A scoping review of the feasibility, acceptability, and effects of physiotherapy delivered remotely

Abstract

Purpose

To review the feasibility, acceptability, and effects of physiotherapy when delivered remotely.

Materials and methods

CINAHL, MEDLINE, EBM Reviews, and Cochrane Library databases (January 2015–February 2022) were searched and screened for papers (of any design) investigating remote physiotherapy. Data were extracted by two independent raters. Methodological quality of the identified papers was not assessed. Thematic content analysis drew out the key issues.

Results

Forty-one papers (including nine systemic reviews and six with meta-analyses) were selected involving musculoskeletal, stroke and neurological, pulmonary, and cardiac conditions. The most commonly delivered intervention was remote exercise provision, usually following assessment which was completed in-person. All studies, which assessed it, found that remote physiotherapy was comparably effective to in-person delivery at lower cost. Patient satisfaction was high, they found remote physiotherapy to be more accessible and convenient. It boosted confidence and motivation by reminding patients when and how to exercise but adherence was mixed. No adverse events were reported. Barriers related to access to the technology; technical problems and concerns about therapists’ workload.

Conclusions

Remote physiotherapy is safe, feasible, and acceptable to patients. Its effects are comparable with traditional care at lower cost.

IMPLICATIONS FOR REHABILITATION

• Remote physiotherapy is safe, feasible, and acceptable to patients with comparable effects to in-person care.

• Remote delivery increases access to physiotherapy especially for those who cannot travel to a treatment facility whether due to distance or disability.

• Remote physiotherapy may increase adherence to exercise by reminding patients when and how to exercise.

• Remote physiotherapy does not suit everyone, thus a hybrid system with both in-person and remote delivery may be most effective.

Keywords:

• Physiotherapy
• telerehabilitation
• telehealth

Introduction

The COVID-19 pandemic has challenged health care professionals to rapidly change the care they provide and how it is provided. In-patient services were redeploying staff to expanded acute care; patients were discharged earlier to reduce risk of infection and to expand capacity for acute care. Out-patient- and community-based services were often closed and staff redeployed to in-patient care. However, there was also a desire to continue with “business as usual” where possible and many services rapidly started to deliver care remotely without face-to-face contact. We defined remote physiotherapy as those delivered using technologies such as the telephone, video conferencing, apps and web-based systems, rather than in-person [1,2]. With so many changes, which were often introduced under urgent pressure of time and other demands, it is important not to “reinvent the wheel” but use pre-existing resources and evidence. Remote physiotherapy is not new, it has been an area of interest over recent years, particularly in remote and rural areas [3–5] and as a way to triage assessment and management in high volume musculoskeletal services [6–8]. As part of a project to evaluate the implementation of remotely delivered physiotherapy in response to the COVID pandemic, we undertook a scoping review to explore the evidence-base for physiotherapy delivered remotely. As our interest was in implementation of remote physiotherapy as well as its effectiveness, our objectives were to investigate the evidence for outcomes, and feasibility and acceptability in terms of patient satisfaction; advantages and facilitators; adherence; safety and barriers to implementation.

Method

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) scoping review guidelines [9,10] were used as a framework for the review. The review was registered with Open Science Framework (http://osf.io/up62r/). Three electronic databases (CINAHL, MEDLINE, and the Cochrane library) were searched using key words relating to physiotherapy, remote delivery and experiences or outcomes (Table 1). We selected primary research studies or systematic reviews of any design in English which investigated delivery of remote physiotherapy from 2015 (to ensure currency) to February 2022. We excluded studies which delivered electronic treatment in clinic (such as virtual reality); interventions which were not delivered specifically by physiotherapists (e.g., nurse-led cardiac care/rehabilitation); or those which did not examine the effects of remote physiotherapy, such as studies to develop technologies or studies of support to promote exercise after discharge from rehabilitation. Titles, abstracts, and then full text were independently screened by two of the authors against the selection criteria. Reference lists of selected papers were also screened for any other papers which met the selection criteria.

Table 1. Search terms.

Details of the selected papers were extracted and tabulated according to the condition being treated. In keeping with the scoping review approach, the methodological quality of the selected papers was not assessed. Then thematic content analysis was undertaken. Two of the authors independently reviewed each paper and identified and tabulated content regarding each objective (outcomes, feasibility, and acceptability of implementing remote physiotherapy. They then independently coded the material collated under each theme into categories and sub-categories before meeting to revise and iteratively consolidate the information therein. Any differences in interpretation were resolved through discussion.

Results

A total of 2727 papers were identified through the searches. After duplicates were removed and titles and abstracts screened, 540 papers were downloaded for full text screening, with a final 41 papers included in the review [11–57] (Figure 1). Musculoskeletal problems were the most common condition addressed (n = 13, [11–22]). Nine papers covered stroke and neurological rehabilitation [23–31]; 12 pulmonary disease [33–44], 6 looked at cardiac rehabilitation [45–50], and 8 looked at generic physiotherapy rather than a specific condition [50–57]. A range of study designs were used; with nine systematic reviews [11,12,23–25,32,33,44,45]. Six of which found sufficiently homogenous data to perform meta-analysis [12,24,25,33,34,45]. There were also eight randomised controlled trials [13–15,26,34–37] and two clinical controlled trials [16,46]. Fourteen were cohort studies [17,18,28,29,38–41,46, 54] and two were service evaluations [27,50]. There were five surveys [19,20,42,46,48] using questionnaires or mixed methods, one focus group study [43] and four semi-structured interviews [21,22,30,31]. Further details of the selected studies are found in Tables 2–6.

Figure 1. PRISMA flow diagram.

Table 2. Selected studies of musculoskeletal conditions.

