Effectiveness of adapted physical activity and therapeutic exercise programme in improving chronic fatigue syndrome in long COVID, delivered via hospital-based rehabilitation versus telerehabilitation

Abstract

Background

Long COVID is a prevalent condition characterised by pain, fatigue, disability, and a multitude of health issues. There are various treatment options for managing long COVID symptoms, including non-pharmacological interventions like physiotherapy and rehabilitation, which can be effectively delivered either in institutional care settings or via telerehabilitation.

Methods

This three-arm randomised controlled trial included 145 participants selected from a population-based cohort in eight administrative divisions in Bangladesh. Participants aged 18 and above diagnosed with chronic fatigue syndrome (CFS) secondary to long COVID were included and history of fatigue, cardiovascular, neuro-musculoskeletal, or respiratory diseases, or red flag signs were excluded. Participants were allocated to three groups: hospital-based rehabilitation (HBR), telerehabilitation (TR), or a home programme (HP). Interventions consisted of an individualised exercise programme. The HBR and TR groups received physiotherapist-supervised sessions with sessions lasting 45 min, twice weekly for 8 weeks. And the HP group performed exercises independently following structured instruction. Fatigue, the primary outcome, was measured using the Chalder fatigue scale, while secondary outcomes were quality of life measured using the 36-item Short Form Survey (SF-36), disability-adjusted life years (DALYs), and cardiorespiratory parameters (blood pressure, pulse rate, oxygen saturation, and lung capacity).

Finding

Between 1st July 2023 and 31st December 2023, 145 participants were enrolled, with a mean age of 46.1 ± 6.7 years. After 8 weeks of intervention, the among-group within-group comparison showed a significant difference in fatigue level (HBR: P < 0.001; TR: P < 0.001; HP: P < 0.321), physical functioning (HBR: P < 0.001; TR: P < 0.001; HP: P < 0.057), and episodic disability (HBR; TR; HP: P < 0.001) among the participants when comparing them between the groups. In multiple comparisons, results showed that differences were observed in the Chalder fatigue scale, physical functioning, and episodic disability between all groups. Hospital-based rehabilitation showed a lower mean score compared to telerehabilitation (p < 0.0001) and the home programme (p < 0.0001). Additionally, telerehabilitation was significantly better than the home programme (p < 0.0001), indicating hospital-based rehabilitation’s superior efficacy in reducing fatigue, improving physical function, and reducing disability.

Conclusion

Physiotherapy as hands-on implementation in a hospital setting was substantially more effective than telerehabilitation. Training healthcare professionals to improve accessibility to rehabilitation would help mitigate the consequences of long COVID-19.

Trail registration

The trial was registered with the clinical trial registry of India (CTRI/2023/03/050808. Registered on 17/03/2023).

Introduction

Long COVID encompasses a wide range of recurring or chronic health problems following infection with the SARS-CoV-2 virus [1]. The global prevalence of long COVID was recorded at 7% to 43%, with Asian countries reporting higher rates (51%) [2]. The Bangladeshi population has a documented prevalence of 16.1% [3], with a slightly higher prevalence of 24% in the delta variation [4]. Fatigue and pain are the most commonly reported symptoms of long COVID [3, 4].

Chronic fatigue syndrome (CFS), also known as myalgic encephalomyelitis (ME), is characterised by profound exhaustion and distressing symptoms that impact the brain, immunological system, and autonomic nervous system [5]. Moreover, CFS has a negative effect on an individual’s personal and social welfare, as well as their physical and psychological well-being [6]. Up to 45.2% of the global cases of COVID-19 reported experiencing CFS [6]. The rehabilitation being a non-pharmacological intervention includes exercise, cognitive behavioural therapy (CBT), pacing, aerobic activity, and symptomatic approaches [7, 8]. Adapted physical activity and therapeutic exercise (APTE) programmes have been shown to be more effective than passive control, CBT, or oral antidepressants in improving overall symptoms for non-COVID patients [9]. However, the National Institute for Health and Care Excellence (NICE) Guidelines do not promote graded exercise therapy for long COVID [7].

Various studies have investigated the impact of exercise therapy on individuals with CFS, primarily in non-LC patients [7]. Notably, exercise has been shown to significantly improve physical functioning in these individuals. Research also suggests that APTE is more effective than patient education in non-COVID individuals [10]. However, there is currently no research investigating the effects of APTE on CFS in individuals with long COVID symptoms. This study hypothesised that administering APTE through both institutional rehabilitation and telerehabilitation would be effective in reducing fatigue, enhancing physical performance, and minimising disability in long COVID patients with CFS. The study’s objectives were (1) to describe the baseline characteristics of the participants in each group, (2) to compare fatigue levels, cardio-respiratory function, physical functioning, and episodic disability between the groups before and after intervention, and (3) to determine fatigue level, cardio-respiratory function, physical functioning and episodic disability among three groups after intervention.

Methods

Study design and setting

This assessor-blinded, three-armed, randomised controlled trial (RCT) was conducted on patients with long COVID and CFS, as per the World Health Organization working group’s criteria [1]. The study was conducted between 1 July 2023 and 31 December 2023, involving 145 participants from a population cohort of 12,925 COVID-19 cases. The intervention was administered in five hospitals and their affiliated centres across eight administrative divisions in Bangladesh.

Participants

Participants were recruited by a multidisciplinary team comprising physicians and physiotherapists [4]. This study included individuals aged 18 and above, confirmed as COVID-19 positive via reverse transcription polymerase chain reaction (RT-PCR) at least 12 weeks prior, and diagnosed with CFS. Exclusion criteria included a history of fatigue due to cardiovascular, neuro-muscular, or respiratory diseases and red flag signs based on long COVID rehabilitation guidelines [11]. Participants who withdrew within the first week were also excluded from the analysis.

Sample and sampling technique

The sample size was calculated using G*power Software (Version 3.1.9.7), setting the alpha level for the one-way ANOVA at 5% and power at 80%. Using a medium effect size of 0.28 for the Chalder fatigue scale, we determined a sample size of 145 to accommodate an anticipated dropout rate of 15%. A criterion-based purposive sampling technique was employed to select participants.

Randomisation and masking

Participants were randomly assigned to one of three groups: hospital-based rehabilitation (HBR), telerehabilitation (TR), or a home programme (HP), using a computer-generated simple randomisation process conducted by an independent member from a different discipline who was not participating in this research. Sequentially numbered, opaque, sealed envelopes were used to conceal group allocations. The study adhered to blind design, with assessors blinded. Figure 1 the CONSORT flow diagram represents the blueprint of the study.

