Initially, 43 participants were included in the study, but four of them did not appear for the second measurement, leaving a total of 39 patients tested at both time points (T₁ and T₂). Participants were recruited from the pool of patients hospitalized at the Department of Internal Medicine of the Izola General Hospital, Slovenia, due to COVID-19 and its complications. Inclusion criteria for participation in the study were: age ≥18 years, signed informed consent, and hospitalization for COVID-19 disease [confirmed with a positive polymerase chain reaction (PCR) nasal swab test for SARS-CoV-2 virus]. Exclusion criteria were a positive PCR test for the SARS-CoV-2 virus at the time of hospital discharge and severe medical conditions (musculoskeletal, cardiovascular, pulmonary, and neurological) that prevented patients from performing all motor and cognitive tests. All patients were visited by a dedicated physician during their hospitalization, and the study protocol was discussed in detail with them. If the participant agreed to participate in the study, written informed consent was obtained prior to each examination. The clinical trial protocol was registered on ClinicalTrials.gov under the identifier number NCT04860206. All procedures were performed in accordance with the Declaration of Helsinki, and the study was ethically approved by the institutional ethical board of the Izola General Hospital (application number: 1/21).

Patients participated in T₁ measurements on the condition of a negative COVID-19 test, which was performed when one of the following conditions was met: on the day of hospital discharge or the 10th day after confirmation of the disease if discharged from the hospital earlier. The time frame for recruiting patients for T₁ measurements was between 01/01/2021 and 18/03/2021. After the T₁ assessment, the subject received a “Stay Active” brochure (22) with general information on the beneficial effects of physical activity and some comprehensive explanation of how different exercises could be performed at home during the COVID-19 pandemic. The T₂ assessment was conducted 2 months after the T₁ assessment (the time frame for T₂ measurements was between 11/03/2021 and 19/05/2021). All assessments were performed at the Cardiac Rehabilitation Center of Izola General Hospital during the morning hours.

All assessments at T₁ and T₂ were performed by a trained researcher in the same room and using the same equipment. Measurements were taken in this order:

Body mass (kg) and height (m) were measured with a Libela personal scale with a mounted stadiometer (Libela-Elsi Ltd., Slovenia), and the results were rounded to the nearest 0.1 kg and 0.5 cm, respectively. Body composition was measured using the tetrapolar bioimpedance device BIA 101 Anniversary (Akern-Srl, Florence, Italy) after the participants were in the supine position for 30 min. The proportions of fat mass (FM in %) and muscle mass (MM in %) were recorded from the assessment.

Tensiomyography measurements were performed in three muscles of the right leg: the vastus lateralis (VL), the biceps femoris (BF), and the gastrocnemius medialis (GM). All measurements were made during electrically evoked maximal isometric twitch contractions. For the VL, participants were in the supine position with the knee angle at 30° of flexion (where 0° represents a fully extended knee joint). For the BF, they were in the prone position with the knee at 5° of flexion, and for the GM, they were prone with the ankle in a neutral position, as previously reported (23–25). Foam pads were used for joint support. A single 1-ms maximal monophasic electrical impulse was used to elicit a twitch, which caused the muscle belly to oscillate and enlarge. These oscillations were recorded using a sensitive digital displacement sensor (TMG-BMC Ltd., Slovenia) that was placed on the surface of the skin over the mid-belly of the muscle of interest. If needed, the measurement point and electrode positions were adjusted to obtain the maximum amplitude (Dm) of the muscle belly response. Initially, the stimulation amplitude was set just above the threshold and then gradually increased until the Dm of the radial twitch displacement did not increase any further. From two maximal twitch responses, a contraction time (Tc), a delay time (Td), and a transversal velocity (Vc) were calculated, and the average was used for further analysis. Td was defined as the time from the electrical impulse to 10% of Dm. Tc was defined as the time for the amplitude to increase from 10% to 90% of Dm (25). Vc was calculated as the ratio between Dm and the sum of Td and Tc (26).

During the tensiomyography assessment, participants were interviewed to elucidate possible unhealthy lifestyle behaviors and to estimate their risk for sarcopenia and frailty conditions. For this purpose, we used the Slovenian version of the SARC-CalF questionnaire (27), the Mediterranean Lifestyle Index (MEDLIFE) (28), and the Edmonton frail scale questionnaire (29).