Author and country	Main study aim	Study design	Sample	Intervention	Findings
Grona et al. 2018 [11] International	To determine the health, system and process outcomes of remote PT in chronic musculoskeletal disorders	Systematic review. Medline, CINAHL, PsycInfo, and Embase searched for trials and cohort studies on remote PT using secure video-conferencing January 2003–December 2016. Quality analysis using standardised tools.	17 studies were selected. Eight assessed reliability and validity of remote PT assessment (n = 121). Nine were intervention studies: two case studies, one qualitative evaluation, three RCTs, one retrospective and one pre-experimental and one controlled cohort design. Controls were usual care	Secure video-conferencing	Outcomes between remote and traditional care were comparable in terms of pain, function, quality of life, and impairments remote PT compared with usual care. Patient satisfaction was high in the three studies in which it was measured.
Shukla et al. 2016 India [12]	To evaluate the effectiveness of remote PT in after total knee arthroplasty	Systematic review and meta-analysis. Embase, PubMed, and Cochrane were searched for studies in English 2000–2014. Outcomes = ROM, strength; balance, mobility, pain, stiffness, and function	Six (n = 408) trials. Controls = usual in-person care.	Video-conferencing delivered by experienced PTs.	Patients reported high levels of satisfaction. There were comparable outcomes in knee range of movement, quadriceps strength, pain, stiffness, physical activity, and functional status between remote PT and usual care. Different study designs and outcomes prevented pooled analysis and methodological quality was mixed.
Bini and Mahajan 2017 USA [13]	To compare physical therapy delivered through an asynchronous video-based tool to traditional in-person outpatient physiotherapy following knee replacement	Randomised controlled trial. Control = usual care. Outcomes were measured 3 months after surgery, pain, health status, and function. Satisfaction, adherence, and resource utilisation were also assessed.	51 patients were recruited. 14 randomised to intervention and 15 to usual care. Mean age = 62 years. 29 patients completed the final survey.	Asynchonous video-based PT with 23 exercise videos (same as in-person clinic) on an ipad. Exercise videos were sent to the patient who recorded themselves completing it. The PT reviewed and progressed it with advice as needed. Other HCPs and the patient could also review the videos. Technical support was available. Patients could request in-person visits at any time or return to usual care.	29/51 (57%) completed. There were no significant differences in any clinical outcome or satisfaction between groups. Overall cost of hospital-based remote PT was 60% less than in-person therapy, because of lower therapy staff resources. Patients were willing and able to actively participate by taking their own videos and submitting them for review.
Bettger et al. 2020 USA [14]	To evaluate the effect of remote PT on costs, effectiveness, and safety of remote PT vs. usual care following knee replacement.	RCT. Pre-surgery patients were randomised to remote or usual (home or out-patient) PT. Primary outcome = total health-care costs. Secondary outcomes = function; knee ROM; gait speed; and adverse events.	306 patients (mean age, 65 years; 62.5% women) recruited from four sites, randomised November 2016 to November 2017. 287 (143 in the remote PT group and 144 in the usual care group) completed the trial.	Cloud-based remote PT system using 3D tracking technology to produce an avatar that demonstrates and instructs exercises with immediate feedback on exercise quality. Plus synchronous PT visits who prescribed individualised therapy before surgery and then weekly. Activity, exercise quality, and adherence were monitored by the remote therapist asynchronously.	Remote PT had lower costs at 12 weeks after discharge than usual care (median, US$1050 vs. $2805; p < 0.001) with fewer re-hospitalisations (12 vs. 30; p = 0.007). Outcomes were comparable in terms of function at 6 weeks and 12 weeks; knee range of movement and gait speed, pain, and falls.
Levinger et al. 2017 Australia [15]	The feasibility of a 3-month Internet-based remote PT on recovery following ACL reconstruction in terms of knee pain, function, self-efficacy, and fear of pain were also assessed.	Pilot randomised controlled trial assessed before surgery and 3-month follow-up. Control = usual care.	32 (16 in each group) participants were recruited. Mean age of 29.2 ± 7.4 years; 9 females; 23 males. Fifteen dropped out (38% and 56% from the intervention and control groups, respectively). Data from only 17 participants were available for analysis (n = 10 intervention group; mean age of 32.2 ± 10.2 years, 4 females and n = 7 control group; mean age of 28.5 ± 9.1 years, 4 females).	Intervention = usual post-surgery rehabilitation plus open-access interactive website-support comprising information about ACL reconstruction, recovery milestones, how to overcome problems, recommended exercises, and reminders. Communication between patients and therapists via regular text or emails questions on symptoms and progress from which the PT identified any problems, gave additional advice or treatment.	Remote PT was considered useful and important for information and reminders to exercise (mean 7.75/10). Reviewing the exercises (pictures and videos) reminded patients how to do them. They valued how it supplemented rather than replaced contact with the PT. But adherence was low. No significant differences between the groups over time in any measures of knee pain, function, self-efficacy, and fear of pain were seen.
