021 | Ecuador | Prospective cohort study | 6 months from disease onset | 52 patients; 41 controls Mean age = 62.6 ± 11 years Comorbid condition | design, the occurrence of cognitive decline among individuals with a history of mild symptomatic SARS-CoV-2 infection was assessed | MoCA | Post- COVID MoCA scores were notably lower than the mean scores before the pandemic in seropositive individuals (21.7 ± 4 vs. 19.6 ± 4.2; p = 0.010), but not in those who tested negative for antibodies (21.5 ± 5 vs. 21 ± 4; p = 0.618 |
| Duindam et al., 2022 | Netherlands | Prospective observational cohort | 6 months after hospital discharge | 96 patients IQR age = 61 (55–69) years 38% patients = low level education 90.6% had comorbid condition | was to study long term cognitive outcomes following severe COVID-19 and the relation to anti-inflammatory therapies | MOCA, TMT Part A & B, Letter Digit Substitution Test, Digit Span, NART-IQ39,, HADS, CFS,BSI40, CFQ41 | After 6.5 ± 1.3 months, 27% of patients exhibited cognitive impairment. There was no correlation between objective cognitive dysfunction and subjective cognitive dysfunction. Patients who received both dexamethasone and tocilizumab showed poorer executive functioning scores (Trail Making Test). Specifically, 21% of patients scored impaired on both TMT-A and TMT-B, while 18% scored impaired on the Digit Span Test |
| Ferrando et al., 2022 | USA | Cross-sectional study | 6–8 months after | 60 patients (clinical group = 32 and nonclinical group = 28) Mean age = 41 years 39% prior psychiatric history On average 1.5 comorbid medical comorbidities | to describe neuropsychological, medical, psychiatric, and functional correlates of cognitive complaints experienced after recovery from acute COVID-19 infection | IADL42,Chalder Fatigue Scale, PHQ-943, QLESQ44, Test Of Premorbid Function, PAOF45,RBANS TMT Part A & B, Verbal Fluency, Stroop Color- Word Test, Posttraumatic Stress Disorder Checklist, Serological samples, | Subjective cognitive function suggests mild to moderate perceived cognitive difficulties related to memory, language, and cognition. The overall sample scored significantly lower than normative values on the RBANS total score, as well as on subtests measuring immediate and delayed memory and language |
| Ferrucci et al., 2021 | Italy | Cross-sectional | 4–5 months after (mean_ SD = 4.43–1.22 months) | 38 patients Years range = 22–74 Mean age education = 12.39 | To study the occurrence of cognitive abnormalities in the months following hospital discharge | MoCA, BRB-NT46, BDI-II, SSD questionnaire47 | Cognitive impairment was present in 60.5% of the sample, indicated by scores falling below cutoffs on at least one task of the BRB-NT. Among them, 42.1% exhibited deficits in processing speed, while 26.3% displayed delayed verbal recall deficits, and 10.5% showed deficits in immediate verbal recall |
| Frontera et al., 2021 | USA | Prospective cohort study | 6 months after | 196 patients with new neurological and 186 patients non neurological | conducted a prospective study of 6-month outcomes of hospitalized COVID-19 patients. Patients with new neurological complications during hospitalization who survived were propensity score-matched to COVID-19 survivors without neurological complications hospitalized during the same period | Barthel Index for activities of daily living, Telephone MoCA, modified Rankin Scale, Neuro-Qol48 Via tele assistance | Half of the individuals exhibited impaired cognition, indicated by a telephone MoCA score below 18 |
| García-Sánchez et al., 2022 | Spain | Prospective cohort study | 187 days after | 63 patients (33 previous hospitalized) Mean age (SD) =51.5 (12.5) Mean level education = 14.4 years | explore the frequency of impaired performance across cognitive domains in post-COVID patients with subjective complaints and uncover whether impairment existed within a single domain or across multiple. | MoCA, CPT-II49,RAVLT, ROCFT50, Digit Span Forward and Backward, BNT, Block Design Test, Symbol Search, TMT Part A & B, Stroop Task, Phonetic and Semantic Fluency Tasks, 15-OT51, HADS | In patients with single-domain impairment, attention deficits were the most common, observed in 19.0% of cases, significantly surpassing deficits in executive functioning (p = 0.01), short-term/working memory (p = 0.001), and language (p < 0.001). Additionally, among patients with multiple-domain impairment, attention was the cognitive domain most frequently affected, particularly in combination with deficits in learning, long-term memory, and executive functioning |
| Gautam et al., 2022 | United Kingdom | Retrospective cases series | Mean days after COVID onset (SD) = 143.4 (42.4) | 200 Discharged patients. Mean age(SD) = 56.5(13.2) Half of patients with comorbid condition | assessed the medium- term effects of COVID-19 on survivors of severe disease | MOCA | Some degree of cognitive impairment, particularly affecting concentration, and short-term recall, was observed in 12.5% of patients |
| Graham et al., 2021 | USA | Prospective cohort study | On average at 4.72 months + group compared to 5.82 month | 100 ambulatory patients; 36 with cognitive evaluation | to characterize the spectrum of neurologic manifestations in non-hospitalized Covid-19 “long haulers | PROMIS52,NIH Toolbox53 48 patients through tele visits and 52 in-person | The cognitive performance of COVID patients on tasks related to attention and working memory was inferior when compared to a demographically similar population from the USA |
| Hampshire et al., 2022 | United Kingdom | Cohort Study | On average 179 days after illness onset (SD) = 62 | 46 patients Mean age = 51 years Age range = 28–33 | to study a possible association between severe COVID-19 and persistent cognitive deficits | Custom Computerized cognitive assessment Battery, GAD-754,PHQ-9, PCL-555 | Deficits exhibited a notable effect size among individuals who had undergone hospitalization |
| Hartung et al., 2022 | Germany | Prospective multicenter study | ≥6 months after | 969 patients; matched control group | To investigate frequency of post-COVID fatigue and cognitive impairment | MoCA, FACIT Fatigue Scale, PHQ-8, GAD-7, PSQI | 26% of patients experienced mild cognitive impairment, while 1% exhibited moderate impairment. Older age, male gender, shorter education, and a history of neuropsychiatric disease were associated with cognitive impairment. There was no significant correlation found between fatigue and cognitive impairment |
| He et al., 2023 | China | Observational cross-sectional study | At 15 months after discharge | 66 patients-majority mild (median age = 35.50) and 79 healthy controls Range age total = 17–61 years | To investigate cognitive function and neuropsychiatric symptoms of 66 COVID-19 survivors and 79 healthy controls (HCs) using a digital, gamified cognitive function evaluation tool and questionnaires at 15 months after discharge. Venous blood samples were collected to measure cytokine levels | THINC-it tool56 | Criteria for determining cognitive impairment from the cognitive tests involve comparisons to a healthy control group matched by sex, age, and education level. COVID-19 survivors showed long-term cognitive impairment in executive function even at 15 months after illness onset |
| Krishnan et al., 2022 | USA | Observational Study | 5.5 after infection On average 168 days after illness onset | Total 20 patients—average education of 15 years 85% of patients with previous mental- medical history | to document cognitive function among individuals recovering from COVID-19 and to identify factors that were associated with outcomes, including preexisting health conditions and severity of COVID-19 illness | WMS-IV57,Logical memory, RAVLT, Brief Visuospatial Memory Test—Revised, Reading subtest from the wide Range Achievement Test-IV, BNT- Second Edition, Lexical and semantic verbal fluencies, Line orientation, Digit Span, Matrix Reasoning, Similarities subtests58,DKEFS59,TMT, Wisconsin Card Sorting Test, Conners Continuous Performance Test-3, SDMT, Coding and Symbol Search60 | Defines cognitive impairment in post-COVID-19 patients as performance on neuropsychological tests that falls 1.5 SD or more below the mean based on published normative data. Mild cognitive deficits were seen on test involving attention and processing or executive function. |
| Lamontagne et al., 2021 | Canada and USA | Observational study | Mean (SD) 123.63 (94.71) days after infection | 50 patients (3 subgroups; acute Covid-19,PASC and post PASC) and 50 healthy control Patients age range = 20–53 years Mean age education (SD) = 16.2(2.95) | investigated mood and cognitive functioning in individuals who have recovered from COVID-19 infection | Self- reported measures of stress, depression and anhedonia, Attention Network Test and cognitive abilities (Attentional Control Scale) | The criteria for identifying impairment appear to rely on substantial statistical distinctions in task performance between the COVID-19 group and controls, rather than employing a set numerical threshold. The most notable effects were observed in individuals diagnosed 1–4 months before evaluation. The COVID-19 recovery group exhibited significantly higher scores on perceived stress. Within the COVID-19 group, there was a specific deficit in attention, characterized by impaired executive functioning while alertness and orienting abilities remained unaffected |
| Lauria et al., 2022 | Italy | Single-center study | On average 96.5 days after the onset of COVID-19 symptoms | 100 patients Mean age(SD) = 73.4(6.2) Mean years education (SD) = 12.7(8.7) | To investigate the neurologic and cognitive features of a sample of elderly patients with confirmed diagnosis of COVID-19 evaluated in the post-acute phase through a direct neuropsychological evaluation | MMSE, FAB61,RAVLT, Multiple Features Target Cancellation Test, TMT, Digit Span forward and backward | For the MMSE, the adjusted score averaged 28.2 ± 1.7, as anticipated for a study population comprising relatively educated individuals with no history of cognitive impairment. For the FAB, 20% of patients exhibited either pathologic or borderline performances. On the Rey Auditory Verbal Learning Test, 11% of patients showed either pathologic or borderline performances on immediate recall, and 17% on delayed recall. In the Multiple Features Target Cancellation Test, 4% of patients demonstrated either pathologic or borderline performances. Regarding the Trail Making Test, 33% of patients displayed either pathologic or borderline performances. Lastly, in the Digit Span forward and backward tests, 12% and 23% of patients, respectively, exhibited either pathologic or borderline performances |
| Matias-Guiu et al., 2023 | Spain | Mean 484.60 ± 174.46 days after the acute infection | 404 Patients; mean age = 48.6 ± 9.2 145 controls Clinical comorbidities in the post COVID group | To develop objective criteria for cognitive dysfunction associated with the post-COVID syndrome | SDMT, Stroop Test, FCRST62,ROCF63, Digit Span Backwards, VOSP64, BNT, Semantic Fluency, WAIS-IV 65Matrix reasoning, JLO66,TAVEC67,Brief Test of Attention, Object Naming | Using cutoff points of −1 and − 1.5 SD below the normative mean for the cognitive tests administered. Processing speed and attention were the cognitive domains most affected, with 41.2% and 17.3% of patients classified as having impairment in at least one cognitive domain. Younger age and lower levels of education were linked to cognitive impairment, whereas hospitalization did not show any association | |
| Miskowiak et al., 2021 | Denmark | Prospective cohort study | 3–4 months after | 29 Discharged patients; matched 100 healthy controls. Mean age post-COVID = 56.2 (10.6) Mean age education = 14.3(3.9) Charlson comorbidity score = 2.9(2.4) | To investigate the frequency, pattern and severity of cognitive impairments after COVID-19 hospital discharge, their relation to subjective cognitive complaints, quality of life and illness variables | SCIP-D68,TMT Part B | Using a 0.5 standard deviation cut-off for SCIP total scores, a total of 19 patients, constituting 65% of the sample, were categorized as cognitively impaired |
| Miskowiak et al., 2023 | Denmark | Prospective cohort study | an average of 7 (±4) months | 194 patients; mean age = 50.8(15.4); mean age education = 15.3(3.9) 150 healthy controls Charlson comorbidity score = 34 (19) | To investigate the frequency, pattern, and functional implications of cognitive impairments in patients at a long-COVID clinic who were referred after hospitalization with COVID-19 | SCIP-D, TMT Part B | SCIP total score: statistically significant lower performance in patients than healthy controls (p < .001). Verbal learning and memory: statistically significant lower performance in patients than healthy controls (p = .001 and p = .005, respectively). Working memory: statistically significant lower performance in patients than healthy controls (p < .001). Verbal fluency: statistically significant lower performance in patients than healthy controls (p = .004). Processing speed: no statistically significant differences between patients and healthy controls. MT-B: statistically significant lower performance in patients according to expected scores based on age, sex, and education |
| Ortelli et al., 2021 | Italy | Cross-sectional | 9–13 weeks after | 12 Discharged patients; age 67 ± 9.6 12 healthy controls;age 64.3 ± 10.5 Total patients develop neurological complications after COVID | to provide a comprehensive clinical, neurophysiological, and neuropsychological profile of fatigued patients suffering from neurological manifestations related to SARS-CoV-2, who recovered from the acute phase of COVID-19 | MoCA, FAB, three computerized attentive tasks69 | MoCA & FAB: statistically significant lower scores in patients compared to healthy controls (p < .001) Computerized tasks: Reaction times were significantly longer in patients than healthy controls in both SIT (p = .015) and NT (p = .046). Percentage of errors was significantly larger in patients than healthy controls in all three tasks (p < .001) |
| Pihlaja et al., 2023 | Finland | Cohort study | At 3 and 6 months after infection | 184 patients (mean age(SD) =53.4 (12.8) Reported comorbidities. Non-COVID control group = 53 individuals (mean age(SD) = 54.9(12.3) | To describe the prevalence of subjective and objective cognitive dysfunction at 3 and 6 months after COVID-19 and the associations of subjective cognitive symptoms and psychological and disease-related factors | MoCA | At 6 months, there was a statistically significant difference between ICU, WARD, and HOME groups (p < .001). However, there was no statistically significant difference between all COVID-19 patients and control group |
| Pilotto et al., 2021 | Italy | Prospective cohort study | 6 months after | 105 Discharged patients (no premorbid neurological conditions) Three subgroups (mild, moderate, severe) Average age = 64.8 (±12.6) | to evaluate prevalence and predictors of long-term neurological manifestations. | MoCA | 17.5% of participants had scores indicative of cognitive impairment |
| Rass et al., 2021 | Austria | Multicenter observational cohort study | 3 months after COVID-19 (102 days [IQR] = [91–110] days) | 135 patients Median age 56 (range = 19–87) 33 patients pre-diagnosed neurological disorders before COVID | to assess neurological manifestations and health-related quality of life (QoL) | MoCA | Cognitive impairment, as measured by the MoCA, were identified in 23% of the patient population. Among those with severe COVID-19, the prevalence was 29%, while it was 30% in moderate cases and 3% in mild cases |
