Early motor and respiratory re-education in patients hospitalized for COVID-19

INTRODUCTION

At the onset of the Coronavirus Disease 2019 (COVID-19) pandemic, health services around the world had to reinvent themselves to cope with a new unknown emergency¹. The COVID-19 pandemic presented significant challenges for healthcare systems worldwide, requiring hospitals to adapt and develop comprehensive care strategies for patients affected by the disease. In Italy, the Tor Vergata University Hospital (TVUH) was converted into a COVID-19 Hospital and dedicated beds were established².
During the acute phase of COVID-19, depending on severity and comorbidities, some patients may exhibit a range of debilitating symptoms, including respiratory distress, muscle weakness, fatigue, and impaired functional capacity³. This results in prolonged hospital stays, in intensive and non-intensive care areas, and prolonged periods of bed rest. During such periods, the stay is complicated by the need to maintain invasive and non-invasive ventilation therapy, hypoxic brain damage in older individuals, and profound asthenia. Moreover, isolation limits the movement space of the patient, who may already be debilitated by fever, fatigue and muscle pain. This can result in reduced muscle strength and insufficient drainage of sputum; it can also result in a significantly increased risk of deep vein thrombosis, pressure skin lesions, as well as psychological problems such as anxiety, depression, and lack of motivation⁴.
Survivors of acute respiratory distress syndromes, including those not associated with COVID-19, can present persistent weakness up to one year after hospitalization, with functional implications and reduction of Quality of Life (QoL)⁵. Lower muscle strength and higher mortality can still be evident even at five years follow up compared to the general population⁶. Patients admitted with SARS-CoV-2 infection are particularly fragile and require a well-designed Individual Rehabilitation Project (IRP)^7–10. Such symptoms may as well persist long term after discharge, particularly involving respiratory function, implying a decrease in QoL^11,12.
The complications caused by prolonged immobilization, such as reduction in muscle mass and muscle strength, progressive bone demineralization, increased thromboembolic risk, increased cardiovascular risks, and reduced respiratory capacity are reduced by early mobilization^13,14. Active mobilization, defined as voluntary movement performed by the patient, has been shown to significantly enhance muscle strength and functional outcomes^15,16. The physiological rationale behind this is that active engagement of muscle fibres stimulates protein synthesis and helps maintain muscle mass by counteracting the atrophy typically associated with disuse and immobilization¹⁷. In addition, passive mobilization, consisting of movements performed on the patient by a therapist or through mechanical means, also plays a vital role in muscle preservation¹⁸. Research has demonstrated that passive mechanical loading can attenuate muscle mass loss and maintain force-generating capacity during periods of immobilization, probably due to reduced oxidative stress and a decrease in the loss of myosin^15,19. Passive stretching and repetitive movements have also been shown to enhance muscle mass and cross-sectional area, particularly in populations at risk of sarcopenia; these are believed to act by activating the Akt/mTOR signalling pathway, which regulates muscle protein synthesis²⁰. Furthermore, cyclic passive mobilization has been associated with enhanced muscle metabolism, which is vital for maintaining muscle health during periods of reduced activity²¹. Thus, while active mobilization is renowned for its effectiveness in preventing muscle mass reduction, the integration of passive mobilization into an IRP can serve as a crucial strategy for preserving muscle mass and function in vulnerable populations.
Consequently, early re-educational interventions based on a well-defined IRP appear as a fundamental component of multidisciplinary care, aiming to mitigate the detrimental effects of prolonged immobilization^3,22. Therefore, such rehabilitation aims to enhance respiratory function, counteract musculoskeletal deconditioning and immobility, minimize complications, restore cognitive and emotional well-being, reduce disability, and enhance patient QoL during both the acute and post-critical phases, with the ultimate goal of facilitating an eventual return home^23,24.
COVID-19 causes an obstruction in the small airway, reducing lung capacity and respiratory muscle strength from early in the convalescence phase; abnormalities in muscle mass and reduced muscle strength have also been reported ^25–28. Early rehabilitation care (i.e. motor and respiratory re-education) limits the burden of respiratory disease and reduces the impact on QoL in different settings^29,30. Therefore, implementing a comprehensive IRP in an acute setting may be of help for patients with COVID-19 in the early phase^31–33.
During the pandemic period, all clinical activities in TVUH were re-organized to assist patients hospitalized for the most severe consequences of COVID-19. Rehabilitation was no exception and IRPs, i.e. functional re-educational programs, were also adapted; these were planned and carried out by a multidisciplinary team consisting mainly of physicians specialized or residents in Physical and Rehabilitation Medicine (PRM) and physiotherapists (PT).
The IRPs were performed as respiratory or motor re-educational programs, or both, whose goals were mainly to prevent damage due to hypomobility, to recover lost functions and to reduce the burden of returning home after hospitalization.
The purpose of this study was to evaluate the effects of IRP implementation in patients with COVID-19 in the acute phase within a dedicated setting. The effects of early rehabilitation on strength, according to the Medical Research Council (MRC) scale, and dyspnoea, based on the Barthel Dyspnea Index (BDI), were considered as main outcome measures; in addition, their effects on patient independence in activities of daily living (ADL), based on the Modified Barthel Index (MBI), were used as secondary outcomes.