Correia et al. 2018 Portugal [16]	To compare remote PT and usual care after knee replacement. To assess uptake and safety.	Non-randomised controlled cohort feasibility study. Single centre. Control = usual care. Outcomes = balance mobility, function, range of knee movement.	69 patients were recruited (37 to remote PT, 32 to control) according to their address. 10 did not complete the study; 8 in the intervention (21% dropout rate) and two in the control (7% dropout rate).	A digital biofeedback system for home PT using body-attached movement trackers which provided real-time feedback on performance while exercised independently at home through a mobile app. Plus a web-based platform in which PTs prescribed, monitored and adapted treatment remotely. The number of “correct” exercise repetitions contribute to patients’ exercise goals.	Intervention group reported superiority for all outcomes at all time points with less human resource. Treatment time was > in the intervention group (median 31.5 vs. 24 h, p = 0.005) with less PT contact: controls received 24 in-person sessions. Intervention group received 3 in-person and two phone contacts, plus (mean) 0.4 (range 0–2) assistance contacts and 2.5 extra therapist calls (range 1–12). 60% needed help with the equipment.
Malliaras et al. [17] International	To investigate use and views towards remote therapy among allied health clinicians treating people with musculoskeletal conditions during the pandemic.	Cross-sectional international survey of demographics, professional factors, telehealth use, attitudes, perceived barriers, and enablers. Content analysis based on the Capability-Opportunity-Motivation Behavioural Model.	827 allied health clinicians who used telehealth to manage musculoskeletal conditions during the coronavirus pandemic. Mostly PTs (82%) working in Australia (70%).	Telehealth to manage musculoskeletal conditions during the coronavirus pandemic	71% reported less income (mean 62% SD 25%) since the pandemic. % of people seen remotely increased from 0% pre (IQR 0–1) to 60% (IQR 10–100). 42% found remote as effective as face-to-face care. 25% felt patients value remote care as much as in-person. 21% had sufficient telehealth training. Lack of physical contact hampered accurate diagnosis and management.
Hoogland et al. 2019 Netherlands [18]	To evaluate the feasibility and patient experience of remote PT after hip replacement	6-Month prospective cohort study	30 independently living adults (18–75 years) who had undergone hip replacement for osteoarthritis. December 2015 and February 2017	Patients followed a 12-week strengthening and mobility exercise program (at least 5×/week) with video instructions on a tablet PC and daily physical activity monitoring via a motion sensor worn around the neck. There were weekly phone contact with a PT.	26 patients (87%) completed. Average adherence to exercising 5×/week =92%. Main reasons for non-adherence = holiday/“day off” (25%) and “work” (15%). 8 technical problems were reported. Mean user evaluation score = 4.55/5.
Boissy et al. 2016 Canada [19]	To record the use and reliability of remote PT following knee replacement; PTs satisfaction and its impact on clinical objectives.	Cohort study embedded in a randomised controlled trial	97 patients who received 1431 remote PT sessions post-surgery.	Intervention = Remote PT via videoconferencing comprising pre- and post-exercise assessment (structured interview and observation), supervised exercises ∼30 min (mobility, strengthening, function, and balance), home exercises to perform on days without supervised sessions, an advice on pain control, walking aids, and the return to activities.	75% needed a new Internet connection. Mean time to un/install the technology = 308 min. 97% of planned sessions were delivered, 21% of which required reconnection. 43% of sessions needed remote technical support but impact on communications was minimal. The technical environment was good or acceptable in 96% of sessions. Clinical objectives were met in 99% of sessions.
Cottrell et al. 2018 Australia [20]	To identify patients’ interest in remote PT and barriers to accessing remote PT services.	Cross-sectional survey	A convenience sample of 20 participants with chronic musculoskeletal conditions attending traditional orthopaedic out-patient PT services (n = 6) between November 2015 and April 2016.	Participants were receiving traditional clinical-based care. Asked about their interest in, and feasibility of remote PT	85 patients (71%) participated. >1/2 were willing to use remote PT if it reduced the costs (53%) and time (57%) of treatment. 43% would prefer remote PT over travelling to appointments. Working patients were more likely (65%) to be willing to use remote PT if it reduced work absence. 78% of patients had the technology to enable remote PT at home.
Martinez de la Cal et al. 2021 [21].	To explore PTs views of the efficacy and disadvantages of remote PT for chronic low back pain	Semi-structured interview study	19 physical therapists from academic and healthcare fields in both the public and private sector treating people with chronic low back pain.	Web-based PT for low back pain	PTs believed remote care could improve access and reduce costs, but was only successful if patients engaged and the exercises were individualised to them. It also enabled more follow-up but long-term adherence to treatment can be challenging.
Abramsky et al. Canada [22]	To explore patients’ perspectives on mobile-based home exercise programme	Semi-structured, in-person interviews	10 participants (7 women, 3 men aged 22–55 years) who received remote PT for a musculoskeletal condition from an out-patient-based private PT practice between December 2015 and May 2016.	A mobile application (Embodia) featuring exercise notification reminders, ability to track symptoms, video-based exercises and education, and instant messaging with the PT. Used for at least a month.	Patients were generally positive about the app, preferring it to paper handouts as it reinforced the correct exercise technique and engagement. Half of the participants found the exercise reminders useful if they were customised to fit into daily routines. The app fostered an interactive, therapeutic relationship between patient and PT. Some said the app extended but did not replace their PT.