| Rubega et al., 2022 | Italy | Observational study | 12 months after discharge | 33 total patients (2 subgroups) n = 16 ICU patients age interval = 50–80 n = 17 non-ICU patients age interval = 53–78 12 healthy controls matched | to study long-term cognitive and affective sequelae and sleep high-density electroencephalography (EEG) in people with a previous hospital admission for acute COVID-19. | MoCA, FAB, Stroop task, Digit Span forward and backward, RAVLT, TMT PART A & B, SDMT | Only on the TMT-B there was a trend towards worse performance in COVID-19 patients, specifically in non-ICU patients (CTRL vs. non-ICU: p = .04). On the remaining tests, no statistically significant differences between the groups were found |
| Vannorsdall et al., 2022 | USA | Prospective cohort study | Mean time diagnosis (SD) = 126.5 (70.1) | 82 Discharged patients (n = 48 post -ICU; n = 34 post non ICU) 95% participants = ≤12 years education Mean age 54.5 years | to prospectively characterize cognition, mental health symptoms, and functioning approximately 4 months after an initial diagnosis of COVID-19 in a racially and ethnically diverse group of patients. | RAVLT, Oral Trail Making Test70,Digit Span forward and backward, Semantic and phonemic fluency via tele assistance | RAVLT acquisition and delayed recall: 26.8% of patients were mildly/moderately impaired and 14.6% were severely impaired. TMT-A: 35.4% of patients were mildly/moderately impaired and 25.6% were severely impaired. TMT-B: 15.9% of patients were mildly/moderately impaired and 13.4% were severely impaired. Digit Span forward: 1.2% of patients were mildly/moderately impaired and 1.2% were severely impaired. Digit Span backward: 13.4% of patients were mildly/moderately impaired and 4.9% were severely impaired. Phonemic fluency: 25.6% of patients were mildly/moderately impaired and 8.5% were severely impaired. Semantic fluency: 31.7% of patients were mildly/moderately impaired and 9.8% were severely impaired |
| Walle-Hansen et al., 2021 | Norway | Prospective cohort study | 6 months after | 106 Discharged patients (subdivided by age into 2 subgroups) (ambulatory evaluation) Mean age = 74.3 (range 60–96) Comorbid condition | To study age-related change in health-related quality of life, functional decline and mortality among older patients following hospitalization due to COVID-19 | MoCA | A negative change in cognitive functions was reported by 43% of patients 6 months after hospitalization. MoCA scores were lower in the oldest age group (mean: 21.7) compared to younger participants (mean: 25.3) |
| Woo et al., 2020 | Germany | Cross-sectional study | Median: 85 days from COVID-19 recovery [range: 20–105 days] | 18 patients (mean age [SD] = 42.2 [14.3]) 10 healthy controls (mean age [SD] = 38.4 [14.4]) | Study focused on neurocognitive deficits in mild COVID-19 young patients | TICS-M (by phone or directly) | The performance of COVID-19 patients was significantly lower (mean: 38.83; range: 31–46) than the healthy controls (mean: 45.8; range: 43–50), especially on short-term memory, attention, and concentration/language tasks |
Montreal Cognitive Assessment Scale
Rey Auditory Verbal Learning Test
Hospital Anxiety and Depression Scale
Executive functions
Repeatable Battery for the Assessment of Neuropsychological Status
Colour-Word Interference Test
Delis-Kaplan Executive Function System
Multidimensional Fatigue Inventory
Beck Depression Inventory (Version I)
Fatigue Severity Scale
Epworth Sleepiness Scale
Impact of Event Scale—Revised
Assessment of Neurological and Behavioral Function
Patient Health Questionnaire-9
Generalized Anxiety Disorder-7
Trial Making Test Part A & Part B
Clinical Interview Schedule-Revised
Structured Clinical Interview for DSM-5, Research Version
Ask Suicide-Screening Questions
Post-Traumatic Stress Disorder Checklist
Alcohol Use Disorder Identification Test
Memory Complaint Scale
As obtained from the Mini Mental State Examination
Verbal Fluency Test
Clinical Frailty Scale
International Physical Activity Questionnaire -Short Version
Symbol Digit Modalities Test
Boston Naming Test
Judgment Line Orientation
Rey-Osterrieth Complex Figure (copy and recall at 3,30 min and recognition)
Free and Cued Selective Reminding Test
Visual Object and Space Perception Battery
With the following tests: Trail making Test (TMT), Figural Memory Test(FGT,S11 form), Tower of London (TOL-F),Inhibition Response (INHIB, S13 Form),N-Back Verbal Test,Cognitrone, Reaction Test, Determination Test, WAB Battery (S1 Form) of perception and attention functions
Brief Smell Identification Test
State- Trait Anxiety Inventory
Beck Depression Inventory -II
Pittsburgh Sleep Quality Index
Magnetic Resonance Imaging
National Adult Reading Test
Brief Symptom Inventory
Cognitive Failure Questionnaire
Lawton–Brody Instrumental Activities of Daily Living Scale
Patient Health Questionnaire-9
Endicott Quality of Life Enjoyment and Satisfaction Scale
Patient Assessment of Own Function
Brief Repeatable Battery Neuropsychological Tests. The BRB-NT includes the Selective Reminding Test, Spatial Recall Test, Symbol Digit Modalities Test, the Paced Auditory Serial Addition Test, and the word List Generation Test
Subjective Scale of Damage questionnaire
Quality of life in Neurological Disorders
Commers Continuous Performance Test II
Rey-Osterrieth Complex Figure Test
15- Objects Test
Patient-reported Outcome Measurement Information System
National Institutes of Health Toolbox v2.1
Custom Computerized Assessment Battery
PTSD Checklist for Diagnostic and Statistical Manual of Mental Disorders
Digital Simplified Chinese Version
Wechsler Memory Scale -IV
From the Wechsler Adult Intelligence Scale -IV
Color-Word Interference
Subtests from the WAIS-IV
Frontal Assessment Battery
Free and Cued Selective Reminding Test
Rey–Osterrieth Complex Figure
Visual Object and Space Perception Battery
Wechsler Adult Intelligence Scale 4th Edition
Judgment Line Orientation
Test de Aprendizaje Verbal Espana-Complutense
Screen for Cognitive Impairment in Psychiatry Danish Version with five subtests of verbal learning and memory, working memory, verbal fluency and processing speed
Vigilance Task (VT), Stroop Interference Task (SIT) and Navon Task (NT
Part A and B
Studies Assessing Cognition 12 weeks post Covid-19 infection
| Author/year | Country | Study design | Time cognitive assessment from onset infection COVID-19 | Participants | Aim of study | Cognitive test | Results |
|---|---|---|---|---|---|---|---|
| Abramoff et al., 2023 | USA | Observational cohort study | >12–30+ weeks | Total 324 outpatients 74 patients were hospitalized, mean age (SD) =46.6(14.0) Education age = 37 patients (11.9%) has <12 formal education | To examine the factors associated with variation in their psychological and cognitive functioning and health-related quality of life | MoCA1, Hospital anxiety and depression scale, European Quality of Life Scale instrument | Scores below 18 on the education-adjusted MoCA scale, serve as the threshold for classifying patients with cognitive impairment. Alongside reporting group means, the analysis also includes the distribution of scores falling within low or very low ranges, accompanied by effect sizes and odds ratios. Among survivors seeking care for persistent COVID symptoms, 17% exhibited cognitive impairment. Individuals from poor and African American backgrounds experienced more severe psychological and cognitive effects following infection |
| Albu et al., 2021 | Spain | Cross-sectional | 89–124 days after | discharged 30 patients (16 post-ICU patients and 14 non-ICU patients | describes sequelae and persistent symptoms at more than 3 months after acute COVID-19 and their impact on activities of daily living, sleep, and quality of life in patients in an outpatient rehabilitation program | Barcelona Test (orientation), Digit Span forward subtest of the Wechsler Adult Intelligence Scale III (immediate attention), RAVLT2 (verbal memory), Digit Span backward subtest of the Wechsler Adult Intelligence Scale III (working memory), PMR task (executive function), HADS3 | Low scores were observed across various cognitive domains, including orientation, attention, verbal learning, long-term verbal memory, verbal recognition, working memory, and executive control. However, none of these differences were statistically significant, as indicated by X2 values ranging from 0.18 to 1.77 and p-values ranging from .18 to .89. Cognitive impairment was detected in 63.3% of patients, with a comparable profile observed in both subgroups |
| Amalakanti et al., 2021 | India | Case control study | Outpatients Not clear | 93 asymptomatic patients (mean age 36.2 ± 11.7); 102 controls (mean age 35.6 ± 9.8) | to detect MCI in asymptomatic patients subjects COVID-19 with MOCA | MoCA | For individuals aged 18–19, the average MoCA score in cases was 25.9 with a standard deviation of 2.1, compared to 27 with a standard deviation of 1.7 in controls. Among those aged 30–49, controls had an average MoCA score of 25.6 with a standard deviation of 4.3. For individuals aged 50 and above, cases had an average MoCA score of 24 with a standard deviation of 3.5, while controls had an average score of 24.5 with a standard deviation of 3.5 |
| Becker et al., 2021 | USA | Cross-sectional | 7.6 months after | Ambulatory or Discharged Total = 740 mean age = 49 (38–59) years 464 patients were women 516 individuals had other comorbidities (103) patients have educational level ≤ 12 years | to investigate rates of cognitive impairment in survivors of COVID-19 who were treated in outpatient, emergency department (ED), or inpatient hospital setting | Number Span Forward and Backward (attention and working memory), TMT A & B (processing speed and e.f4), phonemic and category Fluency (language) and the Hopkins Verbal Learning Test-revised (memory encoding, recall, and recognition) | Hospitalized patients demonstrated higher odds ratios of impairments in attention (OR: 2.8; 95% CI: 1.3–5.9), executive functioning (OR: 1.8; 95% CI: 1.0–3.4), category fluency (OR: 5.2; 95% CI: 3.0–9.0), memory encoding (OR: 2.3; 95% CI: 1.3–4.1), and memory recall (OR: 2.2; 95% CI: 1.3–3.8) compared to the outpatient group. Emergency department (ED) patients exhibited greater impairment in category fluency (OR: 1.8; 95% CI: 1.1–3.1) and memory encoding (OR: 1.7; 95% CI: 1.0–3.0) compared to outpatients |
| Birberg Thornberg et al., 2022 | Sweden | an ambidirectional population-based cohort study | 5 months after discharge | total 133 patients | to explore objective neurocognitive deficits in COVID-19 patients five months after discharge, and any associations with demographic factors and disease severity indicators | RBANS5, CWIT6, D-KEFS7, HADS, MFI8 | In the RBANS global cognition index, 45 patients (36.9%) scored below 1.5 SD, and 28 (23.0%) scored below 2 SD. The most prevalent deficits were observed in the Coding subtest, assessing psychomotor speed, with 47 patients (36.2%) affected, followed by the Semantic fluency subtest, evaluating verbal ability, with 34 patients (25.8%), and the Digit span subtest, assessing short-term memory and auditory attention, with 31 patients (23.7%). Additionally, performance on the CWIT was below the cutoff for 10.6% of the patients |
| Boesl et al., 2021 | Germany | Retrospective cohort study | Meantime 184.5 days after | 100 ambulatory patients Range age (20–79) Mean age 45.8 | assessed clinical data of their first 100 ambulatory patients regarding clinical presentations and a screening assessment for detecting cognitive impairment ≥12 weeks after the acute infection with SARS-CoV-2. | MoCA, BDI9, FSS10,ESS11 | A total of 30% of the patients with cognitive deficits showed pathologic MoCA scores (≤25/30 points) |
| Brown et al., 2022 | USA | Observational cohort study | Mean 22.4 weeks (SD 12.8) Range 1–51 | total 297 patients (66 hospitalized; 20 ICU) Middle age 46.9 9(SD) = 14.1 | studied specific associations between cognitive impairment, self-reported disruptions in memory, psychiatric symptoms (depression, anxiety and pos-ttraumatic depression) and sleep disorders | HADS, IES-612(post-traumatic stress reactions), MoCA -Blind, Self-reported memory disturbance | Depression symptom severity was significantly associated with MoCA total score. MoCA Digit Span-Backward, MoCA Fluency, MoCA Delay recall were all significantly associated with depression |
| Chen et al., 2022 | Georgia | Prospective Cohort Study (primarily findings) | on average 125 days after | 200 patients (81 patients with comorbid condition) Mean age 44.6 (range 19–82) | iwas established to investigate the severity and chronicity of these neurologic findings over the 5-year period following infection | MoCA, NIH-TB13 (List Sorting Working Memory Test and Picture Vocabulary Test), PHQ-914,GAD-715 | Over time, there was a notable improvement in performance (d = 0.97, p = .03). However, scores remained below the cutoff for detecting cognitive impairment (<26 out of 30). NIH Toolbox cognitive scores, adjusted for demographic factors, showed that scores below the 25th national percentile indicated impairment. 47% of participants who completed the MoCA met the criteria for mild cognitive impairment. 30% of participants demonstrated impairment on vocabulary and 32% demonstrated impairment on working memory |
| Crivelli et al., 2022 | Argentina | Prospective cohort study | 142 days from disease onset | 45 outpatients; 45 controls Median age = 50 years Median age education = 17 years | to investigate the impact of COVID- 19 on cognitive functions in forty-five post COVID patients and forty -five healthy controls. Patients were evaluated for an average of 142 days after illness | MoCA, TMT A & B16,Digit Span Forwards, Digit- Symbol Coding, Craft Story, RAVLT, Benson Figure, Wisconsin Card Sorting Test, Stroop Test, Phonological Fluency, Clock Drawing Test, Multilingual Naming Test, Semantic Fluency, HADS and WHODAS 2.0 Functional Scale | No significant differences were found in the screening measures (MoCA p = .15). Significant differences between groups were found in cognitive composites of memory (p = .016, Cohen’s d = 0.73) attention (p < .001, Cohen’s d = 1.2), executive functions p < .001, Cohen’s d = 1.4), and language (p < .002, Cohen’s d = 87) |
| Damiano et al., 2022 | Brazil | Cohort study (preliminary findings) | 6–9 months after (mean 207 days, SD 20.4) | 425 patients Mean age = 55.7 55.5% participants <12 years of education | to investigate the occurrence of psychiatric and cognitive impairments in a cohort of survivors of moderate or severe forms of COVID-19 | CIS-R17,SCID-5-RV18,HADS,ASQ19,PCL-C20, AUDIT21,MCS22,Temporal and Spatial Orientation23,TMT Part A, VFT24,CFS25,IPAQ-SV26 | The study found deficits on executive and attentional domains. Memory Decline was subjectively reported by 51.1% of the patients. All patients performed worse in TMT- A across all ages. Older age and disorientation were associated with a worse performance on the TMT-A. Older age, higher Frailty Scores (prior COVID) and temporal- spatial disorientation were associated with worse performance in the Verbal Flency Test |