MATERIALS AND METHODS

This retrospective observational study was based on data obtained from patients admitted to the Department of Infectious Diseases at TVUH from March 2020 until July 2022.
The clinical protocol was conducted, recorded, and reported according to Good Clinical Practice guidelines and the Declaration of Helsinki. The study itself was approved by the Lazio Area 2 Ethical Committee (262.24). Before data was collected, all participants completed an informed consent form.

Inclusion and exclusion criteria

The study included patients who tested positive for SARS-CoV-2 via nasopharyngeal swab and were prescribed either a motor re-educational program, a respiratory re-educational program, or both. Additionally, eligible patients must have undergone at least two PRM evaluations, with the final evaluation conducted at hospital discharge. Patients were excluded if they had not been assessed using functional evaluation scales or were not subjected to an IRP.

Assessment and Outcome Measures

All participants had undergone an initial PRM assessment (T0), re-educational intervention (I), and PRM reevaluation (T1) at the time of discharge.
The initial assessment (T0) consisted of a clinical and functional evaluation based on PRM. Respiratory status, cardiovascular status, ability to move, and level of independence in ADLs were evaluated. Following the assessment, the PRM specialist designed a suitable IRP consisting of a motor re-educational program, a respiratory re-educational program, or both.
As per our clinical practice, each patient received an individually-designed IRP; these were designed while considering specific physical, clinical and functional characteristics that could impact training, such as dyspnoea, low oxygen saturation, and motor impairments. The re-educational programs could be performed in various positions, including active or passive mobilizations. Supine exercises, requiring minimal energy expenditure, were ideal for patients with low oxygen saturation and poor trunk control. Seated exercises suited patients with hypotension but good trunk control. As clinical and functional status improved, standing exercises were introduced. The re-educational intervention (I) consisted of 45–60-minute sessions, conducted five to six times a week, in line with the institution protocols.
The motor re-educational program included therapeutic exercises aimed at improving mobility and functional independence, such as passive, assisted, or active limb mobilization; strengthening of the upper and/or lower limbs; training in postural transitions (e.g., sitting to standing); trunk and pelvic stabilization exercises; and walking/gait training with assistive devices if required. Concurrently, the respiratory re-educational program focused on enhancing pulmonary function through techniques like breath-movement coordination, diaphragmatic relaxation maneuvers, bronchial drainage exercises (e.g., postural drainage), lung expansion strategies (e.g., deep breathing), and respiratory muscle training to optimize airflow and endurance. Both programs were tailored to individual patient needs, with components selected based on clinical assessments.
The PRM reevaluation (T2) examined the clinical and functional status of the patient at discharge, i.e. after the re-educational phase. Depending on the progress in the clinical condition and the abilities recovered, the patients were either discharged or transferred to a post-acute rehabilitation facility; moreover, further aids or devices needed to facilitate ADLs (e.g., walking, bed/chair transfer) could be prescribed at this stage.
Functional status was evaluated based on the MRC scale for muscle strength, the MBI for dependence in ADL, and the BDI for dyspnoea perception. Peripheral oxygen saturation (SpO2) was also measured at the beginning and end of each session.
The MRC is a six-point scale (0 to 5) developed by the Medical Research Council (MRC) of Great Britain in 1942; it is used for measuring strength through observation of movement and muscle behavior³⁴. While the scale is simple to administer, due to its manual administration, it is difficult to use the result to objectify (for grades 4 and 5) muscle strength evaluation³⁵.
The Barthel Index (BI) scale is used to measure performance in ADLs, with the aim of defining degree of independence in basic activities. In the original version, the BI examines independence in 10 ADLs such as eating, clothing (dressing/undressing), personal hygiene, bathing/showering, bowel sphincter control, bladder sphincter control, chair/bed transfers and vice versa, toilet use (getting on and off the toilet), mobility (walking on level ground), and climbing and descending stairs. The score ranges from 0 to 100, with the latter indicating complete independence. The score assigned for each task can be 15, 10, 5, or 0. The maximum score is awarded only if the patient performs the task independently, without assistance. Lower scores indicate dependence in performing ADLs³⁶.
The Modified BI (MBI)³⁷ was developed to better define the functional situation of a patient and staging their level of dependence and need for assistance. The activities to be examined are the same of the original BI, but in different order and with the addition of the item wheelchair use if the person is not ambulatory. Each activity can be graded as 0,1,2,3,4,5,8,10,12 or 15, depending on functional status.
The BDI³⁸ measures perceived level of dyspnoea in performing basic ADLs, ranging from 0 to 4. The combined administration of the MBI and the BDI is useful in defining multifactorial disability, i.e. both motor issues and those related to the impact of dyspnea, and the corresponding components to be included in different re-educational programs.