ACL: anterior cruciate ligament; MCID : minimum clinically important difference; RCT: randomised controlled trial; PT: physiotherapy or physiotherapist; ROM: range of movement; HCP: health care professionals.

Table 3. Selected studies of stroke and neurological conditions.

Author and country	Main study aim	Study design	Sample	Intervention	Findings
Tchero et al. 2018 France [23]	To investigate the efficacy of remote PT in people with stroke	Systematic Review and meta-analysis. MEDLINE, Cochrane Central, and Web of Science databases were searched for RCT of remote PT in stroke. Continuous data were extracted for relevant outcomes and analysed as the standardised mean difference and 95% CI in a fixed-effect meta-analysis model. Control = usual care	15 studies (1339 patients) selected with 12 included in the pooled analysis.	Remote PT methods were varied. Some studies used only telephone calls, while others used videoconferencing, educational videos, Web-based chats, and virtual reality systems	Changes in ADL (BI SMD −0.05, 95% CI −0.18 to 0.08) balance (Berg SMD −0.04, 95% CI −0.34 to 0.26); UL function (FMA UL scale SMD 0.50, 95% CI −0.09 to 1.09); and QoL (SIS mobility subscale; SMD 0.18, 95% CI −0.13 to 0.48), Caregivers Strain Index and satisfaction were comparable. One study showed cost of remote PT was < usual care by US$ 867. Advantages were easier access to PT, and (self) monitoring for disabled patients. Barriers = administrative licensing, medico-legal ambiguity, and financial sustainability.
Laver et al. 2020 Australia [24]	To determine the efficacy of remote PT for ADL in stroke survivors compared with (1) in-person rehabilitation or (2) no rehabilitation or usual care.	Cochrane Systematic review and meta-analysis. Cochrane library, MEDLINE, Embase, and eight additional databases were searched for RCTS of remote physiotherapy in stroke compared with in-person or no treatment.	Included 22 trials (n = 1937, sample size in selected studies = 10–536). Reporting quality was often poor for randomisation and concealed allocation. Selective outcome reporting and incomplete outcome data were apparent in several studies.	Interventions included telephone (n = 8), videoconferencing (n = 11) or a combination of phone, video, texts, email, and /or a website (n = 3) for post-discharge support, upper limb, lower limb, and mobility retraining. Studies were either on discharge from hospital or with people in the subacute or chronic phases	No evidence of any differences between remote PT and either control. Evidence was moderate for ADL (SMD-0.00, 95% CI 0.15 to 0.15); QoL (SMD 0.03, 95% CI −0.14 to 0.20) and mood (SMD −0.04, 95% CI −0.19 to 0.11), and low for ADL between remote and in-person (SMD 0.03, 95% CI −0.43 to 0.48) balance (SMD 0.08, 95% CI −0.30 to 0.46); UL function (MD 1.23, 95%CI −2.17 to 4.64). Five studies reported remote PT utilisation or costs were < usual care. Two studies assessed adverse events, with none reported.
Ramage et al. 2021 [25]. International	To map the safety, efficacy, and feasibility of remotely supervised weight-bearing or standing activities for people after stroke.	Scoping review of stroke exercise and telehealth. Two reviewers independently appraised all studies. Data were charted by one reviewer, checked by another, and results synthesised narratively.	7 studies (2 randomised trials, 1 mixed-methods, and 4 pre–post studies) involving 179 stroke survivors, some with cognitive impairment, and 2 (29%) excluded those who could not walk independently.	Interventions involving activities in weight-bearing or standing positions, supervised in real-time remotely were included.	Adherence (n = 3) and satisfaction (n = 4) were good with no serious adverse events (n = 4). Strategies to overcome technical barriers to optimise safety and feasibility were physiological monitoring, caregiver assistance, and in-person exercise prescription.
Burridge et al. 2017 UK [26]	To asses feasibility and acceptability of remotely delivered self-directed exercises and ADL in people with stroke	Feasibility RCT including semi-structured interviews. Assessed before, after 3 weeks of intensive daily treatment and 6-month follow-up.	19 stroke survivors (from 83 screened) recruited: 11 were randomly allocated to the intervention group and 8 to the control group (usual care).	Participants were taught how to do constraint induced movement therapy and then used LifeCIT (web-based constraint-induced movement programme) to continue the therapy. Participants had access to it for 21 days and asked to use it 5 days/week.	Intervention was well accepted with changes in UL function > MCID at both assessment points. Positive effects on self-efficacy and confidence to use the affected UL. Mean treatment time was 3.2 (SD 1.7) h/day on 13.6 (SD 2.1) days. No adverse events
Yang et al. 2021 [27] Canada	We studied the implementation and the effectiveness of the virtual Graded Repetitive Arm Supplementary Program (GRASP) delivered and evaluated via videoconferencing in individuals with stroke.	Mixed methods service evaluation using the Reach, Effectiveness, Adoption, Implementation, Maintenance (RE-AIM) framework to evaluate the implementation of the GRASP program.	17 individuals with stroke who participated in the virtual GRASP programme delivered by a non-profit organisation during the pandemic.	Graded Repetitive Arm Supplementary Program (GRASP) delivered and evaluated via videoconferencing in	Reach: 65% participated. Effectiveness: significant improvement in UL post-treatment. Adoption: uptake facilitators = well-planned, thorough implementation, screening, and feedback from audits. Staff were comfortable using videoconferencing despite technical difficulties. Factors for ongoing participation=: patients liked the group, instructor’s encouragement and the improvements. Implementation was as intended (fidelity checklist). Adherence was high. Maintenance: the organisation continued to offer the program.
Held et al. 2018 Europe [28]	To assess the safety, usability, and patient acceptance of remote PT for balance and gait in stroke survivors’ homes	Cohort study. Patients were assessed before and after using the Rewire system for 12 weeks	Fifteen stroke survivors (6 in Switzerland and 9 in Spain) with mild to moderate lower limb impairments who used the REWIRE system for 12 weeks. Assessed before and after	rEWirE = asynchronous remote balance and mobility training monitoring by staff. The platform = patient’s station (virtual reality system using Kinect camera, TV, and force plate), hospital station (cloud-based service to review, schedule and personalise treatment sessions), and a networking station (at the health provider site) which mines data for common features and trends	Patients completed 71% (range 39–92%) of the scheduled sessions and reported excellent scores on the Technology Acceptance Measure. Mean training duration/week = 99 ± 53 min. No adverse events were recorded.
Kizony et al. 2017 Israel [29]	To retrospectively document the effects of a remote PT program.	Retrospective cohort study assessed before and after 8 week intervention with clinical assessment of UL movements and function. Usability measures using the System Usability Scale and a focus group with clients	N = 82 (46 men), aged 22–85 years, 63 (76.8%) with stroke, 6 with traumatic brain injury, 6 = other; 8 had multiple sclerosis. Clients had mild–moderate upper limb impairment. Most did not use their weak upper limb for ADL. Recruited September 2013 to August 2015.	CogiMotion provides long-term PT to improve UL impairments and function after discharge. It is a hybrid synchronous–asynchronous system using Kinect 3D sensor in client’s home connected to clinicians’ computer. The client selects from ∼20 interactive games and tasks using arm, leg and trunk movements to control the games. They are taught to use two games during 1st session then progressed according to client’s abilities.	Only 26% completed Usability Scale but they found it usable (mean ± SD 89.1 ± 12.1) and enjoyable (mean ± SD 4.1 ± 1.1). Only half were assessed after treatment showing improvements in UL impairments (FMA n = 35, p = 0.002, shoulder flexion n = 42 p = 0.003) but not function.
Tyagi et al. 2018 [30] Singapore	To explore the stroke survivors’, caregivers’, and therapists’ barriers and facilitators of remote PT	Semi-structured in-depth interviews and focus group discussions.	Participants (n = 37) including stroke patients, their caregivers, and PTs delivering remote care	iPad-based system to deliver individualised exercise programme using resistance bands. The system comprised ipad (with exercises), exercise bands, portable sensors and interface with instructions, displays of the exercises and instant feedback (to patient and therapist) on performance via the sensors. Patients were asked to exercise 5×/week plus weekly FaceTime “visits” with remote PT.	Facilitators for patients = affordability and access. For PTs, it filled a service gap. Barriers for patients = setup difficulties; PTs-patient assessments were difficult, interface problems. Both limited scope of exercises and connectivity problems.
Muller et al. 2015 UK [31]	Participants’ experience of remote vestibular rehabilitation ± expert telephone support.	Semi-structured interviews	33 participants (10 men; mean age 27–84 years) with chronic dizziness recruited from primary care as part of a RCT.	Booklet-based vestibular rehabilitation, with (intervention) or without (control) expert telephone support	Phone support group felt they had a genuine relationship with the PT within three short sessions, and found the PT reassuring, encouraging and motivational.

