Post–COVID-19 Ongoing Symptoms and Health-Related Quality of Life: Does Rehabilitation Matter?

Objective

Individuals with persisting symptoms after coronavirus disease 2019 have reported a decrease in health-related quality of life. This study explores the outcome of 50 subjects with post–coronavirus disease 2019 ongoing symptoms including “long COVID”(symptoms lasting over 3 mos), after a rehabilitation program focused on three symptoms: fatigue, breathlessness, and pain. The aims were as follows: to assess the feasibility of the program, to observe a possible change in symptoms and quality of life (null hypothesis: no differences in variables before and after treatment), and to investigate a possible relationship between symptoms and quality of life.

Design

This is a retrospective observational study.

Results

Symptoms intensity measured with numeric rating scale decreased. Mean differences are as follows: breathlessness, −2.91; fatigue, −2.05; and pain, −2.41 (P > 0.001). Quality of life measured with Euroqol-5D improved. Mean differences are as follows: Euroqol-5D index, 0.104, and Euroqol-5D visual analog scale, 19.21 (P < 0.001). Effect size of these changings is classifiable as large (Cohen d > 0.8 and <1.3) except for Euroqol-5D index (Cohen d = −0.575), resulting medium (>0.5 and <0.8). Correlation (Pearson r) between symptoms and Euroqol-5D resulted moderate for pain and fatigue (−0.609 and −0.531, P < 0.001) and low for breathlessness (−0.533, P < 0.001).

Conclusions

Rehabilitation can be feasible and may improve symptoms and quality of life. Further research is needed.

What Is Known

• Relevance of ongoing post–COVID symptoms has been increasing^1–7; the spectrum of symptoms is wide, but all of them play a role in the resulting negative impact on physical function and quality of life.^1,8 The need for a specific rehabilitation approach has emerged^9,10 to address the great variety of lasting sequelae.

What Is New

• This study reports the results of a rehabilitation program aimed to reduce three common symptoms requiring limited resources (simple evaluation methods, patient education, physiotherapy). While supporting the results of existing studies, it highlights the possible influence of the improvement of this limited group of symptoms on a global outcome, such as health-related quality of life.

The ongoing severe acute respiratory syndrome coronavirus 2 pandemic causing coronavirus disease 2019 (COVID-19) represents a heavy burden for healthcare systems globally. The need for resources to support COVID-19 survivors has become evident, and a rehabilitation approach early on during a hospital stay has been recommended.^11–13 This disease may bring chronic consequences: Carfì et al.¹ showed how 87% of a sample of 143 postdischarge patients reported one or more persisting symptoms 60 days after onset of first COVID-19 symptoms. Quality of life, measured with Euroqol 5 Dimension (EQ-5D) worsened for 44.1% of them. These first findings were confirmed by other studies; for example, Halpin et al.² identified persisting symptoms in a sample of 100 COVID-19 survivors at 4- and 8-wk postdischarge follow-up with 45.6% of them reporting negative impact on quality of life at EQ-5D score (the rate rose to 68.8% among those treated in an intensive care unit); Taboada et al.³ observed ongoing symptoms in a group of 91 COVID-19 survivors 6 mos after intensive care unit admittance with a decrease in EQ-5D score for 67% of them, whereas Sigfrid et al.⁴ reported persistence of symptoms and worsening of EQ-5D5L score (the five levels version of EQ-5D) in 93% of their 327 postdischarge patients at 3-mo follow-up. The COVID-19 long-term consequences are not limited to severe infection survivors: in their study on 1077 hospitalized COVID-19 patients, Evans et al.⁵ observed how the extent of the persistent health burden was only weakly linked to the severity of acute illness. These conclusions are confirmed by the study of Augustin et al.¹⁴ who detected a significant presence of ongoing symptoms (up to 34.8%) at 4- and 7-mo follow-up in a sample of 958 nonhospitalized COVID-19 patients who initially presented no or minor symptoms. With the increase in relevance of post–COVID-19 symptoms, the term “long COVID”(or post–COVID syndrome or long-haul COVID-19) has been used to define the persistence of symptoms for more than 3 mos after the first onset.⁹ The need for a specific rehabilitation approach has emerged^9,10 to address the great variety of physical, cognitive, and psychological symptoms, such as fatigue, breathlessness, “brain fog,” pain, anxiety, and depression.^1,6,7 The spectrum of symptoms is wide, but all of them play a role in the resulting negative impact on physical function and quality of life.^1,8 This study reports the early results of a rehabilitation program for patients with ongoing post–COVID-19 symptoms, conducted by the post–COVID-19 reactivation clinic of the rehabilitation medicine unit of Parma University Hospital. The program includes physiotherapy and educational aspects, and it is focused on the improvement of three of the most common symptoms, fatigue, pain, and breathlessness. The study hypothesis is that after the rehabilitation program, a significant reduction in symptoms perception and an improvement in health-related quality of life will be observed.