The SARC-CalF questionnaire is a screening tool for sarcopenia in older adults. It addresses five domains: strength, assistance in walking, rising from a chair, climbing stairs, and falling (30), and also includes a calf circumference measurement (31). Each answer is scored from 0 to 2 points according to the reported difficulty in performing the task in question. For calf circumference, zero represents normal muscle mass and 10 represents very low muscle mass. The sum of the points gives a score that can range from 0 to 20, with zero indicating the best result and 20 the worst. Individuals with a SARC-CalF score ≥11 are at increased risk for sarcopenia.

The Mediterranean Lifestyle index (MEDLIFE) is a tool that measures the adherence of the individual to the principles of the Mediterranean lifestyle (28). A total of 28 items are divided into three blocks of questions (food consumption frequency, Mediterranean dietary habits, and social habits). For each item, 1 point is awarded (28 points in total) if the answer meets certain criteria. The final score of the MEDLIFE index ranges from 0 (the worst) to 28 (the best).

The Edmonton frail scale (EFS) is a brief, valid, and reliable tool for assessing frailty. It can be administered by researchers without special training in geriatric medicine. The EFS assesses nine subscales: (1) cognition; (2) general health status; (3) functional independence; (4) social support; (5) medication use; (6) nutrition; (7) mood; (8) continence; and (9) functional performance. The EFS scores range from zero to 17. Severe frailty is defined as a score of 12–17, apparent frailty as a score of 6–11, and absence of frailty as a score of 5 or less (29).

The level of physical activity was determined with the Slovenian translation of the short version of the International Physical Activity Questionnaire (IPAQ), which was assessed at T₂ (32, 33). A total of seven items assess the frequency (days per week), duration (time per day), and intensity (light, moderate, or vigorous) of PA, which was performed during the previous week.

Blood pressure and pulse wave velocity were measured using a Vicorder (Vascular Model, SMT Medical Ltd., Germany). Prior to measurement, participants were placed in the supine position in a quiet room, with the head elevated to approximately 15°, so that the skin and muscles over the carotid artery were relaxed, but not too tense. Pulse wave velocity was measured with a cuff placed over the right carotid artery and the right thigh. The length between the common carotid artery (CCA) and the superficial femoral artery (SFA) was measured between the suprasternal notch and the midpoint of the thigh cuff. Measurements were taken until pressure waveforms across the CCA and SFA were clear and reproducible. During the FU, the same distance between the measurement sites was used. Blood pressure was measured in the sitting position with the cuff placed on the left upper arm.

Spirometry was performed with the participant comfortably seated in a chair. A clip was used to close the nostrils. Participants were instructed to breathe normally for 10 s and then to inhale as deeply as possible and exhale as quickly as possible into a tube (microQuark, COSMED Ltd., Italy). The best forced vital capacity (FVC) and forced expiratory volume measurement in one second (FEV1) of the three tests performed were used. An FEV1/FVC Tiffeneau-Pinelli index was calculated automatically (34).

The grip strength test (35) was performed using an analog hydraulic handheld dynamometer (Jamar Dynamometer, Sammons Preston, USA). During the measurement, participants sat on a chair with no armrests. The dominant upper arm was parallel to the torso, while the elbow was positioned at 90° flexion. After two test attempts, participants performed three consecutive handgrips with a 1-minute rest in between. The best result was used for further analysis.

A “Timed Up and Go” test (36) was used to assess mobility at a distance of three meters. Participants wore their regular walking shoes. The test started with the participant sitting on a 46-cm-high chair, then standing up, walking around a marked bar, returning to the chair, and sitting on it. Each participant completed one test trial and two trials without walking aids. The trial with the shortest completion time was used for further analysis.

Gait speed was evaluated (37) at the self-selected speed and the fastest speed over a distance of 4 meters using timed gates (Beam Trainer timing system, Seedgrov d.o.o., Ljubljana, Slovenia). Two meters were provided before and after the timed distance for acceleration and deceleration. Each gait modality (self-selected speed and fast speed) was assessed twice, and the best result was used for further analysis.

A 30-second chair-stand test was used to assess the number of stands a participant could complete in 30 s (38). Participants initially sat on a 46-cm-high chair without armrests. They placed their hands on the opposite shoulder, crossed at the wrists. Their feet were flat on the floor, and their backs were straight. They rose to a full standing position, then sat back down again and repeated this for 30 s. Only completed repetitions were counted.

The Montreal Cognitive Assessment (MoCA) was used to evaluate the general level of cognitive functioning and screen for cognitive impairment (39). The MoCA test addresses several cognitive domains, namely visuospatial ability, short-term memory, executive function, attention, concentration, working memory, language, and orientation to time and place. The final score ranges from 0 to 30 points, with values ≥26 indicating no cognitive impairment.