| Delgado-Alonso et al., 2022 | Spain | Cross-sectional study | 9.12 ± 3.46 months after the onset of disease | 50 patients; 50 controls Mean age = 51.06 ± 11.65 Mean years education = 13.58 ± 4.01 | to determine the characteristics of cognitive dysfunction in patients reporting cognitive complaints after COVID-19 and to evaluate the correlation between cognitive function and anxiety, depression, sleep, and olfactory function | Corsi Block-Tapping Test, SDMT27.BNT28,JLO29, ROCF30, FCSRT31,Stroop Color-Word Interference Test, VOSP32,Computerized Neuropsychological Battery Vienna Test System33,BSIT34, STAI35, BDI-II36, PSQI37, Modified Fatigue Impact Scale,3.0 T MRI38 | Cognitive impairment based on performance scores on various neuropsychological assessments compared to normative data -scaled scores (this score is equivalent to a percentile of ≤5 or a z-score of < −1.65, which categorizes a performance as cognitively impaired), normative data which provide representative scores for a cognitively healthy population in Spain and for tests administered via a computerized system, cognitive impairment is assessed by calculating z-scores.COVID-19 group showed worst scores in the recall and recognition trials of the FGT, Inhibition Test, NBV, TMT Part A & B and in several visual tasks of the WAF Battery. Cognitive performance was correlated with olfactory dysfunction and sleep quality and anxiety to a lesser extent, but not depression |
| Del Brutto et al., 2021 | Ecuador | Prospective cohort study | 6 months from disease onset | 52 patients; 41 controls Mean age = 62.6 ± 11 years Comorbid condition | design, the occurrence of cognitive decline among individuals with a history of mild symptomatic SARS-CoV-2 infection was assessed | MoCA | Post- COVID MoCA scores were notably lower than the mean scores before the pandemic in seropositive individuals (21.7 ± 4 vs. 19.6 ± 4.2; p = 0.010), but not in those who tested negative for antibodies (21.5 ± 5 vs. 21 ± 4; p = 0.618 |
| Duindam et al., 2022 | Netherlands | Prospective observational cohort | 6 months after hospital discharge | 96 patients IQR age = 61 (55–69) years 38% patients = low level education 90.6% had comorbid condition | was to study long term cognitive outcomes following severe COVID-19 and the relation to anti-inflammatory therapies | MOCA, TMT Part A & B, Letter Digit Substitution Test, Digit Span, NART-IQ39,, HADS, CFS,BSI40, CFQ41 | After 6.5 ± 1.3 months, 27% of patients exhibited cognitive impairment. There was no correlation between objective cognitive dysfunction and subjective cognitive dysfunction. Patients who received both dexamethasone and tocilizumab showed poorer executive functioning scores (Trail Making Test). Specifically, 21% of patients scored impaired on both TMT-A and TMT-B, while 18% scored impaired on the Digit Span Test |
| Ferrando et al., 2022 | USA | Cross-sectional study | 6–8 months after | 60 patients (clinical group = 32 and nonclinical group = 28) Mean age = 41 years 39% prior psychiatric history On average 1.5 comorbid medical comorbidities | to describe neuropsychological, medical, psychiatric, and functional correlates of cognitive complaints experienced after recovery from acute COVID-19 infection | IADL42,Chalder Fatigue Scale, PHQ-943, QLESQ44, Test Of Premorbid Function, PAOF45,RBANS TMT Part A & B, Verbal Fluency, Stroop Color- Word Test, Posttraumatic Stress Disorder Checklist, Serological samples, | Subjective cognitive function suggests mild to moderate perceived cognitive difficulties related to memory, language, and cognition. The overall sample scored significantly lower than normative values on the RBANS total score, as well as on subtests measuring immediate and delayed memory and language |
| Ferrucci et al., 2021 | Italy | Cross-sectional | 4–5 months after (mean_ SD = 4.43–1.22 months) | 38 patients Years range = 22–74 Mean age education = 12.39 | To study the occurrence of cognitive abnormalities in the months following hospital discharge | MoCA, BRB-NT46, BDI-II, SSD questionnaire47 | Cognitive impairment was present in 60.5% of the sample, indicated by scores falling below cutoffs on at least one task of the BRB-NT. Among them, 42.1% exhibited deficits in processing speed, while 26.3% displayed delayed verbal recall deficits, and 10.5% showed deficits in immediate verbal recall |
| Frontera et al., 2021 | USA | Prospective cohort study | 6 months after | 196 patients with new neurological and 186 patients non neurological | conducted a prospective study of 6-month outcomes of hospitalized COVID-19 patients. Patients with new neurological complications during hospitalization who survived were propensity score-matched to COVID-19 survivors without neurological complications hospitalized during the same period | Barthel Index for activities of daily living, Telephone MoCA, modified Rankin Scale, Neuro-Qol48 Via tele assistance | Half of the individuals exhibited impaired cognition, indicated by a telephone MoCA score below 18 |
| García-Sánchez et al., 2022 | Spain | Prospective cohort study | 187 days after | 63 patients (33 previous hospitalized) Mean age (SD) =51.5 (12.5) Mean level education = 14.4 years | explore the frequency of impaired performance across cognitive domains in post-COVID patients with subjective complaints and uncover whether impairment existed within a single domain or across multiple. | MoCA, CPT-II49,RAVLT, ROCFT50, Digit Span Forward and Backward, BNT, Block Design Test, Symbol Search, TMT Part A & B, Stroop Task, Phonetic and Semantic Fluency Tasks, 15-OT51, HADS | In patients with single-domain impairment, attention deficits were the most common, observed in 19.0% of cases, significantly surpassing deficits in executive functioning (p = 0.01), short-term/working memory (p = 0.001), and language (p < 0.001). Additionally, among patients with multiple-domain impairment, attention was the cognitive domain most frequently affected, particularly in combination with deficits in learning, long-term memory, and executive functioning |
| Gautam et al., 2022 | United Kingdom | Retrospective cases series | Mean days after COVID onset (SD) = 143.4 (42.4) | 200 Discharged patients. Mean age(SD) = 56.5(13.2) Half of patients with comorbid condition | assessed the medium- term effects of COVID-19 on survivors of severe disease | MOCA | Some degree of cognitive impairment, particularly affecting concentration, and short-term recall, was observed in 12.5% of patients |
| Graham et al., 2021 | USA | Prospective cohort study | On average at 4.72 months + group compared to 5.82 month | 100 ambulatory patients; 36 with cognitive evaluation | to characterize the spectrum of neurologic manifestations in non-hospitalized Covid-19 “long haulers | PROMIS52,NIH Toolbox53 48 patients through tele visits and 52 in-person | The cognitive performance of COVID patients on tasks related to attention and working memory was inferior when compared to a demographically similar population from the USA |
| Hampshire et al., 2022 | United Kingdom | Cohort Study | On average 179 days after illness onset (SD) = 62 | 46 patients Mean age = 51 years Age range = 28–33 | to study a possible association between severe COVID-19 and persistent cognitive deficits | Custom Computerized cognitive assessment Battery, GAD-754,PHQ-9, PCL-555 | Deficits exhibited a notable effect size among individuals who had undergone hospitalization |
| Hartung et al., 2022 | Germany | Prospective multicenter study | ≥6 months after | 969 patients; matched control group | To investigate frequency of post-COVID fatigue and cognitive impairment | MoCA, FACIT Fatigue Scale, PHQ-8, GAD-7, PSQI | 26% of patients experienced mild cognitive impairment, while 1% exhibited moderate impairment. Older age, male gender, shorter education, and a history of neuropsychiatric disease were associated with cognitive impairment. There was no significant correlation found between fatigue and cognitive impairment |
| He et al., 2023 | China | Observational cross-sectional study | At 15 months after discharge | 66 patients-majority mild (median age = 35.50) and 79 healthy controls Range age total = 17–61 years | To investigate cognitive function and neuropsychiatric symptoms of 66 COVID-19 survivors and 79 healthy controls (HCs) using a digital, gamified cognitive function evaluation tool and questionnaires at 15 months after discharge. Venous blood samples were collected to measure cytokine levels | THINC-it tool56 | Criteria for determining cognitive impairment from the cognitive tests involve comparisons to a healthy control group matched by sex, age, and education level. COVID-19 survivors showed long-term cognitive impairment in executive function even at 15 months after illness onset |
| Krishnan et al., 2022 | USA | Observational Study | 5.5 after infection On average 168 days after illness onset | Total 20 patients—average education of 15 years 85% of patients with previous mental- medical history | to document cognitive function among individuals recovering from COVID-19 and to identify factors that were associated with outcomes, including preexisting health conditions and severity of COVID-19 illness | WMS-IV57,Logical memory, RAVLT, Brief Visuospatial Memory Test—Revised, Reading subtest from the wide Range Achievement Test-IV, BNT- Second Edition, Lexical and semantic verbal fluencies, Line orientation, Digit Span, Matrix Reasoning, Similarities subtests58,DKEFS59,TMT, Wisconsin Card Sorting Test, Conners Continuous Performance Test-3, SDMT, Coding and Symbol Search60 | Defines cognitive impairment in post-COVID-19 patients as performance on neuropsychological tests that falls 1.5 SD or more below the mean based on published normative data. Mild cognitive deficits were seen on test involving attention and processing or executive function. |
| Lamontagne et al., 2021 | Canada and USA | Observational study | Mean (SD) 123.63 (94.71) days after infection | 50 patients (3 subgroups; acute Covid-19,PASC and post PASC) and 50 healthy control Patients age range = 20–53 years Mean age education (SD) = 16.2(2.95) | investigated mood and cognitive functioning in individuals who have recovered from COVID-19 infection | Self- reported measures of stress, depression and anhedonia, Attention Network Test and cognitive abilities (Attentional Control Scale) | The criteria for identifying impairment appear to rely on substantial statistical distinctions in task performance between the COVID-19 group and controls, rather than employing a set numerical threshold. The most notable effects were observed in individuals diagnosed 1–4 months before evaluation. The COVID-19 recovery group exhibited significantly higher scores on perceived stress. Within the COVID-19 group, there was a specific deficit in attention, characterized by impaired executive functioning while alertness and orienting abilities remained unaffected |
| Lauria et al., 2022 | Italy | Single-center study | On average 96.5 days after the onset of COVID-19 symptoms | 100 patients Mean age(SD) = 73.4(6.2) Mean years education (SD) = 12.7(8.7) | To investigate the neurologic and cognitive features of a sample of elderly patients with confirmed diagnosis of COVID-19 evaluated in the post-acute phase through a direct neuropsychological evaluation | MMSE, FAB61,RAVLT, Multiple Features Target Cancellation Test, TMT, Digit Span forward and backward | For the MMSE, the adjusted score averaged 28.2 ± 1.7, as anticipated for a study population comprising relatively educated individuals with no history of cognitive impairment. For the FAB, 20% of patients exhibited either pathologic or borderline performances. On the Rey Auditory Verbal Learning Test, 11% of patients showed either pathologic or borderline performances on immediate recall, and 17% on delayed recall. In the Multiple Features Target Cancellation Test, 4% of patients demonstrated either pathologic or borderline performances. Regarding the Trail Making Test, 33% of patients displayed either pathologic or borderline performances. Lastly, in the Digit Span forward and backward tests, 12% and 23% of patients, respectively, exhibited either pathologic or borderline performances |
| Matias-Guiu et al., 2023 | Spain | Mean 484.60 ± 174.46 days after the acute infection | 404 Patients; mean age = 48.6 ± 9.2 145 controls Clinical comorbidities in the post COVID group | To develop objective criteria for cognitive dysfunction associated with the post-COVID syndrome | SDMT, Stroop Test, FCRST62,ROCF63, Digit Span Backwards, VOSP64, BNT, Semantic Fluency, WAIS-IV 65Matrix reasoning, JLO66,TAVEC67,Brief Test of Attention, Object Naming | Using cutoff points of −1 and − 1.5 SD below the normative mean for the cognitive tests administered. Processing speed and attention were the cognitive domains most affected, with 41.2% and 17.3% of patients classified as having impairment in at least one cognitive domain. Younger age and lower levels of education were linked to cognitive impairment, whereas hospitalization did not show any association | |
| Miskowiak et al., 2021 | Denmark | Prospective cohort study | 3–4 months after | 29 Discharged patients; matched 100 healthy controls. Mean age post-COVID = 56.2 (10.6) Mean age education = 14.3(3.9) Charlson comorbidity score = 2.9(2.4) | To investigate the frequency, pattern and severity of cognitive impairments after COVID-19 hospital discharge, their relation to subjective cognitive complaints, quality of life and illness variables | SCIP-D68,TMT Part B | Using a 0.5 standard deviation cut-off for SCIP total scores, a total of 19 patients, constituting 65% of the sample, were categorized as cognitively impaired |
| Miskowiak et al., 2023 | Denmark | Prospective cohort study | an average of 7 (±4) months | 194 patients; mean age = 50.8(15.4); mean age education = 15.3(3.9) 150 healthy controls Charlson comorbidity score = 34 (19) | To investigate the frequency, pattern, and functional implications of cognitive impairments in patients at a long-COVID clinic who were referred after hospitalization with COVID-19 | SCIP-D, TMT Part B | SCIP total score: statistically significant lower performance in patients than healthy controls (p < .001). Verbal learning and memory: statistically significant lower performance in patients than healthy controls (p = .001 and p = .005, respectively). Working memory: statistically significant lower performance in patients than healthy controls (p < .001). Verbal fluency: statistically significant lower performance in patients than healthy controls (p = .004). Processing speed: no statistically significant differences between patients and healthy controls. MT-B: statistically significant lower performance in patients according to expected scores based on age, sex, and education |
| Ortelli et al., 2021 | Italy | Cross-sectional | 9–13 weeks after | 12 Discharged patients; age 67 ± 9.6 12 healthy controls;age 64.3 ± 10.5 Total patients develop neurological complications after COVID | to provide a comprehensive clinical, neurophysiological, and neuropsychological profile of fatigued patients suffering from neurological manifestations related to SARS-CoV-2, who recovered from the acute phase of COVID-19 | MoCA, FAB, three computerized attentive tasks69 | MoCA & FAB: statistically significant lower scores in patients compared to healthy controls (p < .001) Computerized tasks: Reaction times were significantly longer in patients than healthy controls in both SIT (p = .015) and NT (p = .046). Percentage of errors was significantly larger in patients than healthy controls in all three tasks (p < .001) |
| Pihlaja et al., 2023 | Finland | Cohort study | At 3 and 6 months after infection | 184 patients (mean age(SD) =53.4 (12.8) Reported comorbidities. Non-COVID control group = 53 individuals (mean age(SD) = 54.9(12.3) | To describe the prevalence of subjective and objective cognitive dysfunction at 3 and 6 months after COVID-19 and the associations of subjective cognitive symptoms and psychological and disease-related factors | MoCA | At 6 months, there was a statistically significant difference between ICU, WARD, and HOME groups (p < .001). However, there was no statistically significant difference between all COVID-19 patients and control group |