METHODS

This retrospective observational study reports the experience of the patients included in the outpatient rehabilitation program from October 2020 to April 2021. The study was approved by the North Emilia Area (AVEN) Committee on research ethics (reference 386/2021/OSS*/AOUPR); it is registered through the SIRER-Regional Platform on Health Research in Emilia Romagna¹⁵ Committee on research ethics, and the participants’ written informed consent has been obtained as required. This study conforms all STrengthening the Reporting of OBservational studies in Epidemiology guidelines and reports the required information accordingly (see Supplementary Checklist, Supplemental Digital Content 1, http://links.lww.com/PHM/B776).

Patients were referred to the rehabilitation clinic through the Parma Hospital Pneumological postacute COVID-19 follow-up clinic if the following criteria were met: age older than 18 yrs, previous severe acute respiratory syndrome coronavirus 2 infection demonstrated with molecular test, persistence of breathlessness, fatigue, and muscular/joint pain arisen after COVID-19 (key symptoms) in the absence of significant actual abnormalities emerging in vital parameters and instrumental examinations that could explain them (in particular: thoracic CT scan, spirometry, echocardiography). Exclusion criteria were as follows: demonstration of acute symptoms/medical instability, persistence of altered lung, and or/cardiac function at pneumological follow-up. The rehabilitation program was conceived on the basis of scientific literature review and the experience gained in the acute rehabilitation provided to hospitalized COVID-19 patients.¹⁶ The rehabilitation team included physiatrists, physiotherapists, occupational therapists, and psychologists. Patients referred by pneumologists received a physiatric evaluation including collection of information about comorbidities, acute COVID-19 phase, presence and intensity of breathlessness, fatigue, and pain. The subjective perception of symptoms was quantified by numeric rating scale (NRS),¹⁷ with 0 = no symptoms and 10 = worst possible intensity of symptoms. Patients were enrolled in the rehabilitation program if they scored 5 or more at NRS in at least one key symptom. Health-related quality of life was measured with EQ-5D¹⁸ with its two subscales: EQ index, more linked to individual functioning, and visual analog scale (VAS), a “thermometer” measuring the self-perceived health-related quality of life on a manually administered 100-point scale. The ends of the scale are going from “the best health you can imagine” to “the worst health you can imagine,” with a higher score indicating better health. Physical examination was based on a clinical neurological examination to rule out deficits in strength, reflexes, and sensitivity, including balance evaluation with Romberg test. It was completed by observation of spine, scapolothoracic and scapulohumeral joints mobility, observation of quiet, and maximal breathing pattern to detect signs of dysfunctionality,¹⁹ followed by Manual Assessment of Respiratory Motion²⁰ and manual examination of paravertebral and accessory respiratory muscles to identify pain/contractures using finger pressure. Tandem walking and a 5-meter barefoot walk for qualitative observation were added to further investigate the functionality and safety of gait. The rehabilitation program consisted of 10 × 45-min individual sessions with physiotherapists (2 times/week in general).

The physiotherapy approach was built on three main steps, beginning with breathing pattern awareness, guided by diaphragmatic breathing training and exercises for thoracoabdominal coordination. This baseline step was integrated in the individual exercises program chosen to reach the following goals:

• – release of muscular contractures (including manual physical therapy, soft tissue stretching, kinesiotaping),

• – increase of flexibility and mobility of the spine, shoulders, and hips: (active stretching exercises, static and dynamic exercises),

• – core stability (active exercises such as bridges, 4-point kneeling, exercises sitting on a bobath ball…), and

• – static and dynamic balance (proprioceptive exercises including two-leg stance and one-leg stance exercises).

All exercises were carried out in sets of 10 repetitions: observation of patient’s performance and tolerance during execution guided the progression in difficulty.