A structured interview was used to assess the self-perceived response to the COVID-19 experience in relation to functional, mental, and mood states. The structured interview was used to capture qualitative data. Patients were interviewed twice, at the first measurement and then at the second measurement after self-rehabilitation at an interval of 8 weeks. The interview was conducted after the functional tests between the patient and the researcher, who took notes on the answers.

The first structured interview included questions about the subjective experience of being hospitalized for COVID-19, i.e., what was most distressing for the respondent besides the physical problems: how did they help themselves or what helped them, and what did they miss most? They also rated their physical and mental states (in percentages) after the infection, assuming that their pre-infection state was rated as 100%. At the end of the interview, the patients were given a form to monitor their symptoms or changes in well-being and were asked to bring it to their post-rehabilitation follow-up appointment. Patients also received a “Stay Home—Be Active” manual with instructions on how to begin the exercise program designed for patients with chronic lung disease.

At T₂, patients were asked to compare their physical and mental state and mood again, assuming that their pre-infection state was rated as 100%. They were also asked to evaluate in more detail what they found most difficult in coping with the physical and mental stress, whether they were frustrated, what thoughts came to mind during the recovery period, how they helped themselves if necessary, and what they missed during the recovery period or how it could have been more successful. They also assessed their perception of what it would take to reach 100% physical, mental, and emotional health and when they expected to achieve it. Finally, they were asked to hand over the symptom monitoring form they received at the first measurement, and if they did not bring it with them, they were interviewed by the researcher to collect the data. They were also asked if they had followed the advice in the manual and if they had exercised according to its instructions.

We used G-Power (40) to determine the required sample size. Considering a two-sided α-value of 0.05 and a β-value of 0.20, and a functional decline of less than −11% (indicating an effect size greater than 0.9), we calculated that a sample size of 13 participants would be sufficient. All parameters are presented as the mean and standard deviation (SD). The normal distribution was confirmed by visual inspection using histograms and Q-Q plots, and analytically using the Shapiro-Wilk test. Parameters of the baseline (T₁) and follow-up (T₂) samples were assessed by a paired samples t-test. All statistical analyses were performed using IBM SPSS Statistics 22 (SAS Institute, Cary, NC, USA), with a significance level of p < .05. Figures 1–3 have been prepared using Microsoft Office Excel (Microsoft Corporation, Washington, USA).

During the two-month self-rehabilitation period, patients were asked to record any symptoms and changes in well-being. Only two patients returned a completed form with symptoms recorded. In addition, we were able to collect recalls of symptoms from 18 other patients at follow-up. Thus, in total, we recorded symptoms and changes in well-being in half of the patients who participated in the study (20 out of 40 participants) (Figure 3). Our data also show that the most common symptoms after COVID-19 are joint pain (10 patients), muscle pain, and cough (seven patients), followed by irritability, sadness, shortness of breath, and difficulty sleeping.

We were interested in whether participants changed various lifestyle characteristics between T₁ and T₂. Nutritional and physical activity patterns should be changed after the COVID-19 hospital discharged to overcome negative consequences of disease. The self-reported amount of physical activity during self-rehabilitation period, on average, 5.2 h peer day 5.4 h per day of sitting. It has been suggested that COVID-19 may result in physical inactivity, which in turn may become a long-lasting symptom for individuals who have recovered from the virus (42). On the other hand, regular physical activity can have an important impact on different aspects of health, from cardiovascular to mental health (43). Another behavioral factor affecting COVID-19 rehabilitation is nutrition. Adequate nutrition can play a protective and regenerative role, according to COVID-19 (44). Because of its protective and preventive effects, it has been suggested that the Mediterranean diet may represent a positive approach against COVID-19 (45, 46). In our sample, we did not observe any changes in eating habits or lifestyle in favor of a Mediterranean diet between T₁ and T₂. Compared with other studies (47), our population had a comparable adherence to the Mediterranean lifestyle.

The Edmonton Frail scale and SARC-CalF test questionnaires were used to measure changes related to frailty or sarcopenia (27, 29). COVID-19 patients tend to be more frail after hospital discharge (48). Moreover, early detection of frailty and sarcopenia may help in the future planning of specific rehabilitation. We were interested in the frailty and sarcopenia status of the enrolled participants after COVID-19. At T₁ measurements, two patients had severe frailty according to the Edmonton scale of frailty, three patients had a moderate frailty score, nine patients had a predisposition to frailty, and 25 patients showed no signs of frailty. The SARC-CalF test for sarcopenia showed that one patient had potential sarcopenia on T₁ measurements. Two months after discharge, the frailty results improved, and no patient had severe frailty, while five patients exhibited signs of moderate frailty, and the remaining patients were not at risk for frailty. The SARC-CalF test for sarcopenia after two months showed that the patient no longer reported problems suggestive of sarcopenia. Therefore, the screening instruments (Edmonton Frail scale and SARC-CalF test), which are based on self-reporting and self-assessment, showed improvement in various aspects such as strength, assistance in walking, general health status, and functional independence.