| Pilotto et al., 2021 | Italy | Prospective cohort study | 6 months after | 105 Discharged patients (no premorbid neurological conditions) Three subgroups (mild, moderate, severe) Average age = 64.8 (±12.6) | to evaluate prevalence and predictors of long-term neurological manifestations. | MoCA | 17.5% of participants had scores indicative of cognitive impairment |
| Rass et al., 2021 | Austria | Multicenter observational cohort study | 3 months after COVID-19 (102 days [IQR] = [91–110] days) | 135 patients Median age 56 (range = 19–87) 33 patients pre-diagnosed neurological disorders before COVID | to assess neurological manifestations and health-related quality of life (QoL) | MoCA | Cognitive impairment, as measured by the MoCA, were identified in 23% of the patient population. Among those with severe COVID-19, the prevalence was 29%, while it was 30% in moderate cases and 3% in mild cases |
| Rubega et al., 2022 | Italy | Observational study | 12 months after discharge | 33 total patients (2 subgroups) n = 16 ICU patients age interval = 50–80 n = 17 non-ICU patients age interval = 53–78 12 healthy controls matched | to study long-term cognitive and affective sequelae and sleep high-density electroencephalography (EEG) in people with a previous hospital admission for acute COVID-19. | MoCA, FAB, Stroop task, Digit Span forward and backward, RAVLT, TMT PART A & B, SDMT | Only on the TMT-B there was a trend towards worse performance in COVID-19 patients, specifically in non-ICU patients (CTRL vs. non-ICU: p = .04). On the remaining tests, no statistically significant differences between the groups were found |
| Vannorsdall et al., 2022 | USA | Prospective cohort study | Mean time diagnosis (SD) = 126.5 (70.1) | 82 Discharged patients (n = 48 post -ICU; n = 34 post non ICU) 95% participants = ≤12 years education Mean age 54.5 years | to prospectively characterize cognition, mental health symptoms, and functioning approximately 4 months after an initial diagnosis of COVID-19 in a racially and ethnically diverse group of patients. | RAVLT, Oral Trail Making Test70,Digit Span forward and backward, Semantic and phonemic fluency via tele assistance | RAVLT acquisition and delayed recall: 26.8% of patients were mildly/moderately impaired and 14.6% were severely impaired. TMT-A: 35.4% of patients were mildly/moderately impaired and 25.6% were severely impaired. TMT-B: 15.9% of patients were mildly/moderately impaired and 13.4% were severely impaired. Digit Span forward: 1.2% of patients were mildly/moderately impaired and 1.2% were severely impaired. Digit Span backward: 13.4% of patients were mildly/moderately impaired and 4.9% were severely impaired. Phonemic fluency: 25.6% of patients were mildly/moderately impaired and 8.5% were severely impaired. Semantic fluency: 31.7% of patients were mildly/moderately impaired and 9.8% were severely impaired |
| Walle-Hansen et al., 2021 | Norway | Prospective cohort study | 6 months after | 106 Discharged patients (subdivided by age into 2 subgroups) (ambulatory evaluation) Mean age = 74.3 (range 60–96) Comorbid condition | To study age-related change in health-related quality of life, functional decline and mortality among older patients following hospitalization due to COVID-19 | MoCA | A negative change in cognitive functions was reported by 43% of patients 6 months after hospitalization. MoCA scores were lower in the oldest age group (mean: 21.7) compared to younger participants (mean: 25.3) |
| Woo et al., 2020 | Germany | Cross-sectional study | Median: 85 days from COVID-19 recovery [range: 20–105 days] | 18 patients (mean age [SD] = 42.2 [14.3]) 10 healthy controls (mean age [SD] = 38.4 [14.4]) | Study focused on neurocognitive deficits in mild COVID-19 young patients | TICS-M (by phone or directly) | The performance of COVID-19 patients was significantly lower (mean: 38.83; range: 31–46) than the healthy controls (mean: 45.8; range: 43–50), especially on short-term memory, attention, and concentration/language tasks |
| Author/year | Country | Study design | Time cognitive assessment from onset infection COVID-19 | Participants | Aim of study | Cognitive test | Results |
|---|---|---|---|---|---|---|---|
| Abramoff et al., 2023 | USA | Observational cohort study | >12–30+ weeks | Total 324 outpatients 74 patients were hospitalized, mean age (SD) =46.6(14.0) Education age = 37 patients (11.9%) has <12 formal education | To examine the factors associated with variation in their psychological and cognitive functioning and health-related quality of life | MoCA1, Hospital anxiety and depression scale, European Quality of Life Scale instrument | Scores below 18 on the education-adjusted MoCA scale, serve as the threshold for classifying patients with cognitive impairment. Alongside reporting group means, the analysis also includes the distribution of scores falling within low or very low ranges, accompanied by effect sizes and odds ratios. Among survivors seeking care for persistent COVID symptoms, 17% exhibited cognitive impairment. Individuals from poor and African American backgrounds experienced more severe psychological and cognitive effects following infection |
| Albu et al., 2021 | Spain | Cross-sectional | 89–124 days after | discharged 30 patients (16 post-ICU patients and 14 non-ICU patients | describes sequelae and persistent symptoms at more than 3 months after acute COVID-19 and their impact on activities of daily living, sleep, and quality of life in patients in an outpatient rehabilitation program | Barcelona Test (orientation), Digit Span forward subtest of the Wechsler Adult Intelligence Scale III (immediate attention), RAVLT2 (verbal memory), Digit Span backward subtest of the Wechsler Adult Intelligence Scale III (working memory), PMR task (executive function), HADS3 | Low scores were observed across various cognitive domains, including orientation, attention, verbal learning, long-term verbal memory, verbal recognition, working memory, and executive control. However, none of these differences were statistically significant, as indicated by X2 values ranging from 0.18 to 1.77 and p-values ranging from .18 to .89. Cognitive impairment was detected in 63.3% of patients, with a comparable profile observed in both subgroups |
| Amalakanti et al., 2021 | India | Case control study | Outpatients Not clear | 93 asymptomatic patients (mean age 36.2 ± 11.7); 102 controls (mean age 35.6 ± 9.8) | to detect MCI in asymptomatic patients subjects COVID-19 with MOCA | MoCA | For individuals aged 18–19, the average MoCA score in cases was 25.9 with a standard deviation of 2.1, compared to 27 with a standard deviation of 1.7 in controls. Among those aged 30–49, controls had an average MoCA score of 25.6 with a standard deviation of 4.3. For individuals aged 50 and above, cases had an average MoCA score of 24 with a standard deviation of 3.5, while controls had an average score of 24.5 with a standard deviation of 3.5 |
| Becker et al., 2021 | USA | Cross-sectional | 7.6 months after | Ambulatory or Discharged Total = 740 mean age = 49 (38–59) years 464 patients were women 516 individuals had other comorbidities (103) patients have educational level ≤ 12 years | to investigate rates of cognitive impairment in survivors of COVID-19 who were treated in outpatient, emergency department (ED), or inpatient hospital setting | Number Span Forward and Backward (attention and working memory), TMT A & B (processing speed and e.f4), phonemic and category Fluency (language) and the Hopkins Verbal Learning Test-revised (memory encoding, recall, and recognition) | Hospitalized patients demonstrated higher odds ratios of impairments in attention (OR: 2.8; 95% CI: 1.3–5.9), executive functioning (OR: 1.8; 95% CI: 1.0–3.4), category fluency (OR: 5.2; 95% CI: 3.0–9.0), memory encoding (OR: 2.3; 95% CI: 1.3–4.1), and memory recall (OR: 2.2; 95% CI: 1.3–3.8) compared to the outpatient group. Emergency department (ED) patients exhibited greater impairment in category fluency (OR: 1.8; 95% CI: 1.1–3.1) and memory encoding (OR: 1.7; 95% CI: 1.0–3.0) compared to outpatients |
| Birberg Thornberg et al., 2022 | Sweden | an ambidirectional population-based cohort study | 5 months after discharge | total 133 patients | to explore objective neurocognitive deficits in COVID-19 patients five months after discharge, and any associations with demographic factors and disease severity indicators | RBANS5, CWIT6, D-KEFS7, HADS, MFI8 | In the RBANS global cognition index, 45 patients (36.9%) scored below 1.5 SD, and 28 (23.0%) scored below 2 SD. The most prevalent deficits were observed in the Coding subtest, assessing psychomotor speed, with 47 patients (36.2%) affected, followed by the Semantic fluency subtest, evaluating verbal ability, with 34 patients (25.8%), and the Digit span subtest, assessing short-term memory and auditory attention, with 31 patients (23.7%). Additionally, performance on the CWIT was below the cutoff for 10.6% of the patients |
| Boesl et al., 2021 | Germany | Retrospective cohort study | Meantime 184.5 days after | 100 ambulatory patients Range age (20–79) Mean age 45.8 | assessed clinical data of their first 100 ambulatory patients regarding clinical presentations and a screening assessment for detecting cognitive impairment ≥12 weeks after the acute infection with SARS-CoV-2. | MoCA, BDI9, FSS10,ESS11 | A total of 30% of the patients with cognitive deficits showed pathologic MoCA scores (≤25/30 points) |
| Brown et al., 2022 | USA | Observational cohort study | Mean 22.4 weeks (SD 12.8) Range 1–51 | total 297 patients (66 hospitalized; 20 ICU) Middle age 46.9 9(SD) = 14.1 | studied specific associations between cognitive impairment, self-reported disruptions in memory, psychiatric symptoms (depression, anxiety and pos-ttraumatic depression) and sleep disorders | HADS, IES-612(post-traumatic stress reactions), MoCA -Blind, Self-reported memory disturbance | Depression symptom severity was significantly associated with MoCA total score. MoCA Digit Span-Backward, MoCA Fluency, MoCA Delay recall were all significantly associated with depression |
| Chen et al., 2022 | Georgia | Prospective Cohort Study (primarily findings) | on average 125 days after | 200 patients (81 patients with comorbid condition) Mean age 44.6 (range 19–82) | iwas established to investigate the severity and chronicity of these neurologic findings over the 5-year period following infection | MoCA, NIH-TB13 (List Sorting Working Memory Test and Picture Vocabulary Test), PHQ-914,GAD-715 | Over time, there was a notable improvement in performance (d = 0.97, p = .03). However, scores remained below the cutoff for detecting cognitive impairment (<26 out of 30). NIH Toolbox cognitive scores, adjusted for demographic factors, showed that scores below the 25th national percentile indicated impairment. 47% of participants who completed the MoCA met the criteria for mild cognitive impairment. 30% of participants demonstrated impairment on vocabulary and 32% demonstrated impairment on working memory |
| Crivelli et al., 2022 | Argentina | Prospective cohort study | 142 days from disease onset | 45 outpatients; 45 controls Median age = 50 years Median age education = 17 years | to investigate the impact of COVID- 19 on cognitive functions in forty-five post COVID patients and forty -five healthy controls. Patients were evaluated for an average of 142 days after illness | MoCA, TMT A & B16,Digit Span Forwards, Digit- Symbol Coding, Craft Story, RAVLT, Benson Figure, Wisconsin Card Sorting Test, Stroop Test, Phonological Fluency, Clock Drawing Test, Multilingual Naming Test, Semantic Fluency, HADS and WHODAS 2.0 Functional Scale | No significant differences were found in the screening measures (MoCA p = .15). Significant differences between groups were found in cognitive composites of memory (p = .016, Cohen’s d = 0.73) attention (p < .001, Cohen’s d = 1.2), executive functions p < .001, Cohen’s d = 1.4), and language (p < .002, Cohen’s d = 87) |
| Damiano et al., 2022 | Brazil | Cohort study (preliminary findings) | 6–9 months after (mean 207 days, SD 20.4) | 425 patients Mean age = 55.7 55.5% participants <12 years of education | to investigate the occurrence of psychiatric and cognitive impairments in a cohort of survivors of moderate or severe forms of COVID-19 | CIS-R17,SCID-5-RV18,HADS,ASQ19,PCL-C20, AUDIT21,MCS22,Temporal and Spatial Orientation23,TMT Part A, VFT24,CFS25,IPAQ-SV26 | The study found deficits on executive and attentional domains. Memory Decline was subjectively reported by 51.1% of the patients. All patients performed worse in TMT- A across all ages. Older age and disorientation were associated with a worse performance on the TMT-A. Older age, higher Frailty Scores (prior COVID) and temporal- spatial disorientation were associated with worse performance in the Verbal Flency Test |
| Delgado-Alonso et al., 2022 | Spain | Cross-sectional study | 9.12 ± 3.46 months after the onset of disease | 50 patients; 50 controls Mean age = 51.06 ± 11.65 Mean years education = 13.58 ± 4.01 | to determine the characteristics of cognitive dysfunction in patients reporting cognitive complaints after COVID-19 and to evaluate the correlation between cognitive function and anxiety, depression, sleep, and olfactory function | Corsi Block-Tapping Test, SDMT27.BNT28,JLO29, ROCF30, FCSRT31,Stroop Color-Word Interference Test, VOSP32,Computerized Neuropsychological Battery Vienna Test System33,BSIT34, STAI35, BDI-II36, PSQI37, Modified Fatigue Impact Scale,3.0 T MRI38 | Cognitive impairment based on performance scores on various neuropsychological assessments compared to normative data -scaled scores (this score is equivalent to a percentile of ≤5 or a z-score of < −1.65, which categorizes a performance as cognitively impaired), normative data which provide representative scores for a cognitively healthy population in Spain and for tests administered via a computerized system, cognitive impairment is assessed by calculating z-scores.COVID-19 group showed worst scores in the recall and recognition trials of the FGT, Inhibition Test, NBV, TMT Part A & B and in several visual tasks of the WAF Battery. Cognitive performance was correlated with olfactory dysfunction and sleep quality and anxiety to a lesser extent, but not depression |
| Del Brutto et al., 2021 | Ecuador | Prospective cohort study | 6 months from disease onset | 52 patients; 41 controls Mean age = 62.6 ± 11 years Comorbid condition | design, the occurrence of cognitive decline among individuals with a history of mild symptomatic SARS-CoV-2 infection was assessed | MoCA | Post- COVID MoCA scores were notably lower than the mean scores before the pandemic in seropositive individuals (21.7 ± 4 vs. 19.6 ± 4.2; p = 0.010), but not in those who tested negative for antibodies (21.5 ± 5 vs. 21 ± 4; p = 0.618 |
| Duindam et al., 2022 | Netherlands | Prospective observational cohort | 6 months after hospital discharge | 96 patients IQR age = 61 (55–69) years 38% patients = low level education 90.6% had comorbid condition | was to study long term cognitive outcomes following severe COVID-19 and the relation to anti-inflammatory therapies | MOCA, TMT Part A & B, Letter Digit Substitution Test, Digit Span, NART-IQ39,, HADS, CFS,BSI40, CFQ41 | After 6.5 ± 1.3 months, 27% of patients exhibited cognitive impairment. There was no correlation between objective cognitive dysfunction and subjective cognitive dysfunction. Patients who received both dexamethasone and tocilizumab showed poorer executive functioning scores (Trail Making Test). Specifically, 21% of patients scored impaired on both TMT-A and TMT-B, while 18% scored impaired on the Digit Span Test |
