Physical and cognitive impairments in people suffering from long COVID: protocol for a longitudinal population-based cohort study

Abstract

Introduction

Approximately 33% of people who contracted COVID-19 still experience symptoms 12 weeks after infection onset. This persistence of symptoms is now considered a syndrome itself called ‘long COVID’. Evidence regarding long COVID and its cognitive and physical impacts is growing, but the literature is currently lacking objectively measured data to guide towards adapted healthcare trajectories. The objectives are to describe the physical and cognitive impairments experienced by individuals living with long COVID using self-reported and clinical objective measures, and to compare the evolution over time of the physical and cognitive state between adults living with long COVID (at least one physical or cognitive COVID-19 symptom for more than 12 weeks following infection; long COVID group), people who developed COVID-19 but did not experience persistent symptoms (short COVID group) and people who did not develop COVID-19 (control group).

Methods and analysis

In this longitudinal cohort study, 120 participants will be recruited in each group. Variables will be collected through three evaluation sessions over 6 months (baseline, 3 months, 6 months). Variables include self-administered questionnaires on health-related quality of life, comorbidity, sleep, pain, anxiety, depressive symptoms, fatigue and cognitive function, as well as objective measures of cognitive (attention, memory, executive functioning) and physical (grip strength, balance, gait speed, gait endurance, VO2, frailty) functions. Activity, heart rate and sleep will be monitored with a fitness tracker watch for 7 days following evaluation sessions. Maximum-likelihood analyses of variance (ANOVAs) will be used to compare data at baseline between groups. Repeated measures ANOVAs will be used to compare the longitudinal performance variations across groups of the self-reported and clinical variables.

Ethics and dissemination

Ethics committees of the CIUSSS de la Capitale-Nationale and CIUSSS de l’Est-de-l’Île-de-Montréal approved the project. Results will be disseminated through clinical and community platforms as well as through peer-reviewed manuscripts and international conferences.

Trial registration number

NCT05216536.

STRENGTHS AND LIMITATIONS OF THIS STUDY

• Three distinct groups will be recruited, allowing a better understanding of the impact of COVID-19 on physical and cognitive impairments.

• Sample size will allow for subanalyses by sex, gender and age, among others.

• Study variables were carefully selected based on previous literature, expert recommendations and with the help of patient partners.

• With successive waves of COVID, participants in the control group are at risk of eventually developing COVID, which will make it difficult to keep our control group COVID-free for 6 months.

• Since people with long COVID fatigue quickly, the number of variables had to be reduced to minimise the development of fatigue; thus, it will be impossible to evaluate all persistent physical and cognitive impairments experienced by people with COVID as variables.

Introduction

The situation regarding the COVID-19 pandemic is evolving daily with the ongoing vaccination efforts and the new variants emerging. As of April 2022, over 500 000 000 cases of COVID-19 have been reported worldwide.1 One-third of people infected by COVID-19 self-declare still experiencing symptoms 12 weeks after their primary infection.2 People who have signs and symptoms that developed during or after a COVID-19 infection and that have continued for more than 12 weeks are understood to suffer from long COVID (also known as post-COVID syndrome).3 Multiple definitions of long COVID currently exist, and the absence of a consensual and universal nomenclature makes the understanding of this phenomenon and its prevalence challenging.4 For example, the National Institute for Health and Care Excellence (NICE) refers to long COVID when an individual experiences symptoms for more than 4 weeks and defines more precisely the persistence of symptoms beyond 12 weeks as ‘post-COVID-19 syndrome’.4 However, the National Institute for Health Research defines long COVID as symptoms lasting at least 12 weeks,2 which is the definition adopted for the current project.