Fig. 1.

CONSORT flow diagram of the trial

Procedure

This study adhered to the ethical principles outlined in the Declaration of Helsinki (2013 revision) for the inclusion of human participants and followed the 2010 consolidated standard of reporting trials (CONSORT) guideline. Ethical approval was granted by the ethical review board of Universiti Malaysia Sabah (reference number: UMS/FPSK6.9/100-6/95), and the trial was registered in the clinical trial registry of India (reference number: CTRI/2020/09/028165). Each participant was provided with study information, and written informed consent was obtained. For participants who were illiterate, the ascent was sought from their legal guardian, and the participants provided a thumbprint to confirm their agreement. The intervention was conducted by graduate physiotherapists with 5 years of experience. Prior to the commencement of the study, the physiotherapists underwent 7 days of hands-on training provided by the members of the primary multidisciplinary team (MDT). The outcome measures were recorded at baseline and again at the end of the 8-week intervention period.

Interventions

Hospital-based rehabilitation (HBR)

Physiotherapists with 5 years of experience delivered the rehabilitation sessions under the supervision of a consultant physiotherapist. Each session lasted 45 min and was conducted twice weekly for 8 weeks within the hospital-based place, including both indoor facilities and outdoor areas. The programme began with a 2–3-min cardiorespiratory warm-up, followed by gentle stretching exercises of the upper limbs, lower limbs, and trunk (3 sets of 10-s holds). Mobility exercises, such as cat-cow stretches, wall angles, and gate openers, were performed with ten repetitions for each. The session then transitioned into a relaxation phase, where participants were positioned for 5 to 10 min and concluded with breathing exercises (5 sets, held for 10 to 30 s each). The stretching exercises were executed according to a personalised evaluation plan, each repeated ten times and held for 30 s. Participants were encouraged to incorporate walking or running into their routine for 10–20 min, depending on their tolerance level. Upon completing the exercises, participants were advised to rest adequately, follow a balanced diet, and maintain proper hydration. The intervention also included a home-based programme which involved various exercises and pacing strategies to complement the supervised sessions. The HBR group received hospital-based supervised sessions in addition to the home programme, which was identical to that provided to the HP group. Participants performed the home programme independently, following prior instruction provided at the start of the study.

Telerehabilitation (TR)

A similar programme was delivered via digital platforms (Zoom, Skype, WhatsApp, YouTube or Facebook) to participants in the telerehabilitation group. Physiotherapists provided real-time supervision and guidance, and participants were provided with a booklet containing detailed instructions. Each session lasted 45 min, twice weekly, for 8 weeks. Any adverse events were documented.

Home programme (HP)

During the initial days, the participants attended the medical centre, where they received comprehensive medical guidance tailored to their long COVID symptoms. In addition, they were provided with verbal instructions for the home exercise programme, which was identical to the home exercises prescribed to the HBR group. Participants could also receive support with pharmacological interventions or any other alternative treatments of their choice. Participants independently following structured instruction and documented all activities, regardless of whether an intervention was implemented.

Primary outcome measure

Chalder fatigue scale

This scale is a self-administered questionnaire consisting of 11 items on a 4-point Likert scale, with higher scores indicating greater fatigue. The scale has demonstrated an acceptable internal consistency, with reliability coefficients ranging from 0.78 to 0.96 [12].

Secondary outcome measure

Physical functioning sub-domain of SF-36

The physical functioning sub-domain of the SF-36 questionnaire assesses physical health. Scores range from 0 (poor health) to 100 (excellent health), with the sub-domain having a reliability coefficient of 0.90 [13].

Disability-adjusted life years (DALYs)

DALYs are calculated by summing the years lived with a disability (YLDs) and the years of life lost due to premature mortality (YLLs). The long COVIS’s disability weight (DW), the duration of ME/CFS in weeks (L1), and the incidence of ME/CFS from the previous study (I) were used to determine the YLDs. The YLLs were computed by subtracting the average life expectancy in Bangladesh from the mean age of death due to COVID-19, after determining the number of COVID-19-related fatalities in Bangladesh (N).

Systolic and diastolic blood pressure

Blood pressure (systolic and diastolic) was measured by a digital sphygmomanometer (OMRON R6, OMRON Corp). Measurements were taken at baseline (pre-intervention) and after 8 weeks of intervention, prior to any exercise session, with participants seated, arms resting on a table at chest level, and both feet firmly on the ground. Before collecting data, the participants were instructed to relax for 30 min and were prohibited from eating any meals. The measurement was recorded thrice, and the mean value was calculated. According to the ESH-IP2 guidelines, the Omron RS6 is a reliable and appropriate device for measuring blood pressure.

Pulse oximeter

A pulse oximeter (Mode: YK-88 LED, Yonker, Xuzhou Yongkang Electronic Science Technology Co., Ltd.) was used to measure the SaO2 and pulse rate (PR) of the participants. Measurements were performed at baseline and after 8 weeks of intervention, with participants at rest prior to any exercise. The pulse oximeter is valid and reliable to measure oxygen saturation and PR.

Digital spirometer

A digital spirometer Smart Phone Based Handheld Wireless Spirometer (Model Number: SP80B Origin, Contec brand) was used to measure functional residual capacity and total lung capacity of the participants. Measurements were taken at baseline and after 8 weeks of intervention, prior to any exercise session. A digital spirometer has good agreement and reproducibility. And the portable spirometer is a reliable tool to screen lung volume and capacity in the primary care settings.

Statistical analysis

The data were analysed using the Statistical Package of Social Science (IBM SPSS) version 23.0. The data underwent analysis to determine its normal distribution using the Bell curve, Kolmogorov–Smirnov, and Shapiro–Wilk tests. A descriptive analysis was performed on continuous data, expressed in mean and standard deviation for parametric data and median and interquartile range for non-parametric data. Categorical data were expressed in frequency and percentage. Then, the chi-square test, Fisher’s exact test, and one-way ANOVA were employed to evaluate baseline data compatibility. Subsequently, the paired t-test was applied for within-group comparisons, and the independent t-test for between-group comparisons. Lastly, the comparison between groups was evaluated using one-way ANOVA. This study did not employ an intention-to-treat analysis. The significance level was set at p < 0.05.