| Ferrando et al., 2022 | USA | Cross-sectional study | 6–8 months after | 60 patients (clinical group = 32 and nonclinical group = 28) Mean age = 41 years 39% prior psychiatric history On average 1.5 comorbid medical comorbidities | to describe neuropsychological, medical, psychiatric, and functional correlates of cognitive complaints experienced after recovery from acute COVID-19 infection | IADL42,Chalder Fatigue Scale, PHQ-943, QLESQ44, Test Of Premorbid Function, PAOF45,RBANS TMT Part A & B, Verbal Fluency, Stroop Color- Word Test, Posttraumatic Stress Disorder Checklist, Serological samples, | Subjective cognitive function suggests mild to moderate perceived cognitive difficulties related to memory, language, and cognition. The overall sample scored significantly lower than normative values on the RBANS total score, as well as on subtests measuring immediate and delayed memory and language |
| Ferrucci et al., 2021 | Italy | Cross-sectional | 4–5 months after (mean_ SD = 4.43–1.22 months) | 38 patients Years range = 22–74 Mean age education = 12.39 | To study the occurrence of cognitive abnormalities in the months following hospital discharge | MoCA, BRB-NT46, BDI-II, SSD questionnaire47 | Cognitive impairment was present in 60.5% of the sample, indicated by scores falling below cutoffs on at least one task of the BRB-NT. Among them, 42.1% exhibited deficits in processing speed, while 26.3% displayed delayed verbal recall deficits, and 10.5% showed deficits in immediate verbal recall |
| Frontera et al., 2021 | USA | Prospective cohort study | 6 months after | 196 patients with new neurological and 186 patients non neurological | conducted a prospective study of 6-month outcomes of hospitalized COVID-19 patients. Patients with new neurological complications during hospitalization who survived were propensity score-matched to COVID-19 survivors without neurological complications hospitalized during the same period | Barthel Index for activities of daily living, Telephone MoCA, modified Rankin Scale, Neuro-Qol48 Via tele assistance | Half of the individuals exhibited impaired cognition, indicated by a telephone MoCA score below 18 |
| García-Sánchez et al., 2022 | Spain | Prospective cohort study | 187 days after | 63 patients (33 previous hospitalized) Mean age (SD) =51.5 (12.5) Mean level education = 14.4 years | explore the frequency of impaired performance across cognitive domains in post-COVID patients with subjective complaints and uncover whether impairment existed within a single domain or across multiple. | MoCA, CPT-II49,RAVLT, ROCFT50, Digit Span Forward and Backward, BNT, Block Design Test, Symbol Search, TMT Part A & B, Stroop Task, Phonetic and Semantic Fluency Tasks, 15-OT51, HADS | In patients with single-domain impairment, attention deficits were the most common, observed in 19.0% of cases, significantly surpassing deficits in executive functioning (p = 0.01), short-term/working memory (p = 0.001), and language (p < 0.001). Additionally, among patients with multiple-domain impairment, attention was the cognitive domain most frequently affected, particularly in combination with deficits in learning, long-term memory, and executive functioning |
| Gautam et al., 2022 | United Kingdom | Retrospective cases series | Mean days after COVID onset (SD) = 143.4 (42.4) | 200 Discharged patients. Mean age(SD) = 56.5(13.2) Half of patients with comorbid condition | assessed the medium- term effects of COVID-19 on survivors of severe disease | MOCA | Some degree of cognitive impairment, particularly affecting concentration, and short-term recall, was observed in 12.5% of patients |
| Graham et al., 2021 | USA | Prospective cohort study | On average at 4.72 months + group compared to 5.82 month | 100 ambulatory patients; 36 with cognitive evaluation | to characterize the spectrum of neurologic manifestations in non-hospitalized Covid-19 “long haulers | PROMIS52,NIH Toolbox53 48 patients through tele visits and 52 in-person | The cognitive performance of COVID patients on tasks related to attention and working memory was inferior when compared to a demographically similar population from the USA |
| Hampshire et al., 2022 | United Kingdom | Cohort Study | On average 179 days after illness onset (SD) = 62 | 46 patients Mean age = 51 years Age range = 28–33 | to study a possible association between severe COVID-19 and persistent cognitive deficits | Custom Computerized cognitive assessment Battery, GAD-754,PHQ-9, PCL-555 | Deficits exhibited a notable effect size among individuals who had undergone hospitalization |
| Hartung et al., 2022 | Germany | Prospective multicenter study | ≥6 months after | 969 patients; matched control group | To investigate frequency of post-COVID fatigue and cognitive impairment | MoCA, FACIT Fatigue Scale, PHQ-8, GAD-7, PSQI | 26% of patients experienced mild cognitive impairment, while 1% exhibited moderate impairment. Older age, male gender, shorter education, and a history of neuropsychiatric disease were associated with cognitive impairment. There was no significant correlation found between fatigue and cognitive impairment |
| He et al., 2023 | China | Observational cross-sectional study | At 15 months after discharge | 66 patients-majority mild (median age = 35.50) and 79 healthy controls Range age total = 17–61 years | To investigate cognitive function and neuropsychiatric symptoms of 66 COVID-19 survivors and 79 healthy controls (HCs) using a digital, gamified cognitive function evaluation tool and questionnaires at 15 months after discharge. Venous blood samples were collected to measure cytokine levels | THINC-it tool56 | Criteria for determining cognitive impairment from the cognitive tests involve comparisons to a healthy control group matched by sex, age, and education level. COVID-19 survivors showed long-term cognitive impairment in executive function even at 15 months after illness onset |
| Krishnan et al., 2022 | USA | Observational Study | 5.5 after infection On average 168 days after illness onset | Total 20 patients—average education of 15 years 85% of patients with previous mental- medical history | to document cognitive function among individuals recovering from COVID-19 and to identify factors that were associated with outcomes, including preexisting health conditions and severity of COVID-19 illness | WMS-IV57,Logical memory, RAVLT, Brief Visuospatial Memory Test—Revised, Reading subtest from the wide Range Achievement Test-IV, BNT- Second Edition, Lexical and semantic verbal fluencies, Line orientation, Digit Span, Matrix Reasoning, Similarities subtests58,DKEFS59,TMT, Wisconsin Card Sorting Test, Conners Continuous Performance Test-3, SDMT, Coding and Symbol Search60 | Defines cognitive impairment in post-COVID-19 patients as performance on neuropsychological tests that falls 1.5 SD or more below the mean based on published normative data. Mild cognitive deficits were seen on test involving attention and processing or executive function. |
| Lamontagne et al., 2021 | Canada and USA | Observational study | Mean (SD) 123.63 (94.71) days after infection | 50 patients (3 subgroups; acute Covid-19,PASC and post PASC) and 50 healthy control Patients age range = 20–53 years Mean age education (SD) = 16.2(2.95) | investigated mood and cognitive functioning in individuals who have recovered from COVID-19 infection | Self- reported measures of stress, depression and anhedonia, Attention Network Test and cognitive abilities (Attentional Control Scale) | The criteria for identifying impairment appear to rely on substantial statistical distinctions in task performance between the COVID-19 group and controls, rather than employing a set numerical threshold. The most notable effects were observed in individuals diagnosed 1–4 months before evaluation. The COVID-19 recovery group exhibited significantly higher scores on perceived stress. Within the COVID-19 group, there was a specific deficit in attention, characterized by impaired executive functioning while alertness and orienting abilities remained unaffected |
| Lauria et al., 2022 | Italy | Single-center study | On average 96.5 days after the onset of COVID-19 symptoms | 100 patients Mean age(SD) = 73.4(6.2) Mean years education (SD) = 12.7(8.7) | To investigate the neurologic and cognitive features of a sample of elderly patients with confirmed diagnosis of COVID-19 evaluated in the post-acute phase through a direct neuropsychological evaluation | MMSE, FAB61,RAVLT, Multiple Features Target Cancellation Test, TMT, Digit Span forward and backward | For the MMSE, the adjusted score averaged 28.2 ± 1.7, as anticipated for a study population comprising relatively educated individuals with no history of cognitive impairment. For the FAB, 20% of patients exhibited either pathologic or borderline performances. On the Rey Auditory Verbal Learning Test, 11% of patients showed either pathologic or borderline performances on immediate recall, and 17% on delayed recall. In the Multiple Features Target Cancellation Test, 4% of patients demonstrated either pathologic or borderline performances. Regarding the Trail Making Test, 33% of patients displayed either pathologic or borderline performances. Lastly, in the Digit Span forward and backward tests, 12% and 23% of patients, respectively, exhibited either pathologic or borderline performances |
| Matias-Guiu et al., 2023 | Spain | Mean 484.60 ± 174.46 days after the acute infection | 404 Patients; mean age = 48.6 ± 9.2 145 controls Clinical comorbidities in the post COVID group | To develop objective criteria for cognitive dysfunction associated with the post-COVID syndrome | SDMT, Stroop Test, FCRST62,ROCF63, Digit Span Backwards, VOSP64, BNT, Semantic Fluency, WAIS-IV 65Matrix reasoning, JLO66,TAVEC67,Brief Test of Attention, Object Naming | Using cutoff points of −1 and − 1.5 SD below the normative mean for the cognitive tests administered. Processing speed and attention were the cognitive domains most affected, with 41.2% and 17.3% of patients classified as having impairment in at least one cognitive domain. Younger age and lower levels of education were linked to cognitive impairment, whereas hospitalization did not show any association | |
| Miskowiak et al., 2021 | Denmark | Prospective cohort study | 3–4 months after | 29 Discharged patients; matched 100 healthy controls. Mean age post-COVID = 56.2 (10.6) Mean age education = 14.3(3.9) Charlson comorbidity score = 2.9(2.4) | To investigate the frequency, pattern and severity of cognitive impairments after COVID-19 hospital discharge, their relation to subjective cognitive complaints, quality of life and illness variables | SCIP-D68,TMT Part B | Using a 0.5 standard deviation cut-off for SCIP total scores, a total of 19 patients, constituting 65% of the sample, were categorized as cognitively impaired |
| Miskowiak et al., 2023 | Denmark | Prospective cohort study | an average of 7 (±4) months | 194 patients; mean age = 50.8(15.4); mean age education = 15.3(3.9) 150 healthy controls Charlson comorbidity score = 34 (19) | To investigate the frequency, pattern, and functional implications of cognitive impairments in patients at a long-COVID clinic who were referred after hospitalization with COVID-19 | SCIP-D, TMT Part B | SCIP total score: statistically significant lower performance in patients than healthy controls (p < .001). Verbal learning and memory: statistically significant lower performance in patients than healthy controls (p = .001 and p = .005, respectively). Working memory: statistically significant lower performance in patients than healthy controls (p < .001). Verbal fluency: statistically significant lower performance in patients than healthy controls (p = .004). Processing speed: no statistically significant differences between patients and healthy controls. MT-B: statistically significant lower performance in patients according to expected scores based on age, sex, and education |
| Ortelli et al., 2021 | Italy | Cross-sectional | 9–13 weeks after | 12 Discharged patients; age 67 ± 9.6 12 healthy controls;age 64.3 ± 10.5 Total patients develop neurological complications after COVID | to provide a comprehensive clinical, neurophysiological, and neuropsychological profile of fatigued patients suffering from neurological manifestations related to SARS-CoV-2, who recovered from the acute phase of COVID-19 | MoCA, FAB, three computerized attentive tasks69 | MoCA & FAB: statistically significant lower scores in patients compared to healthy controls (p < .001) Computerized tasks: Reaction times were significantly longer in patients than healthy controls in both SIT (p = .015) and NT (p = .046). Percentage of errors was significantly larger in patients than healthy controls in all three tasks (p < .001) |
| Pihlaja et al., 2023 | Finland | Cohort study | At 3 and 6 months after infection | 184 patients (mean age(SD) =53.4 (12.8) Reported comorbidities. Non-COVID control group = 53 individuals (mean age(SD) = 54.9(12.3) | To describe the prevalence of subjective and objective cognitive dysfunction at 3 and 6 months after COVID-19 and the associations of subjective cognitive symptoms and psychological and disease-related factors | MoCA | At 6 months, there was a statistically significant difference between ICU, WARD, and HOME groups (p < .001). However, there was no statistically significant difference between all COVID-19 patients and control group |
| Pilotto et al., 2021 | Italy | Prospective cohort study | 6 months after | 105 Discharged patients (no premorbid neurological conditions) Three subgroups (mild, moderate, severe) Average age = 64.8 (±12.6) | to evaluate prevalence and predictors of long-term neurological manifestations. | MoCA | 17.5% of participants had scores indicative of cognitive impairment |
| Rass et al., 2021 | Austria | Multicenter observational cohort study | 3 months after COVID-19 (102 days [IQR] = [91–110] days) | 135 patients Median age 56 (range = 19–87) 33 patients pre-diagnosed neurological disorders before COVID | to assess neurological manifestations and health-related quality of life (QoL) | MoCA | Cognitive impairment, as measured by the MoCA, were identified in 23% of the patient population. Among those with severe COVID-19, the prevalence was 29%, while it was 30% in moderate cases and 3% in mild cases |
| Rubega et al., 2022 | Italy | Observational study | 12 months after discharge | 33 total patients (2 subgroups) n = 16 ICU patients age interval = 50–80 n = 17 non-ICU patients age interval = 53–78 12 healthy controls matched | to study long-term cognitive and affective sequelae and sleep high-density electroencephalography (EEG) in people with a previous hospital admission for acute COVID-19. | MoCA, FAB, Stroop task, Digit Span forward and backward, RAVLT, TMT PART A & B, SDMT | Only on the TMT-B there was a trend towards worse performance in COVID-19 patients, specifically in non-ICU patients (CTRL vs. non-ICU: p = .04). On the remaining tests, no statistically significant differences between the groups were found |