Despite positive results at T₂ for the Edmonton Frail scale and the SARC-CalF test, we did not find improvements in physical performance tests. As previously described in the literature, patients may have impaired physical performance after hospital discharge (49). Re-evaluation after two months of self-rehabilitation showed that participants did not achieve the results expected for a healthy population in selected functional parameters such as the 30-sec Sit-to-stand test, TUG, gait speed, grip strength, and clinical parameters such as systolic and diastolic blood pressure and FEV1/FVC. Due to the complexity of rehabilitation after COVID-19, bed exercises can be an effective tool in the initial stages. An example of bed exercises is the Full-body in-Bed Gym program (50, 51), which allows the difficulty of the exercises to be adjusted to the individual's physical and mental capabilities. Thermal water rehabilitation is also a possible suggestion for an effective rehabilitation program tailored to the individual. This type of rehabilitation can improve the inspiratory muscles, which are typically weakened after COVID-19 (52).

Patients had negatively deviated levels of some risk factors compared with recommended levels shortly after discharge. Factors such as increased BMI (53, 54) and higher systolic blood pressure (55, 56) have been previously identified as factors that may contribute to complications of COVID-19. In addition, it has also been suggested that COVID-19 increases systolic and diastolic blood pressure and may cause new-onset hypertension (57), but we did not find correlations between previously reported hypertension and elevated systolic blood pressure in our patients. Therefore, we can assume that the elevated systolic pressure in our patients is somehow related to the COVID-19 consequences.

The literature also suggests changes in mental health status after hospitalization for COVID-19. Despite the anxiety and fear the respondents have experienced and, above all, the desire to regain their pre-infection health status shown on the first measurements at discharge from hospital care, two-thirds of the patients showed no significant improvement in the measured psychological parameters. At the T₁ measurements, like Brown et al. (62), we noted a willingness to try anything to address symptoms after COVID-19, as subjects showed interest in exercising with the manual and in monitoring their health. Although they reported a severe experience with COVID-19, in addition to anxiety and health concerns, their lifestyle habits did not change after they went home, i.e., no systematic self-rehabilitation was reported at the T2 measurements.

The subjects' self-assessment of physical and mental status (index) showed mainly an improvement in physical status, where the assessment of mental status remained unchanged, while some patients (5%) still didn't reach the pre-Covid state, which may indicate that the mental (mood) consequences are lasting longer what confirmed also by the MoCA test. Despite an average higher MoCA score in the whole sample, 15 out of 39 patients had worsened or remained unchanged at 2 months. Nevertheless, patients' adherence appeared to be very low, as they admitted that they did not follow the manual as advised at the initial measurements. How to encourage patients to take a more active approach to rehabilitation and how to increase adherence will certainly be our future challenges.

Improvements in the MoCA test are indicative of recovery of cognitive function in hospitalized COVID-19 patients at 2 months after hospital discharge. The original MoCA validation study reported a test-retest consistency of 0.91 at 2 months, with no significant learning effect (63); however, the indication of improved performance following repeated MoCA administrations (64) stresses the importance of interpreting the results with caution. It has already been established that patients who are critically ill or who have been treated in an intensive care unit (ICU) are at risk of suffering from the long-term consequences of COVID-19, such as impaired physical and cognitive function and psychological disorders similar to post-intensive care syndrome (PICS). Therefore, optimally selected rehabilitation may be required to address these disabilities (65). In addition to the significance of the results indicating overall recovery, it is important not to overlook the 15 (38.5%) participants whose cognitive status either worsened or remained unchanged from post-hospitalization to 2-month follow-up. These participants may be at higher risk or more susceptible to long-term post-infection difficulties and may require additional attention during recovery, if successful. A comprehensive report on the cognitive status of the patients enrolled in this study has been published previously (66).

The single-center design and relatively small number of patients are potential limitations of this study and may limit the generalizability of the results. The severity of the infection was not considered, and no assessment was performed before or during the infection, so changes may not be solely attributable to the infection.