Result

Baseline socio-demographics

A total of 145 participants were initially enrolled, with 137 completing the study after accounting for dropouts. The mean age of the participants was 46.1 ± 7.6 years. Males were more predominant (69.7%) than females (30.3%), and 89.7% of participants lived with their spouses. Most participants were employed in the service sector (32.5%) and resided in urban areas (36.6%). The majority (58.6%) had a bachelor’s degree or higher, and 57.2% earned more than 30,000 BDT (~300 USD) monthly. Table 1 represents the demographic data.

Table 1.

Baseline characteristics of sociodemographic variables among participants

Variables	Overall		Hospital-based rehabilitation (N = 48)		Telerehabilitation (N = 48)		Home programmes (N = 49)		p-value
	Number	Percent	Number	Percent	Number	Percent	Number	Percent
Age (years) (mean ± SD)	46.1	± 7.6	45.2	± 9.2	46.1	± 6.2	46.9	± 7.1	0.566a
Age
21–30 years	04	2.8%	03	6.2%	00	0.0%	01	2.0%	0.259a
31–40 years	31	21.4%	14	29.2%	09	18.8%	08	16.3%
41–50 years	76	52.4%	19	39.6%	29	60.4%	28	57.1%
51–60 years	28	19.3%	08	16.7%	10	20.8%	10	20.4%
Above 61 years	06	4.1%	04	8.3%	00	0.0%	02	4.1%
Gender
Male	101	69.7%	32	66.7%	38	79.2%	31	63.3%	0.202b
Female	44	30.5%	16	33.3%	10	20.8%	18	36.7%
Marital status
Single	15	10.3%	08	16.7%	04	8.3%	03	6.1%	0.250c
Married	130	89.7%	40	83.3%	44	91.7%	46	93.9%
Occupation
Service holder	47	32.4%	13	27.1%	16	33.3%	18	36.7%	0.820c
Health Professional	09	6.2%	01	2.1%	04	8.3%	04	8.2%
Law Enforcement	07	4.8%	02	4.2%	02	4.2%	03	6.1%
House wife	27	18.6%	10	20.8%	08	16.7%	09	18.4%
Day labour	17	11.7%	09	18.8%	04	8.3%	04	8.2%
Student	11	7.6%	05	10.4%	04	8.3%	02	4.1%
Business	27	18.6%	08	16.7%	10	20.8%	09	18.4%
Living area
Rural	46	31.7%	16	33.3%	18	37.5%	12	24.5%	0.472b
Semi-urban	46	31.7%	16	33.3%	11	22.9%	19	38.8%
Urban	53	36.6%	16	33.3%	19	39.6%	18	36.7%
Educational status
No formal education	04	2.8%	04	8.3%	00	0%	00	0%	0.700a
Primary Education	12	8.3%	05	10.4%	03	6.3%	04	8.2%
Secondary education	17	11.7%	06	12.5%	03	6.3%	08	16.3%
Higher Secondary	27	18.6%	07	14.6%	09	18.8%	11	22.4%
Bachelor or above	85	58.6%	26	54.2%	33	68.8%	26	53.1%
Monthly family income
Below 30,000 BDT (~300 USD)	62	42.8%	26	54.2%	16	33.3%	20	40.8%	0.191a
30,001–100,000 BDT (~301–1000 USD)	83	57.2%	22	45.8%	32	66.7%	29	59.2%

^a Spearman rank test

^b Chi-square test

^c Fisher’s exact test

Baseline clinical characteristics

After fatigue, memory and concentration issues (99.3%) and musculoskeletal discomfort (98.6%) were the most prevalent symptoms. The study participants experienced high rates of insomnia (96.6%), difficulties with activities of daily living (90.3%), and respiratory issues (84.1%). The study participants had also a high prevalence of diabetes (57.0%) and hypertension (57.0%), followed by chronic obstructive pulmonary disease (COPD) (21.5%) and ischaemic heart disease (IHD) (19.8%). In this study, 39.3% of the individuals needed to be admitted to the hospital for treatment of COVID-19, while 22.8% required intensive care unit (ICU), high dependency unit (HDU), or critical care support. Following the COVID-19 pandemic, a significant majority of participants, specifically 82.8%, got medical intervention (Table 2).

Table 2.