| Vannorsdall et al., 2022 | USA | Prospective cohort study | Mean time diagnosis (SD) = 126.5 (70.1) | 82 Discharged patients (n = 48 post -ICU; n = 34 post non ICU) 95% participants = ≤12 years education Mean age 54.5 years | to prospectively characterize cognition, mental health symptoms, and functioning approximately 4 months after an initial diagnosis of COVID-19 in a racially and ethnically diverse group of patients. | RAVLT, Oral Trail Making Test70,Digit Span forward and backward, Semantic and phonemic fluency via tele assistance | RAVLT acquisition and delayed recall: 26.8% of patients were mildly/moderately impaired and 14.6% were severely impaired. TMT-A: 35.4% of patients were mildly/moderately impaired and 25.6% were severely impaired. TMT-B: 15.9% of patients were mildly/moderately impaired and 13.4% were severely impaired. Digit Span forward: 1.2% of patients were mildly/moderately impaired and 1.2% were severely impaired. Digit Span backward: 13.4% of patients were mildly/moderately impaired and 4.9% were severely impaired. Phonemic fluency: 25.6% of patients were mildly/moderately impaired and 8.5% were severely impaired. Semantic fluency: 31.7% of patients were mildly/moderately impaired and 9.8% were severely impaired |
| Walle-Hansen et al., 2021 | Norway | Prospective cohort study | 6 months after | 106 Discharged patients (subdivided by age into 2 subgroups) (ambulatory evaluation) Mean age = 74.3 (range 60–96) Comorbid condition | To study age-related change in health-related quality of life, functional decline and mortality among older patients following hospitalization due to COVID-19 | MoCA | A negative change in cognitive functions was reported by 43% of patients 6 months after hospitalization. MoCA scores were lower in the oldest age group (mean: 21.7) compared to younger participants (mean: 25.3) |
| Woo et al., 2020 | Germany | Cross-sectional study | Median: 85 days from COVID-19 recovery [range: 20–105 days] | 18 patients (mean age [SD] = 42.2 [14.3]) 10 healthy controls (mean age [SD] = 38.4 [14.4]) | Study focused on neurocognitive deficits in mild COVID-19 young patients | TICS-M (by phone or directly) | The performance of COVID-19 patients was significantly lower (mean: 38.83; range: 31–46) than the healthy controls (mean: 45.8; range: 43–50), especially on short-term memory, attention, and concentration/language tasks |
Montreal Cognitive Assessment Scale
Rey Auditory Verbal Learning Test
Hospital Anxiety and Depression Scale
Executive functions
Repeatable Battery for the Assessment of Neuropsychological Status
Colour-Word Interference Test
Delis-Kaplan Executive Function System
Multidimensional Fatigue Inventory
Beck Depression Inventory (Version I)
Fatigue Severity Scale
Epworth Sleepiness Scale
Impact of Event Scale—Revised
Assessment of Neurological and Behavioral Function
Patient Health Questionnaire-9
Generalized Anxiety Disorder-7
Trial Making Test Part A & Part B
Clinical Interview Schedule-Revised
Structured Clinical Interview for DSM-5, Research Version
Ask Suicide-Screening Questions
Post-Traumatic Stress Disorder Checklist
Alcohol Use Disorder Identification Test
Memory Complaint Scale
As obtained from the Mini Mental State Examination
Verbal Fluency Test
Clinical Frailty Scale
International Physical Activity Questionnaire -Short Version
Symbol Digit Modalities Test
Boston Naming Test
Judgment Line Orientation
Rey-Osterrieth Complex Figure (copy and recall at 3,30 min and recognition)
Free and Cued Selective Reminding Test
Visual Object and Space Perception Battery
With the following tests: Trail making Test (TMT), Figural Memory Test(FGT,S11 form), Tower of London (TOL-F),Inhibition Response (INHIB, S13 Form),N-Back Verbal Test,Cognitrone, Reaction Test, Determination Test, WAB Battery (S1 Form) of perception and attention functions
Brief Smell Identification Test
State- Trait Anxiety Inventory
Beck Depression Inventory -II
Pittsburgh Sleep Quality Index
Magnetic Resonance Imaging
National Adult Reading Test
Brief Symptom Inventory
Cognitive Failure Questionnaire
Lawton–Brody Instrumental Activities of Daily Living Scale
Patient Health Questionnaire-9
Endicott Quality of Life Enjoyment and Satisfaction Scale
Patient Assessment of Own Function
Brief Repeatable Battery Neuropsychological Tests. The BRB-NT includes the Selective Reminding Test, Spatial Recall Test, Symbol Digit Modalities Test, the Paced Auditory Serial Addition Test, and the word List Generation Test
Subjective Scale of Damage questionnaire
Quality of life in Neurological Disorders
Commers Continuous Performance Test II
Rey-Osterrieth Complex Figure Test
15- Objects Test
Patient-reported Outcome Measurement Information System
National Institutes of Health Toolbox v2.1
Custom Computerized Assessment Battery
PTSD Checklist for Diagnostic and Statistical Manual of Mental Disorders
Digital Simplified Chinese Version
Wechsler Memory Scale -IV
From the Wechsler Adult Intelligence Scale -IV
Color-Word Interference
Subtests from the WAIS-IV
Frontal Assessment Battery
Free and Cued Selective Reminding Test
Rey–Osterrieth Complex Figure
Visual Object and Space Perception Battery
Wechsler Adult Intelligence Scale 4th Edition
Judgment Line Orientation
Test de Aprendizaje Verbal Espana-Complutense
Screen for Cognitive Impairment in Psychiatry Danish Version with five subtests of verbal learning and memory, working memory, verbal fluency and processing speed
Vigilance Task (VT), Stroop Interference Task (SIT) and Navon Task (NT
Part A and B
Assessment of Bias
The Risk of Bias (RoB) assessment was based on the RoB Cochrane tool for Systematic Reviews of interventions. Two reviewers (EP and MCP) independently assessed the RοΒ for each study. In case of discrepancies, a discussion between the two reviewers followed to resolve them. The RoB diagram was generated using the Robvis tool.
Overview of Patient Clinical and Demographics Characteristics
We included studies that evaluated cognitive impairment after or at 12 weeks of COVID-19 infection. Regarding patients’ assessment method, as would be expected, some studies used tele-assistance services during the critical phase of the pandemic (Frontera et al., 2021; Hampshire et al., 2022; Valdes et al., 2022; Vannorsdall et al., 2022). Patients’ characteristics are presented in Table 1. Overall, 3,256 patients were involved. Sample sizes ranged from 18 to 324 patients (Abramoff et al., 2023; Woo et al., 2020). Median age ranged from 35.5 years (He et al., 2023) to 74.3 years (Walle-Hansen et al., 2021). Furthermore, in the majority of the included studies, more than half of the participants were female. Different types of samples were found. Most involved discharged patients, whereas others evaluated only outpatients never hospitalized due to COVID-19.
These studies collectively provided insights into the cognitive and health outcomes of COVID-19 patients, with variations in patient demographics, severity of symptoms, and timing of assessments. It is important to note that the long-term effects of COVID-19, such as cognitive impairment, can vary among individuals and may depend on various factors including long-lasting symptoms and variations in severity based on healthcare settings and patient characteristics (Abramoff et al., 2023; Albu et al., 2021; Amalakanti et al., 2021; Becker et al., 2021; Birberg Thornberg et al., 2022).
Studies provide insights into cognitive, psychiatric, and neurological effects of COVID-19, as well as the chronicity of these symptoms and their association with different demographic and clinical characteristics (Boesl et al., 2021; Brown et al., 2022; Chen et al., 2022).
Other studies aimed to assess the cognitive outcomes and potential risks associated with COVID-19, taking into account factors such as severity of illness, education level, comorbidities, and time since infection or intensive care unit (ICU) discharge (Crivelli et al., 2022; Damiano et al., 2022; Del Brutto et al., 2021; Delgado-Alonso et al., 2022; Duindam et al., 2022; Ferrando et al., 2022; Ferrucci et al., 2021; Frontera et al., 2021; Vannorsdall et al., 2022). Comorbid conditions, such as hypertension, cardiac conditions, chronic obstructive pulmonary disease, asthma, diabetes, and other conditions were noted in COVID -19 patients with higher prevalence in the older age groups.
Studies included patients who experienced different levels of disease severity (García-Sánchez et al., 2022; Hampshire et al., 2022; Miskowiak et al., 2021; Miskowiak et al., 2023; Rass et al., 2021; Rubega et al., 2022; Walle-Hansen et al., 2021; Woo et al., 2020). These studies often included control groups or compared different patient groups to draw associations between specific demographic and clinical characteristics and the presence or severity of post-COVID symptoms.
They also revealed the persistence of symptoms in some cases, even months after the acute phase of illness (Lamontagne et al., 2021; Lauria et al., 2022; Matias-Guiu et al., 2023; Miskowiak et al., 2021; Miskowiak et al., 2023). Many patients continued to experience various symptoms, including cognitive complaints, fatigue, anxiety, and other neurological and non-neurological symptoms, underscoring the importance of understanding and managing the long-term effects of COVID-19 (Gautam et al., 2022; Graham et al., 2021; Hartung et al., 2022; He et al., 2023; Krishnan et al., 2022).
Several studies highlighted the presence of psychiatric symptoms, including anxiety, depression, and mood disturbances, among COVID-19 survivors. Additionally, preexisting psychiatric conditions seemed to worsen in some cases. In summary, this body of research collectively highlights the diverse range of cognitive and neuropsychological effects seen in individuals recovering from COVID-19, emphasizing the need for ongoing research and support for post-COVID syndrome patients.
Neuropsychological Deficits
A wide range of cognitive assessment tools were used among the included studies. Some studies administered screening tests to assess global cognitive function, whereas other studies used comprehensive neuropsychological batteries to assess memory, executive function, attention and information processing speed, language, and visuospatial functions. The way of comparing the performance of post-COVID-19 patients differed among studies, as some studies compared performance using normative data and/or cut-off scores, whereas others compared performance with healthy controls.
The definition of cognitive impairment varied across studies, lacking a standardized approach. Each study employed different criteria or thresholds to classify cognitive impairment, leading to variability in how cognitive deficits were identified and measured among COVID-19 patients. This variability was due to differences in the selection of cognitive assessment tools, cutoff scores, or clinical contexts, making it challenging to establish a standardized definition of cognitive impairment in the context of COVID-19 research. As a result, understanding and comparing findings across studies require careful consideration of the specific criteria used to define cognitive impairment within each study. Generally, there was disagreement on which tests to employ and what criteria to use for defining impairment. Some studies relied solely on screening tools like the Montreal Cognitive Assessment (MOCA) or the Mini-Mental State Examination (MMSE). Others utilized telephonic or online cognitive tests, some with established normative data. Additionally, either computerized or traditional paper-and-pencil testing methods were used among the studies. Neuropsychological batteries of different depths were commonly used, with impairment determined by comparing group scores to controls (Amalakanti et al., 2021; Crivelli et al., 2022; Del Brutto et al., 2021; Delgado-Alonso et al., 2022; Ferrando et al., 2022; Hampshire et al., 2022; Hartung et al., 2022; He et al., 2023; Lamontagne et al., 2021; Miskowiak et al., 2021; Miskowiak et al., 2023; Ortelli et al., 2021; Pihlaja et al., 2023; Rubega et al., 2022; Woo et al., 2020) or comparing performance between patient subgroups according to the severity of COVID symptoms and according to whether they had been hospitalized in ICUs or not (Pilotto et al., 2021; Vannorsdall et al., 2022). Studies utilizing comparisons to published test norms employed varying definitions of impairment, which included scores equal to or greater than 0.5 SD (Miskowiak et al., 2021), equal to or greater than 1 SD (Matias-Guiu et al., 2023; Miskowiak et al., 2021), equal to or greater than 1.5 SD (Becker et al., 2021; Birberg Thornberg et al., 2022; Duindam et al., 2022; Krishnan et al., 2022; Matias-Guiu et al., 2023; Vannorsdall et al., 2022), and equal to or greater than 2 SD (Vannorsdall et al., 2022).
The diverse utilization of demographic adjustments presented additional obstacles when comparing findings among studies. Research employing adjustments for factors like age, education, cultural background, and gender yielded varying rates of impaired scores compared to those studies not implementing such adjustments (Damiano et al., 2022; Pilotto et al., 2021). As expected, research employing looser definitions of impaired scores tended to document elevated rates of cognitive impairment. The use of two cutoffs allowed for the identification of different severity levels of impairment (Matias-Guiu et al., 2023; Miskowiak et al., 2021). Despite significant variations, studies employing stricter standards indicated that cognitive impairment affects 18%–35% of participants (Becker et al., 2021; Delgado-Alonso et al., 2022). The absence of controlled studies restricted the ability to draw conclusions and generalize findings (Crivelli et al., 2022). A more intense initial phase of illness correlated with a heightened presence of persistent cognitive decline (Ferrucci et al., 2021; Hampshire et al., 2022). However, cognitive deficits were commonly observed even in less severe cases of initial illness, including individuals who were never hospitalized (Becker et al., 2021).
Global Cognitive Function
For the assessment of global cognitive function, 18 studies (Abramoff et al., 2023; Amalakanti et al., 2021; Boesl et al., 2021; Brown et al., 2022; Chen et al., 2022; Duindam et al., 2022; Frontera et al., 2021; García-Sánchez et al., 2022; Gautam et al., 2022; Hartung et al., 2022; Ortelli et al., 2021; Pihlaja et al., 2023; Pilotto et al., 2021; Rass et al., 2021; Rubega et al., 2022; Walle-Hansen et al., 2021), administered the MoCA, of which 12 (Abramoff et al., 2023; Boesl et al., 2021; Chen et al., 2022; Duindam et al., 2022; Frontera et al., 2021; Gautam et al., 2022; Hartung et al., 2022; Ortelli et al., 2021; Pilotto et al., 2021; Rass et al., 2021; Walle-Hansen et al., 2021) recorded low performance in COVID-19 patients, which was mainly indicative of mild cognitive impairment. The percentages of patients with impaired scores ranged from 5% in the study of Duindam and coworkers (2022) to 50% in the study of Frontera and coworkers (2021). However, in four studies (Amalakanti et al., 2021; Crivelli et al., 2022; Pihlaja et al., 2023; Rubega et al., 2022), the performance of post-COVID-19 patients was normal. In the study of Amalakanti and coworkers (2021), while no statistically significant difference was found between the patients and the control group in terms of global cognitive function, the patients performed lower in the domains of visuospatial function, naming, and verbal fluency compared to the control group.
The Modified Telephone Interview for Cognitive Status (TICS-M) was administered in one study (Woo et al., 2020), in which post-COVID-19 patients had poorer performance compared to the control group. However, total scores were not correlated with disease severity and treatment, nor with the presence of depressive symptoms or fatigue (Woo et al., 2020). Finally, one study (Lauria et al., 2022) administered the MMSE, where no pathological scores were detected in older patients with COVID-19.