Commonly reported symptoms that persist in long COVID include fatigue, shortness of breath, muscle weakness, joint pain, headache, cognitive dysfunction, symptoms of anxiety and depression, and sleep-related difficulties.3 5 6 Some individuals also experience post-traumatic stress disorder.7 8 The spectrum of reported symptoms in those experiencing long COVID has been so large that it is assumed that after COVID-19, a patient may develop persisting dysfunction of almost any organ system.9 In addition to being highly heterogeneous,10 symptoms vary in time, with some people experiencing constant symptoms, and others, intermittent symptoms.11 Furthermore, evidence is only emerging regarding who will develop long COVID. For example, women seem more likely to develop long COVID.12 Other findings show that individuals who were hospitalised and more severely ill during hospitalisation seem to present more severe persistent symptoms.13

Although evidence regarding the clinical presentation of COVID-19 is growing exponentially since the beginning of the pandemic, the duration and importance of the physical and cognitive symptoms experienced over time are not well understood.11 Results of studies to date have predominantly been based on self-reported symptoms, answers to self-reported questionnaires, retrospective data from hospital medical charts or on fundamental research. For example, studies using self-reported questionnaires showed that the median frequency of self-reported dyspnoea is 36%, of self-reported fatigue is 40.0%, of self-reported depression and anxiety are 15% and 22%, and of self-reported cognitive deficits is 17.6%.14 Although self-reported symptoms have high value to understand the symptoms experienced,15 16 current evidence does not provide sufficient prospective objectively measured data on the physical and cognitive impairments experienced by people with long COVID.9 Among the studies that have looked at these impairments, a cross-sectional study showed an increased prevalence of muscle weakness in long COVID patients, with muscle weakness up to 86% compared with the predicted normal values.4 Another study showed long COVID sequalae involving the cardiovascular system, as demonstrated by imaging analysis, which may be related to symptoms such as chest pain, dyspnoea and fatigue.4 Imaging analysis has also shown hypometabolism in some regions of the brain in long COVID patients. These abnormalities could explain cognitive impairments, pain and insomnia.4 However, to the best of our knowledge, there are very few studies that quantified prospectively the severity and evolution of those symptoms with objective clinical tests.

Prospective studies on long COVID have so far focused on previously hospitalised patients who experienced a critical form of COVID-19.15 For example, a prospective study with this population reported a desaturation under 88% in 7% of participants by the end of a 6 min walk test, 3 months after symptom onset.16 In another study, 14% of participants had a Short Physical Performance Battery score ≤10 at 3–6 months after hospital discharge, which shows an overall impaired physical function.17 Still, there is a lack of data regarding those who were not critically ill in the first place, but who still experience long COVID.

In light of the increasing burden of COVID-19, gaining a greater understanding of the physical and cognitive state and level of functioning of persons with long COVID will help better define healthcare needs of these persons and guide the development of appropriate medical and rehabilitation interventions and healthcare trajectories. Hence, the objectives of this study are (1) to describe the physical and cognitive impairments and functional limitations experienced by individuals with long COVID using self-reported and objective clinical measures, and (2) to compare the evolution over time of the physical and cognitive state and level of functioning between: (A) people with long COVID, (B) people who developed COVID-19 but did not experience persistent symptoms and (C) people who did not develop COVID-19.

Methods and analysis

Participants

One hundred and twenty adults with long COVID symptoms (long COVID group), 120 age-matched adults who developed COVID-19 but did not experience persistent symptoms (short COVID group) and 120 age-matched adults who did not develop COVID-19 (control group) will be recruited. All eligible participants will be 18 years of age or older and be able to participate in three in-lab evaluation sessions within 6 months. Participants assigned to the COVID groups will have received a positive result from COVID PCR or rapid antigen testing at least 12 weeks prior to inclusion. Participants assigned to the long COVID group will present at least one physical or cognitive long COVID-19 self-reported symptom for more than 12 weeks following the infection such as fatigue, shortness of breath, muscle weakness, joint pain, headache, cognitive dysfunction or sleep-related difficulties, as defined by NICE.18 Participants assigned to the short COVID group will not have experienced any persistent symptom for more than 4 weeks after having developed COVID-19.18 Participants assigned to the control group will not have received a diagnosis of COVID-19 and will not have experienced any COVID-19 symptoms since February 2020 such as fever, sudden loss of smell, dyspnoea, great fatigue, muscle or body aches or shortness of breath.19 No antibody or serology tests will be performed to determine if participants in the control group have been in contact with COVID but have not developed or experienced COVID symptoms.