Baseline compatibility of clinical presentation among participants

Variables	Overall		Hospital-based rehabilitation (n = 48)		Telerehabilitation		Home programmes		P
	Number/mean/median	Percent/IQR/SD	Number/mean/median	Percent/IQR/SD	Number/mean/median	Percent/IQR/SD	Number/mean/median	Percent/IQR/SD
LC symptoms
Fatigue	145	100.0%	48	100.0%	48	100.0%	49	100.0%	-
Musculoskeletal pain	143	98.6%	48	100.0%	48	100.0%	47	95.9%	0.137a
Memory and Concentration problem	144	99.3%	48	100.0%	48	100.0%	48	98.0%	0.373a
Headache	75	51.7%	25	52.1%	25	52.1%	25	51.0%	0.993a
Respiratory Problem	122	84.1%	40	83.3%	41	85.4%	41	83.7%	0.956a
Anxiety	72	49.7%	26	54.2%	26	54.2%	20	40.8%	0.315a
Depression	93	64.1%	29	60.4%	28	58.3%	36	73.5%	0.241a
Insomnia	140	96.6%	47	97.9%	46	95.8%	47	95.9%	0.818a
Problem in ADL	131	90.3%	43	89.6%	43	89.6%	45	91.8%	0.910a
Problem with work	73	50.3%	23	47.9%	27	56.2%	23	46.9%	0.603a
Vertigo	46	31.7%	16	33.3%	12	25.0%	18	36.7%	0.443a
Hot flash/burning sensation	61	42.1%	25	52.1%	18	37.5%	18	36.7%	0.228a
Mood fluctuation	38	26.2%	13	27.1%	12	25.0%	13	26.5%	0.971a
Suicidal tendency	14	9.7%	06	12.5%	06	12.5%	02	4.1%	0.267b
Stammering	31	21.4%	12	25.0%	09	18.8%	10	20.4%	0.741a
LC symptoms duration (weeks)
Fatigue duration	52.00	48.00–55.00	52.00	50.00–56.00	52.00	50.00–53.00	52.00	49.00–56.00	0.623c
Musculoskeletal pain duration	45.00	40.00–55.00	47.50	44.00–50.00	42.00	40.00–47.75	45.00	40.00–52.00	0.069c
Memory and concentration problem duration	49.00	44.00–53.00	50.00	44.00–52.00	49.00	47.25–53.00	48.00	40.00–53.00	0.278c
Headache duration	20.00	0.00–31.50	21.50	0.00–30.00	22.50	0.00–45.75	20.00	0.00–30.00	0.344c
Respiratory problem duration	51.00	48.00–54.00	52.00	48.00–56.00	49.00	48.00–52.00	52.00	48.00–55.00	0.052c
Anxiety duration	0.00	0.00–48.00	29.00	0.00–51.75	40.00	0.00–47.75	0.00	0.00–40.00	0.272c
Depression duration	30.00	0.00–40.00	25.50	0.00–40.00	30.00	0.00–38.00	30.00	0.00–40.00	0.280c
Insomnia duration	53.00	48.00–57.00	51.50	48.00–56.00	54.50	50.00–66.75	53.00	48.00–56.50	0.052c
Problem in ADL duration	50.00	44.00–53.00	51.00	45.25–54.75	48.00	40.00–51.75	50.00	47.50–52.00	0.053c
Problem with work duration	28.00	0.00–52.00	0.00	0.00–51.75	33.50	0.00–49.00	0.00	0.00–53.00	0.945c
Vertigo duration	0.00	0.00–30.00	0.00	0.00–26.00	0.00	0.00–22.50	0.00	0.00–36.00	0.214c
Hot flash/burning sensation duration	0.00	0.00–40.00	22.00	0.00–30.00	0.00	0.00–48.75	0.00	0.00–38.00	0.654c
Mood fluctuation duration	0.00	0.00–25.00	0.00	0.00–23.00	0.00	0.00–22.50	0.00	0.00–30.00	0.834c
Suicidal Tendency duration	0.00	0.00–0.00	0.00	0.00–0.00	0.00	0.00–0.00	0.00	0.00–0.00	0.264c
Stammering duration	0.00	0.00–0.00	0.00	0.00–18.00	0.00	0.00–0.00	0.00	0.00–0.00	0.605c
Comorbidity
	Frequency	Percent	Frequency	Percent	Frequency	Percent	Frequency	Percent
DM	69	57.0%	22	53.7%	24	60.0%	23	57.5%	0.935a
HTN	69	57.0%	21	51.2%	23	57.5%	25	62.5%	0.798a
IHD	24	19.8%	10	24.4%	06	15.0%	08	20.0%	0.547a
COPD	26	21.5%	11	26.8%	07	17.5%	08	20.0%	0.587a
CKD	19	15.7%	08	19.5%	08	20.0%	03	7.5%	0.218b
Hypothyroidism	13	10.7%	06	14.6%	02	5.0%	05	12.5%	0.389b
UTI	01	0.8%	01	2.4%	00	0%	00	0%	0.662b
Osteoporosis	28	23.1%	08	19.5%	09	22.5%	11	27.5%	0.807b
High cholesterol	02	1.7%	02	4.9%	00	0%	00	0%	0.216b
Number of COVID-19 infection
Single times	136	93.8%	43	89.6%	46	95.8%	47	95.9%	0.469b
Couple times	09	6.2%	05	10.4%	02	4.2%	02	4.1%
Treatment received for COVID-19
Medication	30	20.7%	11	22.9%	08	16.7%	11	22.4%	0.967a
Hospital admission	57	39.3%	19	39.6%	19	39.6%	19	38.8%
ICU/HDU/critical care	33	22.8%	09	18.8%	13	27.1%	11	22.4%
None	25	17.2%	09	18.8%	08	16.7	08	16.3
Treatment received after COVID-19
Yes	120	82.8%	37	77.1%	41	85.4%	42	85.7%	0.477a
No	25	17.2%	11	22.9%	07	14.6%	07	14.3%
CFS score	26.1	± 2.2	26.2	± 2.6	26.0	± 2.2	26.1	± 1.8	0.881a
Systolic BP (mmHg)	124.4	± 11.7	125.5	± 18.5	125.0	± 5.8	122.9	± 13.5	0.620a
Diastolic BP(mmHg)	84.7	± 11.6	86.0	± 14.2	83.9	± 9.4	84.2	± 10.9	0.623a
SaO2(%)	95.4	± 1.4	95.4	± 1.5	95.5	± 1.2	95.4	± 1.2	0.765a
Pulse rate (bpm)	78.8	± 10.5	82.7	± 13.3	81.0	± 5.9	82.3	± 7.2	0.666a
Force residual capacity (ml)	2193.0	± 61.1	2201.8	± 63.5	2208.1	± 46.3	2201.0	± 57.3	0.773a
Total lung capacity (ml)	5612.4	± 66.1	5606.7	± 62.5	5600.2	± 58.2	5623.1	± 57.3	0.206a
Physical functioning score	27.6	± 6.4	24.3	± 8.1	24.5	± 1.7	25.4	± 2.0	0.460a

^aChi-square test;^bFisher’s exact test; ^cKruskal-Wallis test; ^aOne-way ANOVA; DM Diabetes mellitus, HTN Hypertension, IHD Ischaemic heart disease, COPD Chronic obstructive pulmonary disease, CKD Chronic kidney disease, UTI Urinary tract infection, CFS Chalder fatigue scale, BP Blood pressure

Change of fatigue level

Following an 8-week intervention, there was no significant difference in the home exercise group from baseline (26.11; 95% CI 25.52, 26.71) to post-intervention (25.35; 95% CI 24.92, 25.84) (p = 0.057). A statistically significant difference was observed in hospital-based rehabilitation, where the mean CFS score decreased from 26.23 (95% CI 25.44, 27.31) at baseline to 13.61 (95% CI 12.75, 14.52) post-intervention (p < 0.001), and in telerehabilitation, where the mean score reduced from 26.11 (95% CI 25.44, 26.80) to 16.64 (95% CI 15.80, 17.54) (p < 0.001) after the intervention. In between-group comparisons, the study found no statistically significant difference between hospital-based rehabilitation and telerehabilitation (mean difference (MD) = − 3.04; 95% CI − 4.25, − 1.82; t = − 4.97; p = 0.246). The study identified significant differences in the effectiveness of hospital-based rehabilitation compared to home programmes (MD = − 8.70; 95% CI − 9.63, − 7.77; t = − 18.61; p = 0.023) and telerehabilitation compared to home programmes (MD = − 11.74; 95% CI − 12.76, − 10.72; t = − 22.87; p < 0.001). Also, among-group comparison showed a significant difference in fatigue level between hospital-based rehabilitation (13.60 ± 3.08), telerehabilitation (16.64 ± 2.72), and home programmes (25.35 ± 1.61), with F = 262.27 (p < 0.001). These findings are displayed in Table 3 and Fig. 2.