Memory
For the assessment of memory function, a wide range of tests was administered for different types of memory. Sixteen studies (Albu et al., 2021; Becker et al., 2021; Birberg Thornberg et al., 2022; Crivelli et al., 2022; Delgado-Alonso et al., 2022; Ferrando et al., 2022; Ferrucci et al., 2021; García-Sánchez et al., 2022; Hampshire et al., 2022; Krishnan et al., 2022; Lauria et al., 2022; Matias-Guiu et al., 2023; Miskowiak et al., 2021; Miskowiak et al., 2023; Rubega et al., 2022; Vannorsdall et al., 2022) administered verbal learning and recall tests, and seven studies (Birberg Thornberg et al., 2022; Crivelli et al., 2022; Delgado-Alonso et al., 2022; Ferrucci et al., 2021; García-Sánchez et al., 2022; Krishnan et al., 2022; Matias-Guiu et al., 2023) administered tests that assess visual memory, such as recall of a complex figure or other figures and/or memorization of associative images. Story recall tests were administered in two studies (Birberg Thornberg et al., 2022; Krishnan et al., 2022).
In verbal learning and recall tests, 13 studies referred to the learning condition, of which 12 (Albu et al., 2021; Becker et al., 2021; Birberg Thornberg et al., 2022; Ferrando et al., 2022; Ferrucci et al., 2021; García-Sánchez et al., 2022; Lauria et al., 2022; Matias-Guiu et al., 2023; Miskowiak et al., 2021; Miskowiak et al., 2023; Vannorsdall et al., 2022) recorded impaired scores in a proportion of individuals with COVID-19, while Rubega and coworkers (2022) did not identify differences in performance. The percentage of individuals who showed impairment ranged from 10.5% (Ferrucci et al., 2021) to 26.8% (Vannorsdall et al., 2022). In the delayed recall condition, 13 studies (Albu et al., 2021; Becker et al., 2021; Birberg Thornberg et al., 2022; Delgado-Alonso et al., 2022; Ferrando et al., 2022; Ferrucci et al., 2021; García-Sánchez et al., 2022; Lauria et al., 2022; Matias-Guiu et al., 2023; Krishnan et al., 2022; Miskowiak et al., 2023; Vannorsdall et al., 2022) reported decreased performance in patients with COVID-19, with rates ranging from 10% (Krishnan et al., 2022) to 26.98% (García-Sánchez et al., 2022). Miskowiak and coworkers (2021) and Rubega and coworkers (2022) did not identify differences in performance between patients and controls. Finally, three studies (García-Sánchez et al., 2022; Hampshire et al., 2022; Krishnan et al., 2022) reported on the recognition condition, in all of which measured deficits were noted in patients.
In terms of visual memory, in all studies (Birberg Thornberg et al., 2022; Crivelli et al., 2022; Delgado-Alonso et al., 2022; Ferrucci et al., 2021; García-Sánchez et al., 2022; Krishnan et al., 2022; Matias-Guiu et al., 2023), a specific proportion of the COVID-19 group performed poorly in the immediate and delayed recall conditions. Finally, with respect to delayed story recall tests, in the study of Birberg Thornberg and coworkers (2022), 15.2% of patients performed 1.5 SD above the average and 8% of patients performed 2 SD above the average, while in the Krishnan and coworkers (2022) study, the score of 10% of patients was indicative of impairment.
Executive Function
Twenty of the included studies (Becker et al., 2021; Birberg Thornberg et al., 2022; Damiano et al., 2022; Delgado-Alonso et al., 2022; Duindam et al., 2022; Ferrando et al., 2022; Ferrucci et al., 2021; García-Sánchez et al., 2022; Graham et al., 2021; Hampshire et al., 2022; He et al., 2023; Krishnan et al., 2022; Lauria et al., 2022; Matias-Guiu et al., 2023; Miskowiak et al., 2021, 2023; Ortelli et al., 2021; Rubega et al., 2022; Vannorsdall et al., 2022) administered tests to assess executive function.
Twelve studies (Becker et al., 2021; Delgado-Alonso et al., 2022; Duindam et al., 2022; Ferrando et al., 2022; García-Sánchez et al., 2022; He et al., 2023; Krishnan et al., 2022; Miskowiak et al., 2021, 2023; Rubega et al., 2022; Vannorsdall et al., 2022) administered the Trail Making Test-Part B (TMT-B), in all of which poor performance was recorded in post-COVID-19 patients, with rates ranging from 5% (Krishnan et al., 2022) to 21% (Duindam et al., 2022). A similar pattern was observed in the verbal fluency test, semantic and/or phonological condition, which was administered in 13 studies (Albu et al., 2021; Becker et al., 2021; Birberg Thornberg et al., 2022; Damiano et al., 2022; Delgado-Alonso et al., 2022; Ferrando et al., 2022; Ferrucci et al., 2021; García-Sánchez et al., 2022; Matias-Guiu et al., 2023; Miskowiak et al., 2023, 2021; Vannorsdall et al., 2022). All studies reported low scores in patients with COVID-19, except for two studies (Damiano et al., 2022; Miskowiak et al., 2021). In particular, the study of Miskowiak and coworkers (2021) noted a trend of underperformance, which did not, however, reach a significance level.
In addition, the Stroop Test was administered in six studies (Delgado-Alonso et al., 2022; Ferrando et al., 2022; García-Sánchez et al., 2022; Matias-Guiu et al., 2023; Ortelli et al., 2021; Rubega et al., 2022), of which in only one study (Rubega et al., 2022) post-COVID-19 patients did not score lower than expected. Likewise, in two studies (Birberg Thornberg et al., 2022; Krishnan et al., 2022), where the D-KEFS Color-Word Interference was employed, more than 10% of patients indicated subpar performance.
The Wisconsin Card Sorting Test (WCST) was administered in two studies (Krishnan et al., 2022), in both of which patients scored in the low range based on the WCST-recommended qualitive labels. Meanwhile, Delgado-Alonso and coworkers (2022) reported low performance in a significant proportion of patients when administering the Tower of London Test. Moreover, two additional executive function tests were administered, the Inhibition Response and the N-back Verbal Test, in which the frequency of impairment was at least two to three times more frequent than expected performance, respectively.
The Verbal Analogies and Spatial Planning tests were administered by Hampshire and coworkers (2022), in which low performance of post-COVID-19 patients was noted for both the accuracy and latency index. In the Matrix Reasoning and Similarities subtests from the Wechsler Adult Intelligence Scale-IV (WAIS-IV), administered in one study (Krishnan et al., 2022), 5% of patients achieved a low score on the former test, whereas on the latter test, all patients performed normally.
Graham and coworkers (2021) assessed executive function through the National Institute of Health (NIH) Toolbox, in which no statistically significant differences were found in performance of post-COVID-19 patients compared to normative data. Finally, three studies (Lauria et al., 2022; Ortelli et al., 2021; Rubega et al., 2022) administered the Frontal Assessment Battery (FAB); only the study of Rubega and coworkers (2022) reported no statistically significant differences in terms of the scores of the post-COVID-19 patient group compared to the control group.
Attention, Information Processing Speed, and Working Memory
Twenty studies (Becker et al., 2021; Birberg Thornberg et al., 2022; Damiano et al., 2022; Delgado-Alonso et al., 2022; Duindam et al., 2022; Ferrando et al., 2022; Ferrucci et al., 2021; García-Sánchez et al., 2022; Graham et al., 2021; He et al., 2023; Krishnan et al., 2022; Lamontagne et al., 2021; Lauria et al., 2022; Matias-Guiu et al., 2023; Miskowiak et al., 2021; Miskowiak et al., 2023; Ortelli et al., 2021; Rubega et al., 2022; Vannorsdall et al., 2022) assessed attention and information processing speed in post-COVID-19 patients.
The Trail Making-Part A (TMT-A) test was administered in 11 studies (Becker et al., 2021; Damiano et al., 2022; Delgado-Alonso et al., 2022; Duindam et al., 2022; Ferrando et al., 2022; García-Sánchez et al., 2022; Krishnan et al., 2022; Lauria et al., 2022; Rubega et al., 2022; Vannorsdall et al., 2022), of which only two studies (Ferrando et al., 2022; Rubega et al., 2022) did not observe lower than expected performance in post-COVID-19 patients. In addition, the Symbol Digit Modalities Test (SDMT) was administered in four studies (Delgado-Alonso et al., 2022; Ferrucci et al., 2021; Matias-Guiu et al., 2023; Rubega et al., 2022), of which only the study of Rubega and coworkers (2022) found no statistically significant difference in patients’ performance compared to the healthy sample. Coding and symbol search tests were administered in five studies (Birberg Thornberg et al., 2022; García-Sánchez et al., 2022; He et al., 2023; Krishnan et al., 2022), in all of which reduced performance was reported in a proportion of patients, except for the study of He and coworkers (2023).
Delgado-Alonso and coworkers (2022) additionally administered computerized tests to assess attention and information processing speed. Specifically, they administered the Cognitrone, Reaction Test, Determination Test, and Vigilance - WAFV battery. In the WAF battery, the post-COVID-19 patients performed statistically significantly lower than the control group, in the Reaction and Determination Test, a non-statistically significant trend was observed, while in the Cognitrone, there was no statistically significant difference in performance between the two groups. The Conners Continuous Performance Test (CCPT) was administered in two studies (García-Sánchez et al., 2022; Krishnan et al., 2022), both of which noted low scores in a proportion of patients.
Lamontagne and coworkers (2021) used the Attention Network Test (ANT) to assess three attentional networks: alerting, orienting, and executive control (conflict). Statistically significant differences were found in both accuracy and reaction times of post-COVID-19 patients compared to the control group. In this study, differences were recorded between the two groups in the executive network, but not in the alerting and orienting networks. In parallel, in the study of Ortelli and coworkers (2021), two computerized tests were administered, the Vigilance Task (VT) and the Navon Task (NT). While in both tests the error rates of the post-COVID-19 group were statistically significantly higher compared to the control group, reaction times were statistically significantly higher only in the NT.
Graham and coworkers (2021) assessed attention and processing speed through the NIH Toolbox. In the attention test, patients’ performance differed from expected performance, whereas no differences were found in the processing speed test. The processing speed test from the Screen for Cognitive Impairment in Psychiatry Danish Version (SCIP-D) was used by Miskowiak and coworkers (2021, 2023). Miskowiak and coworkers (2021) reported a non-statistically significant trend of difference in patient performance compared to the control group, whereas in the study of Miskowiak and coworkers (2023), no differences were found between the two groups. Finally, He and coworkers (2023) administered the Reaction Time Paradigm (RTI) and the One-back task; neither of those reported differences were recorded between the patients and the control group.
For the assessment of working memory, 12 studies (Albu et al., 2021; Becker et al., 2021; Birberg Thornberg et al., 2022; Delgado-Alonso et al., 2022; Duindam et al., 2022; García-Sánchez et al., 2022; Krishnan et al., 2022; Lauria et al., 2022; Matias-Guiu et al., 2023; Rubega et al., 2022; Vannorsdall et al., 2022) administered the Digit Span, forward and backward task, two studies (Miskowiak et al., 2021; Miskowiak et al., 2023) administered the Working Memory Test from the SCIP-D, two studies (Chen et al., 2022; Graham et al., 2021) the Working Memory Test from the NIH Toolbox, one study (Duindam et al., 2022) the Letter Digit Substitution Test, one study (Ferrucci et al., 2021) the Pace Auditory Serial Addition Test (PASAT), one study (Delgado-Alonso et al., 2022) the Corsi block-tapping test, and one study (Hampshire et al., 2022) the Spatial Span test. In the aforementioned tests, all studies recorded reduced performance in a proportion of post-COVID-19 patients; one exception was for the study of Rubega et al., 2022, in which patients performed similarly to the healthy control group in the Digit Span forward and backward test post-COVID-19.
Language
A limited number of studies evaluated language function. In particular, four studies (Delgado-Alonso et al., 2022; García-Sánchez et al., 2022; Krishnan et al., 2022; Matias-Guiu et al., 2023) administered the Boston Naming Test (BNT), in all of which impaired performance was reported in a proportion of patients ranging from approximately 9% (Matias-Guiu et al., 2023) to 16% (Krishnan et al., 2022). Ferrando and coworkers (2022) assessed language function through the language index from the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) recording lower than expected performance in patients.
In addition, the Reading subtest from the Wide Range Achievement Test-IV was administered by Krishnan and coworkers (2022), where 5% of patients achieved a low score. In the study of Chen and coworkers (2022), the Picture Vocabulary Test from the NIH Toolbox was used, in which 30% of the patients showed impaired vocabulary performance. Through the administration of the Multilingual Naming Test, no statistically significant differences in performance between patients and healthy subjects was observed.
Visuospatial Functions
For the assessment of visuospatial functions, three studies (García-Sánchez et al., 2022; Matias-Guiu et al., 2023) administered the copy task of the complex figure tests, of which only one did not show poor performance in the post-COVID-19 patient group. In addition, the Judgement of Line Orientation test (JLO) was administered in three studies (Delgado-Alonso et al., 2022; Krishnan et al., 2022; Matias-Guiu et al., 2023) and the Visual Object and Space Perception battery (VOSP) in two studies (Delgado-Alonso et al., 2022; Matias-Guiu et al., 2023), in all of which lower than expected performance was recorded in a proportion of post-COVID-19 patients.
In addition, two studies (Birberg Thornberg et al., 2022; Ferrando et al., 2022) assessed visuospatial functions using the RBANS. In the study of Birberg Thornberg and coworkers (2022) 21.1% of post-COVID-19 patients scored 1.5 SD below the mean and 10.2% scored 2 SD below the mean, whereas in the study of Ferrando and coworkers (2022), the patient group did not score statistically significantly lower than expected. Additionally administered the Block Design test and the 15-Objects test, in which 4.76% and 7.94% of patients, respectively, performed below expected levels.
The 2D manipulation, Target Detection, and 3D Rotation tests were administered in one study (Hampshire et al., 2022), in which only on the Target Detection test, no differences in the performance of the patient group compared to normative data were observed. In parallel, the Multiple Features Target Cancellation Test was administered in one study (Lauria et al., 2022), in which 4% of patients scored lower than expected. Finally, the Clock Drawing Test (CDT) was administered in one study, in which the group of post-COVID-19 patients had statistically significantly lower performance than the control group.
DISCUSSION
The goal of this systematic review was to provide findings from an overview of studies that evaluated specific neurocognitive functions affected by COVID-19 at or after 12 weeks from onset of illness. Participants’ cognitive assessments were conducted at different time points relative to symptom onset, but all the studies included reported data collected at or after 12 weeks from acute infection. Current evidence suggests a highly multifactorial set of components leading to long COVID: direct infection by SARS-CoV-2, the consequence of prolonged time spent in ICUs, persistent inflammation, brain hypoxia, ventilation mechanisms utilized, various pharmaceutical drugs, prior cognitive difficulties, and peripheral organ dysfunction. The uncontrolled inflammatory response, also referred to as cytokine storm was associated disease severity. Increased level of inflammatory cytokines and chemokines and high levels of IL-6, IL-8, and TNF-α were found in COVID-19 patients’ serum. A significant number of these long-COVID sufferers reported experiencing cognitive symptoms that prevented them from returning to work and the resumption of daily activities. Although cognitive symptoms can persist as long as one year after illness onset, there is evidence that they might improve after approximately 4 months (Lamontagne et al., 2021).