Participants will be recruited through medical clinics in Quebec City and Montreal (eg, family medicine groups and long COVID Clinics), electronic mailing lists of students, employees and retired employees at Université Laval, and social media (Facebook). As more than 78 095 individuals in the Quebec City area and 287 035 in the Montreal area have developed COVID-19 as of March 2022 and that up to 33% are known to develop persistent symptoms,2 20 we are confident in being able to recruit the targeted number of participants. Pilot testing began in summer 2021. All participants are expected to be enrolled by winter 2023 with final follow-up assessments finalised by fall 2023.

Patient and public involvement

Two patient partners were recruited among healthcare professionals from the province of Quebec (Canada) who developed COVID-19 during the first wave (spring 2020). Our aim was to have one patient partner who had persistent symptoms of COVID-19 (long COVID) and one who developed COVID-19 but did not have symptoms for more than 4 weeks (short COVID). Patient partners were involved in the selection of outcomes and in the decision process regarding the in-lab evaluations modalities, to ensure an optimal balance between outcome collection and potential exacerbation of symptoms in the long COVID group. They also participated in reviewing the study protocol.

Study design

A longitudinal population-based cohort study of people who developed COVID-19 with or without persistent symptoms, and of people who did not develop COVID-19, with three evaluation sessions over 6 months (baseline, 3 months and 6 months) (figure 1).

Figure 1.

Study design.

Experimental procedures

Eligibility criteria will first be confirmed during a screening call with a research associate. After screening, all included participants will receive an email with a link to the online informed consent form. Once the consent form will have been electronically signed, they will be asked to complete the baseline web-based self-administered questionnaires (hosted on the Research Electronic Data Capture (REDCap) platform) evaluating health-related quality of life (HRQoL), comorbidity, sleep quality, pain and pain-related disabilities, anxiety, depressive symptoms, fatigue and cognitive function. These questionnaires have been selected to address the most commonly reported symptoms by individuals with COVID.3 5 6 The long COVID group will also complete a questionnaire on frequent COVID-19-related symptoms. Participants will have the possibility to complete the web-based questionnaires over a period of 5 days, in case they experience fatigue or cognitive impairments, or by phone if they do not have internet access. As some participants may start filling out the questionnaires, pause and forget to finalise them, the evaluator will verify on REDCap that all questionnaires have been completed before the in-lab evaluation. If they are not, participants will be asked to complete them during the in-lab evaluation.

Within 1 week of inclusion, all participants will complete the in-lab baseline evaluation at the Centre interdisciplinaire de recherche en réadaptation et en intégration sociale (Cirris) in Quebec City or at the Centre de recherche de l’Hôpital Maisonneuve- Rosemont (CRHMR) in Montreal. During this in-lab evaluation, participants will answer questions on sociodemographics (eg, age, sex, gender, ethnicity, smoking, income, education, employment, residential postal code), medications used and vaccination status. Weight and height will also be measured. Participants in the COVID groups will also answer questions on COVID-19 medical history (onset, variant, hospitalisation, intensive care, treatments received, number of episodes) and on COVID symptoms they have experienced since onset (nature, frequency, duration). Then, objective clinical tests will be performed on all participants to assess cognitive (attention, memory, executive functioning) and physical (grip strength, balance, gait speed, gait endurance, VO2 and frailty) functions. Finally, all participants will be asked to wear a fitness tracker watch (Garmin Forerunner 35) for 7 days to monitor heart rate, physical activity and sleep in their real-life environments. Evaluators will not be blind to which group the participants are in. Since individuals with long COVID can fatigue quickly, evaluators need to be aware which participants have long COVID in order to ascertain their condition and avoid deterioration of their condition.