After improvements in respiratory pattern functionality, mobility, and balance, patients worked on endurance training (walking/treadmill based). The Borg-CR10 Scale²¹ was used to guide effort progression. Exercise choice, sequence, and dosage were matched to every subject’s main symptoms and objectives: the physiotherapists could adapt the intervention to meet patient individual needs.¹¹

Brief exercise sets for home practice were assigned, verifying patient adherence and possible difficulties at the start of the following session. An educational part including information accessible from the rehabilitation Website page²² completed the program to create a sort of “tool box” for self-treatment. Main points were as follows: pacing advice, suggestions about integrating correct breathing pattern and posture at work, including simple stretching exercises, and recommendations to maintain regular physical activity such as walking, in accordance with fatigue reduction. The rehabilitation outcome was registered at the end of the physiotherapy program measuring symptom intensity with NRS and quality of life with EQ-5D. Patients’ treatment was considered completed if they attended the scheduled program.

All analyses were performed excluding patients if missing data were observed. A paired t test was used to compare changes before and after the rehabilitation program verifying normality of EQ-5D index and VAS score distributions with the Shapiro-Wilk test before applying Student t test. Cohen d effect size was calculated to estimate the dimension of the changes, considering the following classification of the index: small = 0.2; medium = 0.5; large = 0.8; very large = 1.3.²³ Pearson correlation coefficient was calculated between EQ-5D and symptoms scores to verify possible correlations considering the following classification of the index: 0–0.3 = negligible correlation; 0.31–0.50 = low correlation; 0.51–0.70 = moderate correlation; 0.71–0.90 = high correlation; 0.91–1 = very high correlation.²⁴P values of 0.05 or less were considered statistically significant. All analyses were performed using Jamovi version 2.2.5.

RESULTS

Fifty individuals were included in the rehabilitation program from October 2020 to April 2021: 21 men (42%) and 29 women (58%), mean age was 53 yrs (±11.4). Nineteen individuals (38%) had infection managed at home, 31 (62%) needed hospitalization of which 10 cases (20%) included intensive care unit admission. Duration of post–COVID-19 symptoms since acute onset ranged from 2 to 9 mos (mean = 5.68 ± 3.22 mos, median = 6 mos). Thirty-one patients (62%) met long COVID definition criteria. The physiotherapy program was well tolerated by all the subjects; 5–10 supplementary sessions were added for the majority of patients to reach the intended improvement (mean number of sessions = 14.3 ± 4.16, median = 15). There were seven dropouts because of work circumstances.

Changes in symptom intensity and quality of life are shown in Table 1.

TABLE 1.

Paired samples t test

	Mean (SD)		P	Mean Difference	Effect Size		95% Confidence Interval
Variables	IN	OUT					Lower	Upper
BrNRS	5.614 (2.41)	2.705 (2.51)	<0.001	2.91	Cohen d	1.222	0.826	1.61
FatNRS	5.591 (2.47)	3.545 (2.77)	<0.001	2.05	Cohen d	0.892	0.538	1.239
PainNRS	4.568 (2.8)	2.159 (2.51)	<0.001	2.41	Cohen d	1.002	0.635	1.361
EQindex	0.652 (0.21)	0.756 (0.20)	<0.001	−0.10	Cohen d	−0.575	−0.895	−0.249
EQVAS	60.233 (17.42)	79.442 (16.48)	<0.001	−19.21	Cohen d	−1.094	−1.469	−0.71

BrNRS, breathlessness numeric rating score; EQindex, EQ-5D index; EQVAS, EQ-5D VAS; FatNRS, fatigue numeric rating score; IN, mean variables before rehabilitation treatment; OUT, mean variables after rehabilitation treatment; PainNRS, pain numeric rating score.

The NRS score mean difference was −2.91 (P < 0.001) for breathlessness, −2.05 (P < 0.001) for fatigue, and −2.41 (P < 0.001) for pain. The increase in EQ-5D was noted in both its dimensions: EQ-5D index mean difference was 0.104 (P < 0.001) and EQ-5D VAS mean difference was 19.21 (P < 0.001). The effect size of these changes is classifiable as large (Cohen d > 0.8 and <1.3²³) for every variable except for the EQ-5D index whose increase showed a Cohen d of −0.575 classifiable as medium effect size (>0.5 and <0.8²³). The correlation analysis between symptom and quality of life is shown as a heat plot in Table S1 (Supplemental Digital Content 2, http://links.lww.com/PHM/B777).