RCT: randomised controlled trial; SMD: standardised mean difference; ADL: activities of daily living; BI: Barthel Index; UL: upper limb; FMA: Fugl-Meyer Scale; QoL: quality of life; SIS: Stroke Impact Scale; PT: physiotherapy or physiotherapist.

Table 4. Selected studies of chronic obstructive pulmonary disease.

Author and country	Main study aim	Study design	Sample	Intervention	Findings
Cox et al. 2021 [32] International	To determine the effectiveness and safety of remote PT for people with chronic respiratory disease	Cochrane Review: Searched Cochrane Airways, Central Trials Registers, MEDLINE and Embase to Nov 2020 for RCTs and CCTs. Cochrane risk of bias tool and GRADE used.	15 studies (32 reports, 3 controlled trials, the rest were RCTs) with 1904 participants. 99% COPD.	Remote PT for the delivery of pulmonary rehabilitation including exercise training, with at least 50% of the intervention delivered remotely. Compared with traditional in-person (centre-based) pulmonary rehabilitation, or no rehabilitation.	4 studies (n = 556) had moderate evidence of no difference between remote and in-person PT for exercise capacity (6MWD MD 0.06 m 95%CI −10.82 m to 10.94 m), QoL life (SGRQ MD= −1.26, 95%CI −3.97 to 1.45; low evidence) for breathlessness CRQ MD = 0.13, 95%CI −0.13 to 0.40). Participants were more likely to complete remote PT (93%, 95%CI 90- 96%) that in-patient (70%) care. No adverse effects were noted.
Bonnevie et al. 2021 [33] International	Effectiveness of home-based exercise using advanced telehealth technology	Systematic review and meta-analysis of randomised trials.	15 trials of 1522 people with stable chronic obstructive pulmonary disease (COPD) referred for exercise therapy (ET)	ET delivered using advanced telehealth technology (ATT-ET) compared with: no exercise therapy (ET), in/outpatient ET, and home-based ET without ATT?	Remote exercise had a similar effect to usual care.
Hansen et al. 2020 [34] and Godtfredsen et al. 2020 [35] Denmark	Investigate whether remote pulmonary rehabilitation is superior to conventional care at 3, 6, and 12 months	Single-blinded, multicentre (8 hospitals) randomised controlled trial. Assessed at baseline, end of intervention and at 22 weeks and 12 months follow-up.	134 people with COPD and FEV1 < 50%, eligible for routine hospital-based, outpatient PR. 74 females, mean age 68 ± 9 years, FEV1 33%±9% predicted, 6MWD 327 ± 103 m	10 weeks of groups-based remote pulmonary rehab (60 min, three times weekly) or conventional PR (90 min, two times weekly)	No between-group differences for changes in 6MWD after intervention (9.2 m 95%CI: −6.6 to 24.9) or follow-up (–5.3 m 95% CI: −28.9 to 18.3). More completed the remote intervention (n = 57 vs. 43 p < 0.01). No significant differences between or within groups in the 6MWD one year after completion of the programme
Cerdán-de-Las-Heras et al. 2021 [36] Denmark	To compare the effect of a remote PT program vs. standard rehabilitation on COPD	Non-inferiority RCT. 6MWTD, 7-day pedometry, QoL, adherence, patient satisfaction and safety assessed before and after treatment and 3 and 6 months follow-up.	54 People with stable COPD (27 in tele-rehab group). 70 ± 9 years, male 57%, FEV1% 34.53 ± 11.67, FVC% 68.8 ± 18.81, 6MWT 376.23 ± 92.02)	8 weeks of remote PT [physiotherapist video/chat-consultations and workout sessions with a virtual-autonomous-PT-agent (VAPA)] Control: standard rehabilitation	Non-inferiority in change in 6MWTD at 8 weeks (47.4 ± 31.4), and at 3 (56.0 ± 38.0) and 6 (95.2 ± 47.1) months follow-up was observed. No differences were observed in 7-day pedometry or quality of life. Adherence was 81%; and patient satisfaction was 4.27 ± 0.77 (Likert scale 0–5).
Cerdan-de-Las-Heras 2021 [37] Denmark	Investigated the usefulness and effectiveness of remote PT on exercise capacity.	Randomised controlled trial assessed after treatment, 3 and 6 months follow up.	29 patients with stable Idiopathic Pulmonary Fibrosis (15 intervention, 14 control); aged 71 yrs; male 72.4%, FVC% 83.5, 6MWT 468.4 m.	3 months of tele-rehabilitation: video and chat consultations with a PT and workout sessions with a virtual PT agent (VAPA)	Exercise capacity was > at follow-up compared with control: 6MWTD at 3 (+39.5 m (p = 0.03), 6 (+34.3 m (p = 0.02), and 9 months (+40.5 m (p = 0.15) follow-up. No difference in pedometry or QoL. Adherence was > 63%. Patient satisfaction was high.
Hoaas et al. (2016) and Zanaboni et al. (2017) Norway (two publications from the same study) [38,39]	To explore feasibility, acceptability, adherence to, experience, effects and resource use of long-term remote PT for COPD.	Cohort study over two years of remote PT. Assessed before and after 1 and 2 years using objective measures of adherence; COPD severity, activity and health care utilisation. Plus interviews and focus groups	10 Patients (5 men mean age 55 years) with moderate to severe COPD were recruited after a four week in-patient pulmonary rehabilitation programme	Patients used a treadmill, pulse oximetre and ipad for individualised interval treadmill training and strength exercises 3×/week. Electronic daily record of symptoms, O2 saturation, breathlessness and sputum production, monitored by PT via a website. Also self-management advice and weekly video-conferencing.	No drops out over 2 years. Mean exercise sessions/week = 1.7. Average diary completion = 43%. 56% adherence to online training. Satisfaction with the technology generally high. Increased self-efficacy, and motivation reported. Activity, lung capacity, health status and QoL maintained at two years. 32% reduction in health care utilisation vs. year before intervention due to fewer COPD-related hospitalisations and outpatients visits.
Paneroni et al. 2015 Italy [40]	Feasibility and acceptability of remote PT (pulmonary rehabilitation)	Controlled cohort study (multi-centre)	36 adults with COPD (age 66 yrs, 86% men) completed out-patient pulmonary rehabilitation e. 18 changed to the remote programme and 18 continued with usual care	Daily exercise using an interactive programme and cycle ergometer plus PT contact by phone or video to collect clinical data and encourage and supervise the exercise. These contacts tapered off to encourage independent exercise.	Mean = 28 remote sessions over 40 days, lasting 100 min including strength and cycle ergometer training. No adverse events. 84% were satisfied with the remote PT. Improvements in walking capacity, dyspnoea and quality of life were comparable (p > 0.05) between remote and traditional PT.
Marquis et al. 2015 Canada [41]	to assess effects of, satisfaction with, and adherence to remote pulmonary rehabilitation	Cohort study. Assessing changes in exercise tolerance (6-min walk test and cycle endurance test) and quality of life before and after intervention	26 patients (15 women, mean age 65 years) with moderate to very severe COPD.	15 remote PT sessions (ergo cycling and strength training 3×/week) over 8 weeks via videoconferencing, wireless oximetre and heart rate sensor plus education via self-learning modules.	Significant improvements post- intervention in mobility (p < 0.001), exercise capacity (p < 0.005), dyspnoea (p < 0.001), fatigue (p = 0.002), and emotion (p = 0.002) domains of the quality of life. Participants’ satisfaction and adherence were very high.
Inskip et al. 2018 Canada [42]	Explore the views of patients and PTs on remote pulmonary rehabilitation	Cross-sectional survey using questionnaires and focus groups	26 patients with chronic lung disease (50% men; mean age 72 years; 73% COPD) and 26 respiratory PTs (23 women; mean age = 43 years). None had experience of remote rehabilitation.	None—participants were asked about their needs for, and the necessary features of remotely delivered pulmonary rehabilitation.	Most regularly used technology but some were uncomfortable with it. Both groups valued the social aspect of groups and interaction with PT. Heart rate, oxygen saturation while exercising, and individualised exercises were key. Staff worried about workload if remote work was added to their regular duties, and the need to be technically adept. They appreciated remote PT to increase access to rehabilitation services for underserved communities
Knox et al. 2021 [43] UK	To understand stakeholders’ experiences of virtual pulmonary rehabilitation	Five focus groups. Reflexive thematic analysis was used to analyse the transcripts.	people with chronic lung disease attending virtual PR Plus staff running of the PR	Virtual pulmonary rehabilitation	Participants were positive about the virtual programme, discussing the trade-off between convenience and relatedness. Staff described barriers to the service and solutions including greater training, organisation and understanding of risk.