Statistical Analysis

All data were initially entered into an Excel spreadsheet (Microsoft, Redmond, WA, USA). Data analysis was carried out using IBM SPSS Software (version 29). Descriptive statistics were calculated, including measures of central tendency and their dispersion ranges; mean ± standard deviation (SD) were used to describe parametric data, and median with interquartile ranges (IQRs) for non-parametric data.
For continuous data (i.e. SpO₂), the normality of the data distribution was assessed using the Shapiro-Wilk test due to the moderate sample size (n = 52); the result indicated a significant departure from normality [W (52) = 0.87, p < 0.001]. The remaining outcome measures (i.e. MRC, MBI, BDI) are categorical ordinal variables and were subjected to the appropriate tests. Consequently, Wilcoxon's signed-rank was employed to assess significant changes within all variables between T0 and T1. Statistical significance was set at p < 0.05.
The patients enrolled in this study were further divided into two subgroups: the first consisting of the patients who underwent only a motor re-educational program (motor program group, MPG) and the other including patients who performed both motor and respiratory re-educational programs (motor-respiratory program group, MRPG). The two subgroups were then compared using the non-parametric Mann-Whitney independent-samples U-test. This non-parametric test was chosen because the data did not meet the assumption of normality and were measured on an ordinal/continuous scale.
Covariate analysis was then carried out with linear regression.

RESULTS

The study included data from 52 patients (24 male, 46.1%) with a mean age of 74.9±12.4 years. The mean period of hospitalization was 25.1±13.75 days. The most commonly-represented comorbidities were hypertension (57.7%), type II diabetes mellitus (26.9%), chronic obstructive pulmonary disease (19.2%), chronic atrial fibrillation (19.2%), obesity (13.5%) and chronic renal failure (11.5%). The mean duration of IRP was 12.38±13.75 days.
Upon discharge, all patients were reevaluated by the PRM specialist. Three patients were found to be fully autonomous in walking and ADLs and, therefore, did not receive any prescriptions for aids; however, two of these three patients demonstrated persistent dyspnea, and needed to continue rehabilitation through the activation of home rehabilitation assistance care for respiratory re-education.
All other patients were discharged home and prescribed home rehabilitation care for motor re-education or aids best suited for their clinical-functional status, such as walkers and wheelchairs. For only one of the 52 assessed patients, an indication was placed for transfer to a post-acute intensive rehabilitation facility for continued rehabilitation.

MRC – Muscle Strength

Median muscle strength, assessed using the MRC scale, was 3.75 (IQR: 3–4, range: 2–5) at T0 and 4 points (IQR: 4–4, range: 3–5) at T1. This difference was found to be significant (W = 0, n = 52 pairs, p < .001; Wilcoxon Signed-Rank Test), with a large effect size (r = 0.57). Twenty-one patients improved their muscle strength at T1 compared to T0, while 31 participants showed no change. The MRC rank summary for the Wilcoxon Signed-Rank Test is presented in Table 1.

MBI – Level of independence in ADLs

Median independence in ADLs was found to be 25 (IQR: 18–33, range: 2–65) at T0 and 35 (IQR: 21.5–65, range: 5–100) at T1, assessed using the MBI scale. This difference was significant (W = 10.5, n = 52 pairs, p < .001; Wilcoxon Signed-Rank Test), with a large effect size (r = 0.76). Of the 52 patients enrolled, almost all showed an increase in MBI scores. Only three patients showed a decrease between T0 and T1, while eleven patients showed no change. The MBI rank summary for the Wilcoxon Signed-Rank Test is presented in Table 2.