Table 3.

Change of fatigue level

Within group comparison (paired T-test)
Variables	Hospital-based rehabilitation (n = 46)			Telerehabilitation (n = 45)				Home programmes (n = 46)
	Mean	95% CI	p value	Mean	95% CI	p value		Mean	95% CI		p value
CFS
Baseline	26.23	25.44, 27.31	< 0.001***	26.11	25.44, 26.80	< 0.001***		26.11	25.52, 26.71		0.057
Post-intervention	13.61	12.75, 14.52		16.64	15.80, 17.54			25.35	24.92, 25.84
Between group comparison (independent T-test)
Variable	MD	Std. error diff	95% CI	t				F		p value
Hospital-based rehabilitation VS telerehabilitation
CFS score	− 3.04		0.61	− 4.25, − 1.82	− 4.97			1.37		0.246
Telerehabilitation VS home programmes
CFS Score	− 8.70		0.47	− 9.63, − 7.77	− 18.61			5.33		0.023*
Hospital-based rehabilitation VS home programmes
CFS Score	− 11.74		0.51	− 12.76, − 10.72	− 22.87			14.26		< 0.001***
Among group comparison (one-way ANOVA)
Variable	Hospital-based rehabilitation (n = 46)		Telerehabilitation (n = 45)		Home programmes (n = 46)			F		p value
	Mean	SD	Mean	SD	Mean		SD
CFS score	13.60	± 3.08	16.64	± 2.72	25.35		± 1.61	262.27		< 0.001***

Fig. 2.

Fatigue, physical functioning and episodic disability variation long COVID. After 8 weeks of intervention, hospital-based rehabilitation and telerehabilitation resulted in significant improvements in CFS scores, physical function, and DALYs compared to home programmes

Change of cardio-respiratory function Level

Systolic blood pressure decreased significantly in the hospital-based (from 125.76 mmHg (95% CI 120.18, 131.33) to 121.74 mmHg (95% CI 118.93, 124.54), p = 0.015) and telerehabilitation groups (from 125.00 mmHg (95% CI 123.22, 126.78) to 122.00 mmHg (95% CI 119.81, 124.18), p = 0.022), but not in the home programmes group (from 123.80 mmHg (95% CI 119.85, 127.75) to 123.37 mmHg (95% CI 120.82, 125.91), p = 0.759). Similarly, diastolic blood pressure improved in the hospital-based (from 86.19 mmHg (95% CI 81.92, 90.46) to 80.76 mmHg (95% CI 78.48, 83.04), p < 0.001) and telerehabilitation groups (from 84.18 mmHg (95% CI 81.32, 87.03) to 79.56 mmHg (95% CI 77.61, 81.49), p < 0.001), but the change in the home programmes group (from 84.78 mmHg (95% CI 81.58, 87.99) to 82.61 mmHg (95% CI 80.69, 84.53), p = 0.053) was not significant. Significant improvements in pulse rate were observed in the hospital-based group (from 82.78 (95% CI 78.77, 86.79) to 76.65 (95% CI 74.56, 78.74), p < 0.001) and telerehabilitation group (from 75.53 (95% CI 73.43, 77.64) to 73.36 (95% CI 72.35, 74.35), p = 0.004), with no significant change in the home programmes group (from 78.57 (95% CI 75.74, 81.39) to 78.91 (95% CI 76.29, 81.53), p = 0.176). Significant improvements in oxygen saturation (SaO₂), functional residual capacity, and total lung capacity were also observed only in the two active intervention groups (p < 0.01), with no significant changes in the home programmes group (p > 0.05). However, no significant improvement was observed in the home programme (p > 0.05) (Table 4).

Table 4.

Change of cardio-respiratory function level

Baseline to post-test (within group comparison)
Variables		Hospital-based rehabilitation (n = 46)			Telerehabilitation (n = 45)			Home programmes (n = 46)
		Mean	95% CI	p value	Mean	95% CI	p value	Mean	95% CI	p value
Systolic BP
Baseline		125.76	120.18, 131.33	0.015*	125.00	123.22, 126.78	0.022*	123.80	119.85, 127.75	0.759
Post-intervention		121.74	118.93, 124.54		122.00	119.81, 124.18		123.37	120.82, 125.91
Diastolic BP
Baseline		86.19	81.92, 90.46	< 0.001***	84.18	81.32, 87.03	< 0.001***	84.78	81.58, 87.99	0.053
Post-intervention		80.76	78.48, 83.04		79.56	77.61, 81.49		82.61	80.69, 84.53
Pulse rate
Baseline		82.78	78.77, 86.79	< 0.001***	75.53	73.43, 77.64	0.004**	78.57	75.74, 81.39	0.176
Post-intervention		76.65	74.56, 78.74		73.36	72.35, 74.35		78.91	76.29, 81.53
SaO2
Baseline		95.30	94.89, 95.72	0.001***	95.62	95.26, 95.98	0.006**	95.43	95.08, 95.79	0.255
Post-intervention		96.48	96.26, 96.69		96.27	95.93, 96.59		94.83	93.76, 95.88
Functional residual capacity
Baseline		2168.48	2146.93, 2190.02	0.001***	2210.89	2197.39, 2224.38	0.001***	2198.91	2181.91, 2215.91	0.030*
Post-intervention		2242.39	2224.41, 2260.36		2246.67	2233.69, 2259.64		2213.69	2200.03, 2227.36
Total lung capacity
Baseline		5595.86	5574.80, 5616.92	0.001***	5596.89	5579.28, 5614.49	0.001***	5639.78	5620.98, 658.57	0.790
Post-intervention		5733.69	5716.25, 5751.13		5688.89	5665.59, 5712.18		5636.96	5622.13, 5651.78
Between group comparison (independent T-test)
Variable		MD		Std. error diff		95% CI		t	F	p value
Hospital-based rehabilitation VS telerehabilitation
Systolic BP		− 0.26		1.77		− 3.77, 3.25		− 0.15	2.92	0.091
Diastolic BP		1.21		1.49		− 1.75, 4.16		0.81	3.62	0.060
Pulse rate		3.29		1.16		0.99, 5.59		2.85	24.54	< 0.001***
SaO2		0.21		0.19		− 0.17, 0.59		1.09	3.56	0.063
Functional residual capacity		− 4.28		11.04		− 26.22, 17.66		− 0.39	8.57	0.004**
Total lung capacity		44.81		14.39		16.19, 73.42		3.11	2.08	0.153
Telerehabilitation VS home programmes
Systolic BP		− 1.37		1.67		− 4.68, 1.94		− 0.82	0.99	0.321
Diastolic BP		− 3.05		1.36		− 5.75, − 0.36		− 2.25	0.95	0.333
Pulse rate		− 5.56		1.40		− 8.34, − 2.77		− 3.96	30.18	< 0.001***
SaO2		1.44		0.56		0.33, 2.55		2.584	5.436	0.022*
Functional residual capacity		32.97		9.36		14.38, 51.56		3.52	0.02	0.880
Total lung capacity		51.93		13.64		24.83, 79.03		3.81	6.91	0.010**
Hospital-based rehabilitation VS home programmes
Systolic BP		− 1.63		1.88		− 5.36, 2.10		− 0.87	0.50	0.481
Diastolic BP		− 1.85		1.48		− 4.79, 1.09		− 1.25	1.20	0.276
Pulse rate		− 2.26		1.66		− 5.56, 1.04		− 1.36	1.64	0.203
SaO2		1.65		0.54		0.58, 2.72		3.07	8.29	0.005**
Functional residual capacity		28.69		11.21		6.43, 50.97		2.56	6.93	0.010**
Total lung capacity		96.73		11.37		74.16, 119.32		8.51	2.31	0.132
Among group comparison (one-way ANOVA)
Variable	Hospital-based rehabilitation (n = 46)			Telerehabilitation (n = 45)		Home programmes (n = 46)			F	p value
	Mean		SD	Mean	SD	Mean		SD
Systolic BP	121.74		± 9.44	122.00	± 7.26	123.37		± 8.57	0.49	0.614
Diastolic BP	80.76		± 7.67	79.56	± 6.47	82.61		± 6.48	2.27	0.108
Pulse rate	76.65		± 7.04	73.36	± 3.32	78.91		± 8.80	7.67	< 0.001***
SaO2	96.48		± 0.72	96.27	± 1.09	94.83		± 3.58	7.61	< 0.001***
Functional residual capacity	2242.39		± 60.52	2246.67	± 43.17	2213.69		± 46.01	5.76	0.004**
Total lung capacity	5733.69		± 58.74	5688.89	± 77.52	5637.96		± 49.92	27.16	< 0.001***