COPD: chronic obstructive pulmonary disease; 6MWD: 6-minute walking distance; SGRQ: St George’s Respiratory Questionnaire; CRQ: Chronic Respiratory Questionnaire; PT: physiotherapy or physiotherapist; QoL: quality of life.

Table 5. Selected studies of cardiac conditions.

Rawstorn et al. 2016 New Zealand [44]	To determine the benefits of remote CR	Systematic review and meta-analysis. CINAHL, Cochrane Library, Embase, MEDLINE, PubMed and PsycINFO searched to May 2015 for RCTs comparing remote and centre-based CR or usual care	11 trials involving 1189 patients with coronary heart disease. Outcomes- exercise capacity, modifiable cardiovascular risk factors and exercise adherence	Remote PT delivered by phone (n = 7) and websites (n = 6) with asynchronous PT review and feedback, education, psychosocial support +/or behaviour change. Interventions = exercise prescription +/or monitoring +/or adherence. Exercise prescription = 2–5×/week for 30–60 min at moderate (40–60%) to vigorous (70–85%) peak capacity usually while walking.	Activity, exercise adherence, blood pressure, and cholesterol levels showed greater improvement following remote cardiac rehabilitation than usual care. There were comparable effects on exercise capacity and other modifiable cardiovascular risk factors
Subedi et al. 2020 [45]. International	Effectiveness of implementing remote CR in practice, in factors underlying successful implementation	MEDLINE, Embase, PsycINFO, and Global Health were searched 1990–2018 for studies evaluating implementation of tele-rehabilitation to self-manage Chronic Heart Disease (CDH). A narrative synthesis was carried out.	10 studies (n = 2250). None studied implementation in practice. Outcomes were: acceptability (80%), fidelity (70%), appropriateness (90%), adoption (60%), feasibility (60%), and cost (40%), mostly from the participant perspective	Implementation of remote PT for the self-management of Chronic Heart Disease (CDH).	Remote CR was highly acceptable for most participants, but technical challenges (e.g., Internet connectivity) impacted acceptability and feasibility. Many found remote CR an appropriate alternative to usual care as it was convenient, flexible, and easy to access. Participants valued interactive intervention such as real-time exercise monitoring and feedback as well as individualised support.
Kikuchi et al. 2021 [46]. Japan	Evaluate the feasibility and safety of remote real-time monitoring of home-based CR.	Before/after study of exercise tolerance and lower limb strength. Feasibility = participation and completion rates. Safety = adverse events during the sessions.	10 elderly hospitalised or out-patients with heart failure. Mean age 76 ± 7 years; 60% male	12-week home-based CR remotely supervised using an integrated platform for tele-rehabilitation.	Remote supervised CR was feasible and safe among elderly HF patients. All completed the program. Median participation rate = 94.4% (IQR: 89–98%). No serious cardiovascular events were reported. 6MWD significantly improved from 383 ± 94 m to 432 ± 83 m (p = 0.003).
Laustsen et al. 2020 [47] Denmark	To investigate the effect of remote cardiac rehabilitation.	Cohort study assessed before, 6 and 12 months after remote PT. Outcomes: physical capacity (peak oxygen uptake, muscle endurance, power, and strength) and HRQoL	34 moderate risk patients (82% men; mean age 58 years; 74% had angina) with heart disease 2–6 weeks after hospital discharge	Remotely monitored exercise 3×/week for 12 weeks via a smartphone, a heart rate monitor and app. This displayed heart rate with an alarm indicating training intensity limits. Data were uploaded to a website.	Significant (p < 0.05) increases in physical capacity (peak oxygen uptake, muscle endurance, power, and strength) and health-related quality of life were seen both 6 and 12 months post-intervention except physical capacity at 12 months.
Rawstorn et al. 2018 [48] New Zealand	To evaluate users’ experience feasibility and acceptability of remote PT for people with coronary heart disease	Questionnaire assessment of usability and acceptability at 12-week assessment as part of a randomised controlled non-inferiority trial	67/82 adults with coronary heart disease randomised to remote CR and who completed usability and acceptability assessment at 12-week follow-up.	Intervention = 12 weeks of individualised exercise prescription, real-time physiological monitoring, coaching, and behavioural support, delivered via a bespoke platform	Usability and acceptability were positively evaluated by most (n = 44–66, 66–99%). 87% would choose remote PT as usual care, why because it provided convenient and flexible access to real-time individualised support. Technological challenges were rare and had little effect on user experiences or demand for remote PT.
O'Doherty et al. 2021 [49]. International	If exercise-based CR continued during the pandemic and how technology was used to deliver home-based CR.	Mixed methods survey of service provision, programme diversity, patient complexity, safety, technology use, and barriers.	330 CR professionals worldwide. 89.7% were from the United Kingdom	Exercise-based cardiac rehabilitation programmes	49% suspended programmes due to COVID-19. Of those that continued, 26% used technology beforehand. Most started using technology within 19 days of lockdown but 49% excluded high-risk patients. Telephone most commonly delivered CR but some centres used teleconferencing.