BDI - Perceived level of dyspnoea during ADL

Median perceived dyspnoea increased from 51 (IQR: 40–65, range: 13–90) at T0 to 90 (IQR: 55–90, range: 25–100) at T1, assessed using the BDI scale. This difference was significant (W = 0, n = 52 pairs, p < .001; Wilcoxon Signed-Rank Test), with a large effect size (r = 0.70). Of the 52 patients enrolled, 33 showed an increase in BDI scores, while 19 showed no change. No patients showed a decrease in scores between T0 and T1. The BDI rank summary for the Wilcoxon Signed-Rank Test is presented in Table 3.

Oxygen Saturation – SpO₂

Median oxygen saturation (SpO₂) fell slightly from 98 (IQR: 96–99, range: 90–100) at T0 to 97 (IQR: 95–98, range: 92–99) at T1. This difference was not statistically significant (W = 321, n = 52 pairs, p = 0.065; Wilcoxon Signed-Rank Test), with a small effect size (r = 0.26). Of the 52 patients enrolled, 29 showed a decrease in SpO₂ scores, 14 showed an increase, and nine showed no change. The SpO₂ rank summary for the Wilcoxon Signed-Rank Test is presented in Table 4.

Subgroup Analysis

Of the 52 patients enrolled, 20 (38%) underwent exclusively a motor re-educational program and fell in the MPG, while the remaining 32 (62%) were treated with both motor and respiratory re-educational programs and fell in the MRPG. The Mann-Whitney independent-samples U-Test (Table 5) revealed no significant differences across all the outcome measures analyzed: MRC (U = 273.5, p = 0.339, r = 0.16), MBI (U = 318, p = 0.897, r = 0.02), BDI (U = 317.5, p = 0.889, r = 0.02), and SpO2 (U = 295.5, p = 0.584, r = 0.08) (Table 5).

Covariate Analysis

The regression analysis revealed that the following parameters together explained 35.91% of the variance of final MRC scores (p = 0.002), 45.39% of the final MBI scores (p < .001) and 38.62% of the final BDI scores (p = 0.001): initial respective rating scale scores (i.e. MRC, MBI, BDI), age, cardiologic and respiratory comorbidities (e.g. history of myocardial infraction, hypertension, chronic obstructive pulmonary disease, pulmonary fibrosis), diabetes mellitus type 2 and obesity. These parameters only accounted for 4.37% of the variance of the final SpO₂ scores, which was not statistically significant (p = 0.911).
The initial MRC scores alone explained 33.44% (p < 0.001) of the variance of the final MRC scores, 15.4% for final MBI (p = .004) and 16.55% for final BDI (p = 0.003). When age was added to initial MRC, the results increased to 33.62% (p <0.001) for MRC, 16.48% (p = 0.012) for MBI, 18.43% (p = 0.007) for BDI.
The initial SpO₂ scores alone explained 6.11% (p = 0.077) of the variance for final MRC scores, 6.58% for final MBI (p = 0.066) and 10.43% for final BDI (p = 0.019). When age was added to initial SpO₂, the results increased to 6.32% (p = 0.202) for MRC, 7.12% (p = 0.163) for MBI, 11.64% (p = 0.048) for BDI.