Three and 6 months after the baseline evaluation, all participants will be asked to complete the follow-up web-based self-administered questionnaires on REDCap (same questionnaires as baseline) and to take part in the in-lab follow-up evaluations at Cirris and CRHMR. During the in-lab follow-up evaluations, participants will first be questioned on sociodemographics that may have changed since baseline, on any COVID symptoms experienced, and on any interventions received since the previous evaluation. Then the same cognitive and physical clinical tests will be performed and they will again be asked to wear a fitness tracker for 7 days to monitor the same parameters as in baseline.

Outcome measures

The variables were selected to cover the main long COVID symptoms reported in the literature3 5 6 and according to their established adequate psychometric properties and patient partners’ input.

Self-reported questionnaires

Health-related quality of life

EQ-5D-5L is a generic HRQoL questionnaire that contains five questions covering five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression.21 Each question is rated on a five-point scale. The combined dimensions describe 5⁵ = 3125 theoretically possible states of health that can be converted into a weighted index score ranging from 0 to 1.21 Its validity, reliability and responsiveness (Clinically important difference (CID)=0.32) have been established for various health conditions.22 23

Comorbidities

The Self-Administered Comorbidity Questionnaire (SCQ) is a validated generic questionnaire describing 12 common medical conditions (heart disease, high blood pressure, lung disease, diabetes, ulcer or stomach disease, kidney disease, anaemia or other blood disease, cancer, depression, osteoarthritis, rheumatoid arthritis and back pain), developed to adjust for the impact of comorbidity on functional status.24 The participants indicate if they are known for any of those conditions, if they are treated for those, and if those conditions result in activity limitations. One point is allocated for each positive answer, for a maximum of 3 points for each condition. As there are 12 defined medical problems and 3 open-field medical condition questions, the maximum global score is 45 points. The SCQ is a reproducible measure of comorbidity with moderate associations with standard medical record-based comorbidity charts.24

Sleep Quality

The Pittsburgh Sleep Quality Index (PSQI) is a questionnaire evaluating the quality and patterns of sleep over the last month. The questions are framed in a 4-point Likert scale (0–3) and cover seven factors including subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medications, and daytime dysfunction. Higher scores indicate more severe sleep difficulties. The scores from each component are added to give a sum score ranging from 0 to 21. The total score provides an overall summary of a person’s quality of sleep and sleep health.25 The PSQI has good internal reliability of α= 0.83 and test–retest reliability of 0.85.26

Pain and pain-related disabilities

The Brief Pain Inventory Short Form (BPI-SF) is a questionnaire with 11 items, designed to evaluate the intensity of, and the impairment caused by pain.27 Four items measure pain intensity (pain now, average pain, worst pain and least pain in the last 24 hours), and seven items measure the level of interference with function caused by pain (general activity, mood, walking ability, normal work, relations with other persons, sleep and enjoyment of life).27 BPI-SF is valid, internally consistent and reliable.27

Anxiety

The General Anxiety Disorder (GAD-7) is an instrument recognised as a valid and reliable measure of anxiety.28 It assesses the core symptoms of generalised anxiety disorder through seven questions by asking how frequent participants have been bothered by symptoms of anxiety over a period of 2 weeks.29 Responses range from 0 (not at all) to 3 (nearly every day), generating a sum score between 0 and 21.29 The GAD-7 has a strong construct validity and good-to-excellent test–retest reliability.28

Depression

The Patient-Health Questionnaire (PHQ-9) assesses the nine symptoms of a major depressive episode30 based on a 2-week period using a Likert scale in which ‘0’ means not at all and ‘3’ means nearly every day. The total score ranges between 0 and 27. A higher score signifies more severe symptoms. Cut-off scores for mild,5 moderate,10 moderately severe15 and severe20 depression can be used. A PHQ-9 score ≥10 has a sensitivity of 88% and a specificity of 88% for major depression when compared with a mental health professional assessment.31

Fatigue

The Fatigue Severity Scale measures the severity of fatigue and its effect on a person’s activities and lifestyle.32 It includes nine items scored on a seven-point scale, 1 (strongly disagree) to 7 (strongly agree). The sum score is between 9 and 63. The higher the score, the more severe the fatigue and the functional limitations are.33