Neurocognitive impairment has been commonly reported in patients with a viral infection. Symptoms commonly associated with COVID-19, such as anosmia, have previously been associated with the onset of neurocognitive disorders and neurodegeneration (Chen et al., 2022). Recent research has documented cognitive functioning using brief screening measures, such as the MoCA or MMSE. One study reported a higher rate of decline on the MoCA among individuals who were seropositive for COVID-19 when compared with individuals who were seronegative (Del Brutto et al., 2021). The current evidence suggests that cognitive sequelae are frequent after COVID-19, even in mild cases not requiring hospitalization or ICU admission. These cognitive alterations include executive dysfunction and, to a lesser extent, memory impairment, which progressively improves in most cases (Damiano et al., 2022; Ferrucci et al., 2021; Rubega et al., 2022; Vannorsdall et al., 2022). Impaired quality of life (QoL) and cognitive deficits were reported in 31% and 23% of patients, respectively. Still, functional outcome was good, with almost all patients living independently at 3 months (Rass et al., 2021). A high impact of persistent symptoms on daily life activities and QoL, regardless of acute infection severity, was also detected (Albu et al., 2021).
In SARS and MERS, after recovery from the infection, impairment of memory, attention, concentration, or mental processing speed were reported in more than 15% of patients at a follow-up period ranging between 6 weeks and 39 months (Mazza et al., 2021; Rubega et al., 2022; Vannorsdall et al., 2022). Conversely, patient-reported physical functioning, Activities of Daily Living (ADL), return to work, cognitive and mental function, and health-related quality of life (HRQoL) may be relatively less impaired than the performance-based measures of physical status and in comparison, with pre-COVID-19 studies of Acute Respiratory Distress Syndrome (ARDS).
The high impact of persistent symptoms on daily life activities and QoL advocate for a multidisciplinary evaluation and clinical profiling of post-COVID-19 patients before initiating rehabilitation interventions, regardless of the severity of acute infection (Albu et al., 2021). Furthermore, as COVID-19 can cause critical illness in some patients, ARDS and delirium may occur, both previously related to cognitive impairment. Cognitive impairment was found in 29% of severe COVID-19 patients, 30% of moderate COVID-19, and 3% of mild COVID-19 patients (Rass et al., 2021).
The variability of post COVID-19 physical and neuropsychological impairment requires a complex screening process both in ICU and non-ICU patients. Post-ICU patients present sequelae of critical illness myopathy and polyneuropathy, stroke and encephalopathy and lower forced vital capacity compared to non-ICU patients. Cognitive impairment was found in 63.3% of patients, with a similar profile in both sub-groups. Impairments in executive functioning, processing speed, category fluency, memory encoding, and recall were predominant among hospitalized patients. The relative sparing of memory recognition in the context of impaired encoding and recall suggests an executive pattern (Becker et al., 2021). In fact, 17.5% of survivors had cognitive impairment and their HRQoL was substantially lower than that of the general population and compared to aindividuals with other chronic conditions.
The majority of studies that assessed language performance, using a combination of Picture naming and different versions of Verbal Fluency subtests, reported similar negative findings (Ferrucci et al., 2021; Mazza et al., 2021; Miskowiak et al., 2021).
The association between depression and cognitive impairment was driven primarily by lower performance on tasks of verbal fluency, attention, and delayed recall among patients with higher severity of depression symptoms (Brown et al., 2022). There are also findings consistent with some prior research on associations between depression and memory among individuals recovering from SARS-CoV-2. Symptoms of depression were significantly associated with severity of cognitive impairment among patients recovering from SARS-CoV-2 infection in both univariate and multivariable analyses. Several studies have shown that inflammatory ratios can be used as inflammatory biomarkers in depression. The main finding is the presence of isolated persistent depressive symptomatology at 3-month follow-up after SARS-CoV-2 infection, which is predicted by systemic inflammation during acute infection and by its pattern of change over time (Mazza et al., 2021). Exploratory correlations with proinflammatory cytokines support further research into inflammatory mechanisms and viable treatments.
Positive correlations between executive function deficits and subjective cognitive complaints in a group of COVID-19 patients assessed 4 months after hospitalization were noted (García-Sánchez et al., 2022; Miskowiak et al., 2021). When directly tested with a neuropsychological battery, 33%, 23%, and 20% of participants failed the Trail Making, Digit Span Backwards, and Frontal Evaluation tests, respectively, showing impairment in visual perceptual skills, selective and divided attention, working memory, short-term verbal memory, and executive functions (Lauria et al., 2022).
Furthermore, apart from exploring the frequency of impaired performance across cognitive domains in post-COVID patients and uncovering whether impairment existed within a single domain or across multiple domains, several studies investigated potential links between cognitive performance and other clinical factors. In one of these studies, a strong correlation was found between ferritin levels during acute illness and language composite scores. This correlation could suggest a role for hyper-inflammation in neuronal damage underlying language impairment and especially in those language tasks that are timed and involve cognitive flexibility along with optimal frontal functioning (García-Sánchez et al., 2022).
The PASC phase was uniquely associated with a selective impairment in executive functioning, but not attentional orienting or alerting, highlighting specificity of cognitive dysfunction in post-COVID infection. Critically, none of these effects were moderated by appraisal of current life stress, nor were they a product of psychiatric impairment that predated the COVID-19 pandemic. Findings were not related to severity of illness or hospitalization. In fact, the overwhelming majority of their sample (98%) was non-hospitalized and effects did not vary by acute illness severity (Lamontagne et al., 2021).
Deficits in attention and memory indices were most common, each affecting approximately 30% of the patients (see Table 2). After adjustment for sex, language, level of education, and premorbid function, neurocognitive performance was positively associated with length of hospital stay, but not with disease severity indicators.
Percentage of patients displaying deficits in different cognitive domains
Percentage of patients displaying deficits in different cognitive domains
Findings suggest that comprehensive neuropsychological assessment should be performed when patients report post-COVID-19 symptoms that affect daily life (Birberg Thornberg et al., 2022). Cognitive deficits, when present, appear mild and isolated to domains of attention, processing speed, and executive function in a sample where most patients did not require intensive treatment. This finding may be a consequence of other factors developed as a result of COVID-19 (worsening mood, sleep disruption, fatigue, etc.) rather than a direct result of COVID-19 infection (Krishnan et al., 2022).
Given the diverse geographical origins of the studies included in this review, it’s pivotal to address how cross-cultural factors may have influenced the neuropsychological assessments. Acknowledging the monocultural origins of many neuropsychological tests, such as the Boston Naming Test, highlights the necessity for cultural sensitivity in cognitive evaluations. Our review found that a significant portion of the studies from non-Western countries incorporated culturally adapted tests or utilized locally developed norms to ensure relevance and accuracy in assessing cognitive functions. This adaptation is crucial for accurately interpreting neuropsychological performance across different cultural contexts, underscoring the importance of cultural competence in cognitive assessment. This consideration helps mitigate the risk of bias and ensures that findings are reflective of a truly global understanding of long-COVID’s cognitive impacts, providing a more nuanced and comprehensive analysis of the data.
In summary, the majority of studies reported deficits in cognitive functions and provided evidence of persistent cognitive changes in survivors of COVID-19 in several neurocognitive domains including, language, perceptual motor ability (visuospatial construction), executive function, learning and memory, complex attention, and social cognition. The neurocognitive domains found to be the most frequently impaired were executive function, memory, and attention. Several studies reported comorbidity in the post-COVID group, and this may have implications for participants’ premorbid cognitive ability.
Several studies limited cognitive assessment to screening tools and individual subtests rather than to detailed neuropsychological assessment batteries (Abramoff et al., 2023; Frontera et al., 2021; Lauria et al., 2022; Pilotto et al., 2021). Other studies did not compare performance to that of a healthy control group, and there was no subgroup in any study for comparison with a respiratory infection subgroup other than COVID disease.
Although in several studies there were patients with post-COVID of different severity, in the studies where patients had been hospitalized, we still do not know to what extent the treatment during their hospitalization contributed to their reported cognitive performance. Severely affected COVID-19 cases experience high levels of proinflammatory cytokines and acute respiratory dysfunction and often require assisted ventilation. All these factors have been suggested to cause cognitive decline and several potential mechanisms related to secondary effects of SARS-CoV-2 infection may contribute to cognitive impairment, such as hypoxia, hyperinflammation, hypercoagulability, blood–brain barrier disruption, uremia, septic encephalopathy, or autoimmune mechanisms.
Although most previous studies have extensively reported deficits in neurocognitive domains, little has been reported regarding language domains. The few studies that assessed language functions of post-COVID patients were limited to subtests of verbal fluency or naming. Therefore, there are scarce data describing the cognitive–linguistic or communication profile of individuals who have been infected with COVID-19 in the more chronic stages of recovery. Moreover, little is known about the language expression and comprehension abilities of COVID-19 survivors.
This review study has several limitations worth noting. Firstly, it relied solely on studies obtained from two electronic bibliographic databases, which may not encompass the full breadth of available literature. Moreover, certain studies failed to provide data on the prevalence or incidence of cognitive assessments of patients and neglected to address cognitive impairment subtypes. Additionally, a significant drawback was the absence of controls groups in most studies, which hindered direct comparisons. Furthermore, the diverse array of cognitive assessment instruments used across studies affected the external validity of the findings. Notably, none of the studies included in this review utilized a comprehensive cognitive neuropsychological evaluation, which is widely regarded as the gold-standard assessment method. These limitations collectively precluded the possibility of conducting a meta-analysis. Lastly, the decision to exclude gray literature may have helped maintain the quality of evidence, but we may have overlooked valuable insights from well-conducted studies or reports that may have contributed to a comprehensive analysis.
Moving forward, it is important for clinical research to address these limitations by incorporating control groups and employing comprehensive neuropsychological evaluations to identify the most affected cognitive domains accurately.
Our included studies evaluated patients with COVID-19 at least 12 weeks after illness initiation; however, the timing of neurocognitive assessment of the patients is not always known, as it occurred during a period of global lockdown (i.e., social isolation). Future research should take into consideration whether the period of lockdown is an additional risk factor or may cumulatively contribute to deficient neurocognitive performance. Additionally, we need to be aware of comorbid conditions (e.g., mood, anxiety or other major or minor psychiatric symptoms), lifestyle, fatigue, premorbid intelligence and premorbid cognitive levels, and the cognitive/brain reserve of individuals who were infected in order to thoroughly investigate the long-term cognitive consequences of the disease. It is imperative that future reviews provide systematic considerations of mood and/or other conditions as mentioned previously that may be associated with cognitive impairment in this heterogenous population.
Future studies should employ standardized or similar comprehensive neuropsychological assessment tools, including utilization of biomarkers. Such an approach would enable a thorough assessment of cognitive impairment across the continuum of care allowing for a better understanding of its impact on functional communication and cognition and facilitate improved intervention planning. Additionally, healthcare professionals should play a proactive role in early identification of cognitive impairment and its impact on functional abilities, offering tailored recommendations to optimize patient education and therapeutic goals for the benefit of the patient, their family, and the interdisciplinary healthcare team.
CONCLUSIONS
The primary aim of this study was to bring attention to a relatively unexplored topic and to establish a framework for the increasing evidence emphasizing the significance of evaluating cognitive impairment in individuals experiencing persistent symptoms of COVID-19.
Published epidemiological evidence revealed an uneven distribution of post-COVID cognitive impairment, suggesting the presence of factors that differentiate each individual’s susceptibility. Both genetic (e.g., APOE4 carriers) and environmental parameters have been implicated (Kuo et al., 2020). Research suggests that middle-aged individuals might be affected along with older ones; therefore, contrary to most neurocognitive entities, a younger age does not preclude cognitive implications and clinicians should be alert irrespective of the age of affliction (Matias-Guiu et al., 2023). Of note, healthcare providers should be more watchful in cases of lower educational attainment and socioeconomic status, as these individuals are more strongly predisposed toward post-COVID cognitive impairment (Liu et al., 2021; Matias-Guiu et al., 2023). Parameters associated with the course of the disease, for example, greater disease severity, severe pulmonary dysfunction, need for mechanical ventilation, and admission to the ICU, have been revealed to enlarge the risk and severity of cognitive manifestations (Evans et al., 2022; Miskowiak et al., 2021). Of interest, the use of several targeted therapies (antiviral agents) as well as the concomitant use of some supportive medication (e.g., benzodiazepines) has been associated with cognitive impairment (Izcovich et al., 2022; Kotfis et al., 2022).
Awareness of these factors can raise clinicians’ vigilance over post-COVID cognitive impairment. Individuals at higher risk could be monitored more closely and for a more prolonged period. Earlier identification of cognitive deficits is necessary to improve their management by implementing more timely interventions (e.g., cognitive rehabilitation) and avoiding iatrogenic complications (especially with respect to medications). Of note, awareness of the exact affected domains of cognition may provide a chance for superior, individualized cognitive interventions. At the same time, COVID infection appears to augment the risk of neuropsychiatric and sleep comorbidities, which are well-established risk factors for cognitive impairment and decline (Rogers et al., 2020).
In conclusion, according to the results of our systematic review, the majority of the published studies reported that patients with COVID-19 present cognitive impairment in several cognitive domains. The cognitive abilities most often impaired were executive functions, attention, and memory. Less attention has been paid to language abilities. Specifically, the language measures that were administered are limited mainly to confrontation naming and verbal fluency tasks. Data from several neurological populations such as Parkinson’s disease mild cognitive impairment and aphasia have suggested an association between language skills and non-linguistic cognitive abilities. Given the data derived from our review regarding cognitive impairment in several domains in this group of patients, it is highly possible that patients with COVID-19 will exhibit impaired language skills, especially in high-level language tasks. However, the current published literature has not investigated this area in depth. Future studies should investigate communication disorders in patients with COVID-19, as it will aid researchers and clinicians in better understanding the underlying mechanisms that might contribute to language and communication decline in this population.
ACKNOWLEDGEMENTS
The authors alone are responsible for the content and writing of this paper.
CONFLICT OF INTEREST
None declared.
References
Ortelli, P., Ferrazzoli, D., Sebastianelli, L., Engl, M., Romanello, R., Nardone, R., et al. (2021). Neuropsychological and neurophysiological correlates of fatigue in post-acute patients with neurological manifestations of COVID-19: Insights into a challenging symptom.