PT: physiotherapy or physiotherapist; CR: cardiac rehabilitation

Table 6. Selected studies of physiotherapy.

Author	Main study aim	Study design	Sample	Intervention	Findings
Miller et al. 2021 [50]. United States	To evaluate implementation of telehealth PT in response to COVID-19 and identify implementation strategies to maintain and scale up telehealth PT within a large urban academic medical centre.	Service evaluation using the Reach Effectiveness Adoption Implementation Maintenance (RE-AIM) framework. Patient data were extracted from electronic medical records	4548 physical therapy sessions provided by 40 therapists from 22 March 2020 to 16 May 2021, of which 3883 (85%) were telehealth.	Remote PT within a large urban academic medical centre in the United States.	Remote PT was feasible and acceptable. 94% of patients were satisfied and 92% would attend again. All PTs used remote methods at least once. Strategies to support implementation  = oganisational (policies, pre-existing partnerships), satisfaction surveys, clinical champions/leaders, education, and adaptability.
Albahrouh et al. 2021 [51] Kuwait	To investigate PTs perceptions of and willingness to use remote PT in Kuwait during the COVID-19 pandemic and to explore the barriers that may hinder the use of remote PT	Mixed methods: Cross-sectional survey and semi-structured interviews (n = 6). In the survey, an electronic questionnaire included: perceptions of, comfort with, willingness and barriers to using remote PT.	273 completed questionnaires from 747 sample (36.5%) of 747 Pts working in the governmental health sector. Interviews were conducted with PT managers to explore the barriers and facilitators.	Remote PT	87% found remote PT possible during the pandemic and would integrate it into their practice. More familiarity was associated with more comfort using it. Managers had concerns about patients’ privacy and data confidentiality.
Cox 2021 [52] Australia	To understand barriers to implementing remote PT and design a behaviour support strategy for clinicians to implement.	Hybrid implementation study: surveys, Theoretical Domains Framework to identify implementation barriers, and development of a targeted intervention.	Thirty-one clinicians completed baseline questionnaires identifying key barriers to the implementation of remote PT.	A 6-week brief intervention focused on remote clinician support, and education	Barriers were associated with knowledge, environment, social influences, and beliefs. The intervention was well received and changed practice but barriers remained.
Reynolds et al. 2021 [53]. Ireland	A cross-sectional observational study evaluating the Irish PT's experience of remote PT	Online survey exploring considerations and barriers, advantages and disadvantages, overall experience, and their opinion on the future of remote PT using a conceptual content analysis	205 members of the Irish Society of Chartered PTs. 75% female. Mean age 36–45 years with 17 years of PT experience.	Remote PT	83% had no experience with remote PT pre-COVID. 40% considered it a stop-gap. Considerations re: starting remote PT were service user’s suitability, COVID restrictions, adequate technical and organisational resources, and PT's professional conduct and skills. Benefits = reduced travel time (82%), flexibility (81%), and avoiding COVID risk (92%). Disadvantages = limited physical examination (86%).
Hall et al. 2021a [54] and Hall et al. 2021b [55]. United States	To describe the characteristics of paediatric remote PT and identify its important factors, facilitators, and barriers during COVID-19.	An anonymous, online survey including 3 open-ended questions and 1 multipart Likert Scale question from the questionnaire above. Analysed using thematic analysis and descriptive statistics.	250 paediatric PTs in the United States who transitioned to telehealth during COVID-19.	Remote paediatric PT	96% had no experience of remote PT before COVID. Only 15% had formal training. 35% moved most of their caseload to remote delivery. Most found it effective for carers (90%) and children (77%) and were confident providing it (73%). PTs in schools were less positive. 76% would provide remote PT in the future. 6 themes: carer engagement, technology, resilience and personal attributes, equity, and COVID specific issues. Carer engagement and access to stable technology were most important for effective remote PT.

Methodological quality was not formally assessed in this review, but it was noted that sample sizes were generally small and power calculations seldom used. Furthermore, randomisation processes and concealment of allocation were often poorly reported in randomised controlled trials. Control interventions were traditional in-person physiotherapy whether an out-patient department/clinic or home-based therapy.