The strongest correlation emerges between the reduction of fatigue and the improvement in quality of life being Pearson r coefficient = −0.531 (P < 0.001) with EQ-5D index and r = −0.449 (P < 0.004) with EQ VAS, thus classifiable as “moderate correlation.”²⁴ The reduction of pain shows moderate correlation with EQ 5D index (Pearson r = −0.609, P < 0.001) and low correlation with EQ VAS (Pearson r = −0.449, P < 0.004). Breathlessness emerges as the symptom less strongly related to EQ-5D scores showing low negative correlation with EQ-5D index (Pearson r = −0.378, P < 0.016) and EQ VAS (Pearson r = −0.368, P < 0.02).

DISCUSSION

This experience represents a preliminary attempt to respond to the emerging need linked to the persistence of symptoms and decrease in health-related quality of life after COVID-19.²⁵ It is a retrospective observational study reporting the experience of the first 50 patients treated with a rehabilitation approach focused on relieving three of the most common symptoms: breathlessness, fatigue, and pain of COVID-19–related origin. The choice of these symptoms has been influenced also by practical aspects: the availability of resources in terms of spaces and staff was limited at the time because of the hospital reorganization after COVID-19 pandemic. These restrictions have been taken into account, and the program has been designed on a pragmatic basis, selecting recurring symptoms manageable with physiotherapy, simple and diffuse evaluation methods such as NRS and EQ-5D, and a global outcome measure such as health-related quality of life.

According to the literature recommendation, the program was adapted to individual needs¹¹ and it has been well tolerated by all the patients. The mean age and sex distribution in our sample are consistent with other studies.²⁵ Despite the differences in the acute phase severity, all the patients shared the symptoms of fatigue, breathlessness, and pain emerging with different intensity in each subject: these findings are consistent with other research^3–5,14 that shows how post–COVID-19 sequelae are not necessarily linked to clinical severity of initial infection.

The sample showed large variability in the duration of symptoms, more than 6 mos for half of them: nevertheless, an improvement after rehabilitation was generally observed indicating a possible role for rehabilitation for long COVID patients as well as subjects in the postacute phase.

The reduction of fatigue and pain after rehabilitation showed a moderate correlation with EQ-5D, while the correlation with breathlessness reduction resulted low, although this variable was comparable with the others in terms of perceived intensity before the treatment, mean difference after treatment, and effect size of the changing. Patients in the sample were mainly in working age: it could be hypothesized that fatigue and pain could cause a greater interference with the activities of daily life and with working perfomance, thus influencing quality of life. Still, these preliminary results suggest that perceived breathlessness could be furtherly explored to interpret its possible dual role: commonly reported symptom even in patients without significant respiratory function abnormalities and minor importance as a determinant of quality of life after rehabilitation. As a whole, the correlation of the three symptoms with quality of life suggests that the amelioration in EQ-5D could be influenced by symptom improvement. In this sense, the possible role of rehabilitation in improving health-related quality of life can be hypothesized.

The rehabilitation approach is based on evaluation methods and treatments that can be easily conducted by other rehabilitation facilities with limited resources.

The observed results of this study are in accordance with other experiences^26–28 showing initial but promising effects of rehabilitation programs sharing the same goals: relieving symptoms of dyspnea, improving physical function and quality of life, and including education and development of self-management skills.¹¹ This initial evidence supports the recommendations to implement rehabilitation provision by healthcare systems.^2,29,30

This study has several limitations, including the small sample size, the variability in the composition of the sample, the heterogeneity in duration of treatment and the lack of a control group. Rehabilitation program was individually tailored so that exercises sets and repetitions were not standardized, precluding recommendations about dosing at this time. The opportunity for an accurate consideration of the effectiveness of the intervention is therefore limited. The findings of our study are part of an observational framework that is not generalizable and where the causality can be hypothesized and not demonstrated.

However, they could help design future studies about need-focused rehabilitation strategies and education for post–COVID-19 sequelae.

CONCLUSIONS

This study provides evidence for the feasibility and safety of a rehabilitation program for patients with post–COVID-19 ongoing symptoms, including long COVID patients, with an approach tailored to patients’ needs and resources. After treatment a significant reduction in breathlessness, fatigue and pain has been noted as well as an improvement in health-related quality of life.

These observational findings support the importance of access to rehabilitation resources for post–COVID-19 patients who experience ongoing symptoms. At the same time, they could contribute to the design of future studies.