Cognitive Function

The Patient-Reported Outcomes Measurement Information System Short Form v2.0—Cognitive Function measures perceived changes in cognitive function.34 The questionnaire contains eight items which participants rate on a scale of 1 (not at all) to 5 (very much) according to their experience in the prior 7 days. The total score ranges between 0 and 40. Higher scores stand for better perceived cognitive function. The questionnaire has excellent internal consistency and reliability.34

Persistent symptoms

The Newcastle post-COVID syndrome Follow-up Screening Questionnaire identifies individuals who are experiencing residual COVID-19-related symptoms and for who multidisciplinary management or specialist support may be beneficial. It contains 14 dichotomous questions exploring frequent COVID-19-related symptoms.35

In-lab evaluation

The tests will be administered in the same order for all participants and breaks will be given as needed. All evaluators will be trained to perform the various physical and cognitive tests used in this study. In addition, standardisation meetings will be held between evaluators in Quebec City and Montreal.

Cognitive evaluation

Montreal Cognitive Assessment

The Montreal Cognitive Assessment will be used to assess global cognition.36 It includes items for short-term memory, visuospatial abilities, executive functions, attention, concentration and working memory, language and orientation in time and place. It has a good reliability and construct validity for screening mild cognitive impairments. The score can range from 0 to 30 points, higher scores expressing better global cognitive functioning. An optimal cut-off score of 22 has been reported to detect mild cognitive impairments with a good sensitivity (81%) and specificity (77%).37

Digit Span subtest

The Digit Span subtest of the Wechsler Adult Intelligence Scale, Fourth Edition battery, including Forward and Backward tasks, will be administered as a measure of auditory attention and working memory.38

Trail making test

The Trail Making Test will be used to assess speed of information processing and executive functioning.39 The test is divided into two parts (A and B), in which the participants are first asked to connect numbers randomly distributed on a paper in ascending order (A) and then alternate between numbers and letters in consecutive order (B). The time of completion and number of errors of each part are collected.40 As for part A, completion time greater than 29 s and greater than 78 s are considered to be average and deficient, respectively. For part B, the average has been established at 75 s and a completion time greater than 273 s is considered deficient.41 Norms according to age and education also have been described in the literature.41

Physical evaluation

Grip Strength

Grip strength of both hands will be measured using a handgrip dynamometer (Jamar, JA Preston Corporation, New York, USA). Participants will be seated upright, with their elbows at 90° of flexion and their wrist in a neutral position. The handle of the dynamometer will be adjusted so that the base of the dynamometer is positioned with the rest on first metacarpal (heel of palm), and the handle rested on the middle of the four fingers prior to testing. Three trials will be performed and the maximum strength reached (in kg) will be used in the subsequent analysis. Grip strength is associated with concurrent overall strength, upper limb function and overall quality of life.42

Short performance physical battery

The Short performance physical battery is a valid short (5–10 min) test to measure lower extremity function using tasks that mimic daily activities. It examines three areas of lower extremity function, that is, static standing balance (three timed standing positions), gait speed (time to complete 4 m walk) and getting in and out of a chair (time to complete 5x sit to stand).43 The final score for each area is reported from 0 (inability to complete a given task) to 4 (highest level of performance). A sum score between 0 and 12 is then extracted.

6 min walk test (6MWT) and VO2

The 6MWT is a submaximal test used to assess endurance and functional capacity by measuring the maximal distance covered in a 6 min period.44 The test will be performed in a 15–20 m corridor and the distance will be measured with a measuring wheel.45 Before, during (every minute) and after the 6MWT, the perceived exertion level will be collected using the Borg Scale.46 Oxygen consumption (VO2) and metabolic cost of walking will be measured during the 6WMT using indirect calorimetry collected by a portable gas analysis system (Metamax 3b).47 The 6MWT scores are known to be correlated with quality of life.44 A minimal difference of 45 m between two performances is required to conclude on a real change in one’s condition.46 The 6MWT is validated and is reliable for multiple populations.44