In between group comparison, a statistically significant difference in pulse rate was observed between the hospital-based rehabilitation and telerehabilitation groups (MD = 3.29; 95% CI 0.99, 5.59; t = 2.85; F = 24.54; p < 0.001), and in functional residual capacity (MD = − 4.28; 95% CI − 26.22, 17.66; t = − 0.39; F = 8.57; p = 0.004). There were no statistically significant differences observed in systolic blood pressure (MD = − 0.26; 95% CI − 3.77, 3.25; p = 0.091), diastolic blood pressure (MD = 1.21; 95% CI − 1.75, 4.16; p = 0.060), SaO₂ (MD = 0.21; 95% CI − 0.17, 0.59; p = 0.063), and total lung capacity (MD = 44.81; 95% CI 16.19, 73.42; p = 0.153). This study identified significant differences between telerehabilitation and home programmes in terms of pulse rate (MD = − 5.56; 95% CI − 8.34, − 2.77; t = − 3.96; F = 30.18; p < 0.001), SaO₂ (MD = 1.44; 95% CI 0.33, 2.55; t = 2.584; F = 5.436; p = 0.022), and total lung capacity (MD = 51.93; 95% CI 24.83, 79.03; t = 3.81; F = 6.91; p = 0.010). The study found no statistically significant difference in systolic (p = 0.321) and diastolic (p = 0.333) blood pressure, as well as functional residual capacity (MD = 32.97; 95% CI 14.38, 51.56; p = 0.880). A statistically significant difference was found in the levels of SaO₂ (MD = 1.65; 95% CI 0.58, 2.72; t = 3.07; F = 8.29; p = 0.005) and functional residual capacity (MD = 28.69; 95% CI 6.43, 50.97; t = 2.56; F = 6.93; p = 0.010) when comparing the hospital-based rehabilitation and home programmes. The study found no statistically significant differences in systolic (p = 0.481) and diastolic (p = 0.276) blood pressure, pulse rate (p = 0.203), and total lung capacity (p = 0.132) (Table 4).

After that, in a group comparison following 8 weeks of intervention, there was no significant difference in systolic (F = 0.49; p = 0.614) and diastolic (F = 2.27; p = 0.108) blood pressure among the groups. The study found significant differences in pulse rate (F = 7.67; p < 0.001), SaO₂ (F = 7.61; p < 0.001), functional residual capacity (F = 5.76; p = 0.004), and total lung capacity (F = 27.16; p < 0.001) (Table 4).

Change of physical functioning level

There was no significant difference in the home exercise at baseline (24.80; 95% CI 24.34, 25.35) or after the intervention (25.17; 95% CI 24.59, 25.92; p = 0.321). A statistically significant difference was observed in hospital-based rehabilitation (from 24.19; 95% CI 21.82, 26.63 to 67.39; 95% CI 64.34, 70.55; p < 0.001) and telerehabilitation (from 24.89; 95% CI 24.30, 25.51 to 30.79; 95% CI 28.76, 32.32; p < 0.001) both before and after the intervention. The study identified significant differences in the effectiveness of hospital-based rehabilitation compared to home exercise programmes (MD = 42.22; 95% CI 39.11, 45.33; t = 26.95; p < 0.001) and telerehabilitation compared to home exercise programmes (MD = 5.31; 95% CI 3.46, 7.17; t = 5.69; p < 0.001). The study found no statistically significant difference between hospital-based rehabilitation and telerehabilitation (MD = 36.90; 95% CI 33.38, 40.42; t = 20.83; p = 0.178). There was a significant difference in physical functioning among the participants when comparing them between the groups (hospital-based: 67.39 ± 10.37; telerehabilitation: 30.49 ± 5.88; home programmes: 25.17 ± 2.32; F = 492.752; p < 0.001). These findings are displayed in Table 5 and Fig. 2.

Table 5.