DISCUSSION

The study aimed to evaluate the effects of early rehabilitation on patients with COVID-19 in the acute phase within a dedicated setting. The main outcome measures were muscle strength (MRC scale), perceived level of dyspnea (BDI scale), and patient dependence in ADL (MBI scale); SpO₂ was also evaluated as an objective measure of pulmonary function. A statistically significant improvement was found in MRC, BDI and MBI score, indicating a substantial improvement in muscle strength, perceived level of dyspnea and independence in ADLs. Notably, weaker findings emerged about SpO₂. However, it should be considered that all subjects enrolled in the study were on oxygen therapy or non-invasive mechanical ventilation at T0, while most (29.56%) were eupneic on room air at discharge.
The applied re-educational programs consisted of therapeutic exercises aimed at improving motor and respiratory performance. Depending on the impairment and disability level presented by each patient, a wide variety of therapeutic exercises could be implemented, such as: active and passive mobilizations of the upper and lower limbs, trunk control exercises, proprioception exercises, strengthening exercises, reaching and/or maintaining a sitting or standing position (independently or with assistance), and gait training with or without aids (such as canes, walkers, and axillary crutches). In addition, patients with low coordination between motor and respiratory skills could participate in exercises to improve breath-movement coordination and reduce apnea.
The positive effects of such therapeutic exercises are well known^39,40. The PRM section of the European Union of Medical Specialists (UEMS) states that a comprehensive IRP should be person-centered and interdisciplinary, and should put the focus on the individual; thus rehabilitation is tailored to the needs and goals of the specific person for whom the IRP is developed⁴¹. Consequently, the IRP protocols applied in the present study differed from patient to patient according to their clinical and functional status. Indeed, the development and implementation of individualized treatments is one of the strengths of our work, since these made it possible to address the patient needs in a focused way; in contrast, a standardized program could over or underestimate performance, leading to poor results or overload^42,43. Another strength is the combination of a motor program with a respiratory program, and the overall evaluation of their effects.
Rehabilitation has shown positive effects in several conditions and settings, including respiratory conditions^44–50. Our findings confirm previous observations regarding the positive effects of early re-educational interventions on COVID-19 patients. Several authors have already addressed the effects of early mobilization in such populations. For example, a Brazilian study by Bonorino et al.⁵¹ examined the effects of early intervention programs with early mobilization in patients with COVID-19, especially in Intensive Care Unit (ICU).
Our results are consistent with other studies indicating how inpatient re-educational programs lead to improvements in physical performance status, the ability to perform ADL tasks and pulmonary measures. One retrospective paper on 100 patients found these programs to achieve significant functional, motor, and respiratory improvements. Similarly to the present article, the clinic created dedicated beds for COVID-19 patients, who received re-educational treatment in their rooms. However, while the present study only employed one session per day of 45-60 minutes, the previous paper separated treatments into two different slots of less than 20 minutes; it also included a motor program associated with a respiratory program, as well as with individualized therapeutic exercises⁵².
Saeki et al. also reported an increase in MBI in a case study of a single patient who received positioning and postural drainage re-educational while under mechanical ventilation, followed by gait and endurance training; notably, they also note significant improvement in MRC scale, which is consistent with our results⁵³.
Findings similar to ours were also obtained in two more studies conducted in 2021. The first, a prospective study by Puchner et al., analysed 23 patients who underwent a personalized multidisciplinary IRP; the authors observed significant improvement in mean lung functionality and in the six-minute walking test (6MWT), which could also be considered a measure of independence ⁵⁴. The second is a retrospective study performed on an Italian population but in a different setting to the present study. Briefly, Curci et al. subjected a group of 41 patients to two re-educational sessions of 30 minutes per day; similarly to the present study, MBI and 6MWT scores were found to improve significantly. Significant changes in perceived level of effort and dyspnea were also noted, but using a different scale (Borg Rating of Perceived Exertion) ⁵⁵.
Interestingly, Liu et al ⁵⁶. report how a six-week respiratory program can improve respiratory function and QoL outcome in patients with COVID-19 compared to a control group. Notably, the implementation of an IRP was found to increase anxiety scores but have little effect on depression outcomes. These findings further corroborate the role of re-educational treatment in this population.
The effects of combining motor and respiratory re-education were evident in our population. Our findings indicate significant improvements in the perceived level of dyspnoea, as assessed with the BDI, and muscle strength, according to the MRC scale. However, no statistically significant improvement was found regarding median SpO2 values, falling slightly from 98% to 97%, both values being in the normal range. It is noteworthy that 56% of the patients involved in the study were weaned from oxygen therapy; this result should be considered as a very positive change. Regarding long-term functional outcomes, the implemented IRPs and the early re-educational programs achieved significant increases in autonomy in ADL, as measured by the MBI.
The correlation between improvements in physical performance and ADL status is a significant area of research. Numerous studies have established that enhanced physical performance contributes to better ADL outcomes, thus highlighting the importance of physical activity interventions. Dominguez et al. indicate that exercise interventions, including strength and balance training, lead to notable improvements in ADL performance in individuals with moderate to severe dementia⁵⁷. This is corroborated by research that emphasizes the role of physical performance in facilitating independence in ADL, suggesting that interventions aimed at improving physical capabilities can yield substantial benefits in functional outcomes⁵⁸. This is further supported by evidence indicating that improvements in muscle strength and physical performance are directly associated with enhanced ADL outcomes, particularly in older adults experiencing sarcopenia or other declines in physical function⁵⁹.
Our covariate analysis revealed that initial MRC scores alone explained 33.4% of final MRC variance, highlighting baseline muscle strength as the strongest predictor of recovery. Age and comorbidities added minimal explanatory power (total 35.91%), suggesting that other factors (e.g., nutrition, inflammation severity, or genetic factors) may play significant roles. This aligns with a report by Gil et al., which found that muscle strength seems to predict recovery in patients hospitalized with moderate to severe COVID‐19, stressing the value of muscle health in prognosis of this disease⁶⁰. Early identification of low baseline strength could prioritize intensive rehabilitation to mitigate further decline.
Key factors for independence in ADL were baseline MBI and comorbidities. Age contributed modestly, reflecting how pre-existing functional status and chronic conditions (e.g., diabetes, hypertension) disproportionately affect recovery; this is in agreement with previous findings⁶¹. This supports the notion that tailored interventions for patients with comorbidities can improve ADL outcomes. Baseline BDI, age and comorbidities explained 38.6% of variance of final BDI scores. Notably, initial SpO₂ became a weak but significant predictor when combined with age for BDI (11.6% variance explained, p = 0.048), suggesting older adults with lower baseline oxygenation may experience worse dyspnea despite rehabilitation. Older adults tend to show diminished ventilatory response to hypoxia and hypercapnia, making them more vulnerable to ventilatory failure during vulnerable conditions such as pneumonia and possible poorer outcomes⁶². Thus, respiratory interventions should prioritize older adults with low SpO₂ to reduce dyspnoea-related disability. Only 4.37% of SpO₂ variance was explained by initial values, age and comorbidities. This aligns with the lack of any significant SpO₂ changes, emphasizing that functional gains (e.g., strength, ADL) occur independently of oxygenation metrics. All this considered, baseline function (MRC/MBI/BDI) and comorbidities should guide IRP design. Moreover, improvements in strength/ADL outcomes without significant SpO₂ gains suggest functional recovery is achievable even with persistent oxygenation challenges.
Our experience highlights the positive impact of an IRP addressing both respiratory and physical impairments. Early re-educational interventions during the acute care phase can influence the clinical and functional condition at the time of discharge. Furthermore, this approach has the potential to enhance independence in ADL and overall QoL. Lastly, it is important to consider the reduction in health care costs associated with reduced disability, particularly in a population as vulnerable and delicate as patients with COVID-19^63,64.
The usefulness of early re-education for patients with COVID-19, and the positive effects it can also exert in terms of long-term improvement, are also confirmed by recent reviews and meta-analyses^65,66. Even during periods of low pandemic intensity, the focus on rehabilitation and re-educational treatments for patients with COVID-19 must remain high: particularly when the primary focus is on recovery from long COVID⁶⁷.