Clinical Frailty Scale

The Clinical Frailty Scale (CFS) will be used to assess frailty on a scale of 1 (very fit) to 9 (terminally ill) according to domains such as comorbidity, function and cognition.48 As scores of 5 and above are associated with frailty, it is also used to predict the risk of poor outcomes which is higher with increased score.48 As of today, the CFS has been validated for older populations (65 years of age and over), but has not been widely validated in younger populations.48

Physical activity and Sleep monitoring

Participants will be asked to wear a fitness tracker watch (Garmin Forerunner 35) 24 hrs/day for seven consecutive days after each of the three in-lab evaluations. Outcomes that will be extracted are (1) for physical activity: number of steps, weekly number of minutes of moderate and intense activity; (2) for sleep: sleep time, total wake time and sleep efficiency and (3) for heart rate: average heart rate, resting heart rate and highest heart rate throughout the day.49 50 According to a systematic review, the validity and reliability of Garmin Forerunner for the number of steps is good to excellent, while its validity for heart rate is low to excellent.50 There are no available data on the psychometric properties of Garmin measures of sleep.

Sample size justification

Statistical analyses will be predominantly based on maximum-likelihood generalised analyses of variance (ANOVAs) (generalised estimated equation (GEE)). In this context, the required sample size would be approximately 60 participants per group.51 In this study, 120 participants per group is justified to permit exploration of secondary subgroup analyses for variables such as sex, gender and age. Attrition with time may also happen as some participants could, for example, experience exacerbation of their symptoms over time that could prevent them from participating to the in-lab assessment. Participants assigned to the control group could get infected with COVID-19 during their participation to the project. In those cases, the research team will assign them to a new group 12 weeks after symptom onset according to the nature and the duration of their symptoms (short or long COVID).

Statistical analysis

Descriptive statistics will be used to present the participants’ characteristics and the outcome measures at the different measurement times. Participants withdrawing from the study and reasons for withdrawal will be analysed.

Objective 1 will examine data at baseline. Notably, a sociodemographic and COVID history profile of the long COVID group will be presented according to sex, gender, hospitalisation, time since onset and number of comorbidities. Scores at baseline on self-reported and clinical variables will also be described and compared with the short COVID and control groups via a three-group one-way ANOVA. The type of ANOVA will depend on data distribution. The first choice is a maximum-likelihood ANOVA (SPSS, proc GENLIN). If not possible (eg, data not normally distributed, ordinal variables), the package rankFD of the R statistical software, a rank-based non-parametric equivalent with multivariate capacities, will be used.

For objective 2, repeated measures ANOVA will be used to compare the divergence in longitudinal performance variations across groups in the self-reported and clinical variables up to the 6-month follow-up (3 groups × 3 times). The first option is a generalised ANOVA for repeated measures (GEE), a maximum-likelihood technique. If not possible (eg, unstable distribution across groups or time), a rank-based non-parametric equivalent is offered by the nparLD package of the statistical R software. Both ANOVAs may work with non-normal distributions and unstructured covariance matrices between repeated measures. With these ANOVAs, missing data do not require to withdraw participants or impute data.

Ethics and dissemination

Ethics approval has been obtained from the Comité d’éthique de la recherche sectoriel en réadaptation et intégration sociale du CIUSSS de la Capitale-Nationale (Rehabilitation and Social Integration section; 2022-2328) and the Comité d’éthique de la recherche du CIUSSS de l’Est-de-l’Île-de-Montréal (2022-2811). As for dissemination, targeted materials will be developed for patients, clinicians, managers and decision-makers at the end of the project. It will include infographics, podcasts/webinars, presentations, and videos shared through several clinical and community platforms, including websites and social media of professional associations and stakeholders. Academic dissemination will include peer-reviewed manuscripts and conferences.

Ethics statements

Patient consent for publication

All the participants will be provided with a detailed information and consent form about the nature and objectives of the research project. They will have to sign this form prior to enrolment.