Change of physical functioning level

Within group comparison (paired T-test)
Variables	Hospital-based rehabilitation (n = 46)						Telerehabilitation (n = 45)					Home programmes (n = 46)
	Mean	95% CI			p value		Mean		95% CI	p value		Mean			95% CI		p value
SF-36 (physical functioning)
Baseline	24.19	21.82, 26.63			< 0.001***		24.89		24.30, 25.51	< 0.001***		24.80			24.34, 25.35		0.321
Post-intervention	67.39	64.34, 70.55					30.79		28.76, 32.32			25.17			24.59, 25.92
Between group comparison (independent T-test)
Variable			MD			Std. Error Diff		95% CI			t			F			p value
Hospital-based rehabilitation VS telerehabilitation
SF-36 score (physical functioning)			36.90			1.77		33.38, 40.42			20.83			1.85			0.178
Telerehabilitation VS home programmes
SF-36 Score (Physical Functioning)			5.31			0.93		3.46, 7.17			5.69			37.09			< 0.001***
Hospital-based rehabilitation VS home programmes
SF-36 score (physical functioning)			42.22			1.57		39.11, 45.33			26.95			19.19			< 0.001***
Among group comparison (one-way ANOVA)
Variable	Hospital-based rehabilitation (n = 46)					Telerehabilitation (n = 45)					Home programmes (n = 46)					F	p value
	Mean			SD		Mean	SD				Mean		SD
SF-36 score (physical functioning)	67.39			± 10.37		30.49	± 5.88				25.17		± 2.32			492.752	< 0.001***

Significant level: *p < 0.05; **p < 0.01; ***p < 0.001

Change of episodic disability level

A statistically significant difference was observed in hospital-based rehabilitation (from 0.48 (95% CI 0.39, 0.57) to 0.23 (95% CI 0.18, 0.23), p < 0.001) and telerehabilitation (from 0.47 (95% CI 0.40, 0.54) to 0.39 (95% CI 0.34, 0.44), p < 0.001). Interestingly, the home programmes group showed a significant increase in DALYs (from 0.44 (95% CI 0.37, 0.52) to 0.59 (95% CI 0.53, 0.66), p < 0.001), indicating a worsening of disability over time. The study identified significant differences in the effectiveness of telerehabilitation compared to home programmes (MD = − 0.20; 95% CI − 0.28, − 0.13; t = − 5.192; p = 0.041). There was no statistically significant difference between hospital-based rehabilitation and telerehabilitation (MD = − 0.15; 95% CI − 0.22, − 0.09; t = − 4.508; p = 0.793) and hospital-based rehabilitation compared to home programmes (MD = − 0.38; 95% CI − 0.44, − 0.28; t = − 8.828; p = 0.052). After 8 weeks of intervention, there was a significant difference in episodic disability among the participants when comparing them between the groups (hospital-based: 0.23 ± 0.18; telerehabilitation: 0.39 ± 0.15; home programmes: 0.59 ± 0.21; F = 44.680; p < 0.001). These findings are displayed in Table 6 and Fig. 2.

Table 6.

Change of episodic disability level

Within group comparison (paired T-test)
Variables		Hospital-based rehabilitation (n = 46)						Telerehabilitation (n = 45)					Home programmes (n = 46)
		Mean			95% CI	p value		Mean		95% CI		p value	Mean			95% CI		p value
DALYs
Baseline		0.48			0.39, 0.57	< 0.001***		0.47		0.40, 0.54		< 0.001***	0.44			0.37, 0.52		< 0.001***
Post-intervention		0.23			0.18, 0.23			0.39		0.34, 0.44			0.59			0.53, 0.66
Between group comparison (independent T-test)
Variable			MD				Std. error diff		95% CI		t				F			p value
Hospital-based rehabilitation VS telerehabilitation
DALYs (post-test)			− 0.15				0.03		− 0.22, − 0.09		− 4.508				0.069			0.793
Telerehabilitation VS home programmes
DALYs (post-test)			− 0.20				0.04		− 0.28, − 0.13		− 5.192				4.294			0.041*
Hospital-based rehabilitation VS home programmes
DALYs (post-test)			− 0.38				0.04		− 0.44, − 0.28		− 8.828				3.870			0.052
Among group comparison (one-way ANOVA)
Variable	Hospital-based rehabilitation (n = 46)					Telerehabilitation (n = 45)					Home programmes (n = 46)						F	p value
	Mean			SD		Mean		SD			Mean			SD
DALYs (post-test)	0.23			± 0.18		0.39		± 0.15			0.59			± 0.21			44.680	< 0.001***

Significance level: *P < 0.05; **P < 0.01; ***P < 0.001

Discussion

This RCT demonstrated the efficacy of adapted physical activity and therapeutic exercise programmes, delivered via HBR and TR, in reducing fatigue, improving physical functioning, improving cardiorespiratory function and reducing episodic disability in patients with long COVID and CFS. While HBR and TR were more effective than HP, no significant differences were observed between HBR and TR for most outcome measures, except for PR and FRC. These findings indicate that both structured rehabilitation approaches are superior to HP. Therefore, although the null hypothesis suggested no difference between approaches, both rehabilitation models were clearly more effective than HP across most outcome domains. Importantly, previous studies have typically evaluated single rehabilitation strategies without directly comparing hospital-based and tele-rehabilitation models within the same randomised framework [14]. Moreover, limited evidence has simultaneously examined fatigue, cardiorespiratory function, physical functioning, and episodic disability in one comprehensive trial. Therefore, our study helps bridge this gap by providing comparative evidence across multiple clinically relevant domains.

Fatigue is one of the most prevalent disabling symptoms of long COVID. While several effective treatments are recommended [15], there is still no clear consensus on evidence-based interventions for its management [16]. A prospective cohort study reported that a multidisciplinary rehabilitation programme, including aquatic, respiratory and motor exercises, social integration training, neuropsychological sessions, and light amplification by stimulated emission of radiation therapy, significantly alleviated fatigue in long COVID patients treated in a hospital setting [17]. A prospective longitudinal study also suggests that exercise and ongoing rehabilitation may help manage fatigue following long COVID [18]. Additionally, TR has been described as a valuable tool in managing fatigue among patients who are dealing with long COVID and has the potential to improve their quality of life (QoL) significantly [19]. Consistent with these findings, our study showed that both HBR and TR significantly improved fatigue compared to HP in within-group comparisons. A quasi-experimental study further supports these results, indicating that rehabilitation significantly reduced fatigue in long COVID after 7 days of intervention and 12 weeks of follow-up [20]. Similarly, a pilot telehealth programme recommends physiotherapy and rehabilitation as effective approaches for minimising fatigue associated with long COVID [21]. In contrast many previous studies that evaluated single rehabilitation modalities, our trial directly compared HBR, TR, and HP within the same RCT framework, thereby strengthening the evidence for structured rehabilitation in fatigue management.