Study limitations

This study does not come without limitations. The first is the relatively small sample size, which may have made it impossible to reach statistical significance in subgroup analysis. In addition, due to the retrospective nature of the study, the patients were evaluated in only a limited number of areas. It was also not possible to perform a standardization of the interventions proposed to the study group; however, this eventuality is necessary in order to fit the IRP to the individual patient. Furthermore, it was not possible to acquire accurate data on the precise duration of the re-educational intervention; however, it should be considered that this is a retrospective study, with interventions carried out in the pandemic period, and acquiring data on the precise duration of the re-educational programs was not considered a priority at the time. Nonetheless, as per our institutional protocol, to all patients was administered a re-educational intervention with a timing ranging in 45 to 60 minutes. Finally, it was not possible to include a control group; however this is understandable, as no patients hospitalized for COVID-19 in our institution were left without rehabilitation assistance for obvious clinical and ethical reasons.

CONCLUSION

In conclusion, COVID-19 still presents a challenge for national health systems, which must provide hospitalization for the most complex and severe cases. Long bed rest and disease-specific complications can greatly reduce patient autonomy and quality of life, even in the long term. An IRP based on early re-educational interventions appears to be effective in reducing perceived levels of dyspnea and in improving autonomy regarding ADL and muscle strength. Further research, with larger sample numbers, and including subjects admitted to other departments (e.g., intensive care unit), is needed to deepen and sharpen these findings.

FUNDING

This research did not receive any external funding.

CONFLICTS OF INTEREST

The authors declare no conflict of interest.

REFERENCES