A prospective cohort study was set up at Reims University Hospital.
The inclusion criteria were: patients hospitalized for SARS-CoV2 infection during the health crisis between February 25th 2020 and April 30th 2020, who were aged 18 and over, having French national healthcare cover, and agreeing to participate in the study.
The Covid-19 diagnosis was established by a doctor if the patient tested positive with a PCR test, or on the basis of imaging or a clinical history pointing to this condition.
For all patients, socio-demographic data and medical history were collected at inclusion. Clinical, biological and therapeutic data during hospitalization was also collected, as well as data relating to the evolution of the disease, such as mortality, resuscitation, institutionalization or re-admission to hospital.
Data was collected up to 12 months after the initial hospitalisation.
Follow-up consultations at 6 months and 12 months were carried out by the clinicians who had initially provided care to the patients in the Reims university hospital Covid-19 units.
Data concerning clinical examinations, symptoms experienced by patients (dyspnoea, palpitations, asthenia, etc.) and treatments (anticoagulants, psychotropic drugs, etc.) was collected during the follow-up consultations at 6 and 12 months (M6 and M12).
Description and analyzes concerned only the 120 patients who attended the two follow-up consultations.
An evaluation of dyspnoea and related experiences among patients was graded at M6 and M12, using the modified Medical Research Council score (mMRC) . This score is generally used for patients with a probable pulmonary condition. It is scored on 5 levels from 0 “no dyspnoea except in the case of sustained effort” to 4 “dyspnoea preventing the patient from leaving the house or occurring when dressing and undressing”.
Fatigue was assessed at M6 and M12 using the Fatigue Severity Scale (FSS) . This scale is used to assess fatigue experienced by patients in their daily lives. It is composed of 9 items, each rated on an ordinal scale, from 1 “does not correspond to me at all” to 7 “corresponds to me perfectly”. An average score was calculated (by summing the different item scores and dividing by the number of items) with a global population standard of 2.3, a standard deviation of 0.7  and a threshold at 4 (with a score > 4 indicating great fatigue), as suggested by some authors [13,14,15]. To our knowledge, this score has been used in three other studies to evaluate the degree of subjective fatigue among patients with infectious diseases (Puumala virus, HCV or Covid 19) [16,17,18].
Quality-of-life was assessed using the Short Form 12 scale (SF-12)  at M6 and M12. This scale is an assessment carried out by patients themselves on their quality of life. It is a shorter version of the SF-36. The different items in this scale are grouped under two subscores:
The PCS 12 score: Physical Component Summary score, grouping items concerning patients’ physical state.
The MCS 12: Mental Component Summary score, grouping items concerning patients’ mental state.
A comparison with values observed in the general population was conducted using an abacus relating to a “normal” state in terms of Quality-of-Life in the French population .
Regulations and ethics
This study constituted research involving humans according to French legislation regulating research on human beings (Jardé legislation).
Approval from the French Research Ethics Board was granted. (No. 3838-RM). The study was registered on the database https://clinicaltrials.gov/ct2/home under No. NCT04553575.
All patients who agreed to take part in the study received full information and signed informed consent.
Qualitative variables were described as numbers and percentages, and quantitative variables were described via means or standard deviations, or as medians and interquartile ranges, depending on the variable distribution.
For comparisons of the scales and the scores between M6 and M12, Wilcoxon’s and McNemar’s tests for paired series were used. A p value < 0.05 was considered significant.
All analyzes were conducted using R studio® Version 4.0.5 and SAS Version 9.4 (Institute Inc., Cary, NC, USA).