Most long COVID patients have impaired cardiorespiratory function [22], which contributes to reduced quality of life and increased disability. Several studies have reported a decline in the quality of life and an increase in cardiorespiratory disability, suggesting a need for targeted cardiorespiratory rehabilitation post-COVID infections [22]. Cardiorespiratory rehabilitation is crucial for managing lung disease [23] and cardiovascular issues. Our study found that patients undergoing either HBR or TR experienced significant improvements in cardiorespiratory function (p < 0.05) compared to those who followed a HP. Specifically, improvements were observed in PR, FRC, and SaO₂. These findings suggest that cardiorespiratory rehabilitation can reduce cardiac and respiratory complications while improving cardiorespiratory capacity, volume, and endurance in long COVID patients [22]. It is noteworthy that exercise-based rehabilitation in a specialised setting is highly effective in managing cardiorespiratory symptoms in long COVID patients [22]. Similarly, a study has demonstrated TR’s effectiveness in addressing the cardiorespiratory symptoms in long COVID patients [24]. In contrast, a previous study reported no improvement in cardiorespiratory function in a control group 12 weeks after diagnosis [25], which is comparable to the limited changes observed in our HP group.

Long COVID symptoms can have a profound impact on physical function and mental health. It is important to alleviate these symptoms and work towards improving overall well-being and recovery [26]. A recent study examined physical functioning in individuals who had recovered from SARS-CoV-2 infection [27]. HBR remains crucial for addressing both the physical and psychological needs of long COVID patients [28], and telerehabilitation-based exercise can potentially enhance the physical functioning in long COVID patients [29]. Our findings are consistent with these reports, demonstrating significant improvements in physical functioning in both HBR and TR groups compared with HP.

Long COVID often presents with relapsing and remitting symptoms, resulting in an episodic pattern of disability. This recurring pattern of symptoms can significantly impair daily functioning and quality of life [30]. The experiences of participants in the context of episodic disability were influenced by various factors that played a crucial role in shaping their disability. These factors were found to be integral to their experiences and emerged as a significant component of their episodic disability [31]. According to a recent study, it has been found that people living with long COVID experience a range of disabilities. These disabilities can be categorised into different dimensions, such as physical, cognitive, mental, emotional, and social aspects [32]. In our study, both HBR and TR significantly reduced episodic disability, whereas the HP group demonstrated worsening disability over time. To effectively manage long COVID, it is crucial to include exercise-based therapy as a part of the treatment plan. Telerehabilitation offers a flexible and accessible method for delivering rehabilitation services remotely while maintaining clinical effectiveness [29]. By providing a hospital-based approach, rehabilitation helps these individuals set achievable goals and aims to prevent, mitigate, and address episodic disabilities [30]. So, the findings of this study provide integrated evidence that structured rehabilitation whether delivered in hospital settings or through telehealth platforms produces consistent benefits across multiple domains of long COVID. The comparable effectiveness of HBR and TR suggests that telerehabilitation may serve as a feasible and scalable alternative to hospital-based care, particularly in resource-limited settings.

In our trail the home programme (HP) group showed less improvements, prior research has indicated that structured home-based rehabilitation can be effective in patients who recovered with COVID-19 [33]. Recent findings indicate that supervised home-based exercise interventions can be effective in improving functional capacity and quality of life [33]. The comparatively limited effects observed in our HP group may be explained by reduced supervision, lower adherence, limited exercise progression, or lack of real-time feedback. In contrast to HBR and TR, which incorporated structured monitoring and professional interaction, the HP model depended primarily on patient self-management. Variations in intensity, supervision, and accountability may explore the inferior outcomes identified in this study.

This study is the first large-scale trial to address long COVID symptoms with physiotherapy and rehabilitation. The strengths of this study include its assessor-blinded RCT design and recruitment from a diverse population across eight administrative divisions in Bangladesh. This approach is particularly significant given the severe impact of long COVID on patients’ quality of life and the current lack of effective treatment options. However, several limitations must be acknowledged. Due to the delay in conducting the study, there were challenges in maintaining follow-up with all participants. We also observe some selection bias during referral and multicentre rehabilitation. Additionally, while some cardiorespiratory parameters were measured using digital meters, these may have had variable reliability and validity. Despite these limitations, this study provides a well-defined approach to managing long COVID in a large sample size and offers important direction for future research and clinical practice.

Conclusion

Following an intervention, this study demonstrated the significant benefits of hospital-based rehabilitation and telerehabilitation programmes compared to home-based self-practice with the guidance. Both hospital-based and telerehabilitation interventions significantly improved fatigue levels, physical functioning, and episodic disability. However, home programmes showed comparable efficacy. This study emphasises the importance of integrating structured rehabilitation into long COVID care frameworks to mitigate the impact on individual health, productivity, and the global economy.

Abbreviations

LC

Long COVID

QoL

Quality of life

HBR

Hospital-based rehabilitation

TR

Telerehabilitation

HP

Home programme

APTE

Adapted physical activity and therapeutic exercise

RCT

Randomised clinical trial

CI

Confidential interval

Authors’ contributions

MFK, KNY, MSJ, FBA, SJ, ER, AHG, SA, TH, RA and MAH equally participated in the conceptualization of the study, data curation and collection, validation, data visualization, model construction, formal analysis, manuscript writing, manuscript review and editing, and manuscript approval.

Funding

This work received partial funding for data collection, intervention, and data auditing from a doctoral grant provided by Jashore University of Science and Technology (JUST) under the University Grants Commission Bangladesh (UGC). Grant No: 21-FOHS-1.

Data availability

The data from this trial was uploaded and is accessible through an open data repository (Kabir, Dr Md Feroz (2024), “RCT_Long COVID_Physiotherapy”, Mendeley Data, V1, 10.17632/5hsb3fvmvk.1).

Declarations

Ethics approval and consent to participate

This study obtained ethical approval from the ethical review board of Universiti Malaysia Sabah (UMS/FPSK6.9/100-6/95), and the trial was registered with the clinical trial registry of India (CTRI/2023/03/050808; registered on 17/03/2023). This study obtained written informed consent for participants and their legal guardians.

Patient and public involvement

No.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.