The interventions studied were described in several different ways including tele-rehabilitation; tele-care; remote care, e-health, and m-health. For ease of comprehension and consistency, we have referred to the interventions collectively as remote physiotherapy. Unsurprisingly a wide range of technologies were used. The most common were simple technologies such has such as videoconferencing or mobile phones to deliver the same care as would be delivered in-person (n = 22, [18,27,31–37,40,41,43–46,50–55]) and these were also the focus of the systematic reviews [12,23–25]. The most common system developed specifically to deliver physiotherapy involved videos demonstrating patients’ exercise programmes uploaded to a web-site or app (n = 9, [13,15,17,21,26,30,38,44,47]). One system created an avatar to demonstrate the exercises [14]. These were generally asynchronous. Some also monitored adherence and performance of the exercise which was feedback to the patient and/or therapist. Direct contact with the therapist was usually via a “visit” using video conferencing when the exercise regimen was revised, and self-management advice or any other support given. A few of these interactive systems incorporated coaching, psychological support or behavioural change elements as well as exercise. In some systems, equipment such as body-worn movement sensors, heart rate monitors, or pulse oximetres were used to give synchronous feedback on performance to the patient and/or therapist. Further details of the interventions are found in Tables 2–6.

As befits the wide ranging aims of the selected studies, the concepts and outcomes investigated were very varied, as were the measurement tools used. The aims of the selected studies could be broadly categorised as investigating the feasibility and acceptability of remote physiotherapy and its effects. Each of these are considered in detail subsequently.

The feasibility and acceptability of remote physiotherapy

Feasibility and acceptability of remote physiotherapy were the most frequent objectives of the selected studies (n = 36), four of which were systematic reviews in which feasibility and acceptability were secondary outcomes [12,25,31,45] with the effects of remote physiotherapy as the primary outcome. Four randomised controlled trials assessed feasibility and/or acceptability [14,15,26,27]. Ten cohort studies [16–18,28,38–41,45,46], four implementation studies [27,48,50,52], seven surveys [20,42,47,51,53,–55], and four interview [19,22,30,31] studies investigated feasibility and/or acceptability, all as the primary objective. Three papers [20,22,42] investigated the views of potential users (i.e., patients and therapists without experience of it), while all other studies involved people with experience of remote physiotherapy.

Feasibility and acceptability are broad concepts which we categorised as patient satisfaction, the advantages/facilitators of remote physiotherapy, barriers to using remote physiotherapy, adherence, and safety.

Patient satisfaction. Sixteen studies assessed patient satisfaction [11–13,15,25–29,40,41,43,45,47,50,51]. Although a variety of quantitative and qualitative measures were used, all, including systematic reviews, reported that patient satisfaction was positive—described as high, very high, or excellent. One study observed that there was no difference in satisfaction between groups receiving remote or traditional physiotherapy [13].

The advantages or facilitators of remote physiotherapy was investigated in eight studies [15,19,21–24,30,31,42,45,47,50,53]. The biggest advantage highlighted was that remote physiotherapy increased access to physiotherapy for those who could not travel for in-person treatment, whether because of the distances involved or the degree of disability. Furthermore, patients found remote physiotherapy more flexible and convenient than traditional delivery, enabling them to access therapy when it was convenient and fitted around work or other commitments. This reduced their costs with less travel and absence from work [19,22,24,30,39]. The other main advantage of remote physiotherapy was that it supported patients to perform their exercises [15,21,45,47] through demonstrations of the exercises. For this reason, remote physiotherapy exercise platforms were preferred over paper handouts [21]. Exercise reminders and reinforcement messages were also useful, if individualised to fit in with the patient’s daily routine to promote engagement and motivation to exercise [15,21,31,38,39,45]. Other features that users found useful were feedback on performance and adherence for both patients and therapists and interaction with a “real” therapist [15,21,42]. In several studies, patients highlighted that the remote physiotherapy platform supplemented rather than replaced contact with a physiotherapist but enhanced the therapeutic relationship by creating an interactive and ongoing connection between them and their physiotherapist [15,21,31,45,47]. These views were broadly echoed by potential users of remote physiotherapy—patients and therapists in a traditional pulmonary rehabilitation services were asked about their interest in, and views of remote physiotherapy [19,42]. They both highlighted that the social aspect of rehabilitation was important and wanted a remote system to maintain this, for example, through group exercise and education sessions and opportunities to interact with each other and the physiotherapists. Studies of patients’ experience of using remote physiotherapy reported that patients reported enhanced self-efficacy, confidence and motivation to exercise when using remote physiotherapy, enabled them to monitor their condition and progress, using heart rate monitors and pulse oximetres, for example, and it also increased their confidence to self-manage their condition [26,38,39,44,45].

Adherence to remote physiotherapy regimen was addressed in eleven studies using a variety of measures [15,17,25–27,32,34,36–39]. Overall, adherence was mixed with some patients hardly exercising or using the remote system at all, and others exercising at or above the prescribed dose. Overall, studies which reported the proportion of patients meeting recommended doses varied from 35–95% (overall, 71%) [17,26,28] with 87–95% of patients completing the programme [17,18,32,46] except one study in which only 57% competed [13]. One study reported much lower uptake of remote exercise sessions and completion of an online symptom diary (56% and 43%, respectively) but this was over a 2-year period [38,39].

Safety of remote physiotherapy (in the form of adverse events) was monitored in seven studies of stroke and pulmonary rehabilitation [24–26,28,32,40,46] including three systematic reviews [24,25,32]. Although all reported that no adverse events occurred, few details about how adverse events were measured/defined or monitored were available.

Barriers (whether perceived or actual) to remote physiotherapy relating to the technology, organisation and the physiotherapists’ workload and skills were reported [18,23,27,30,42,47,50,51, 53–55]. Technical barriers included difficulties with setup; connectivity and the interface [18,27,30,51,53,55]. However, few technological problems were reported in the selected studies [18,27,47], and when specifically evaluated satisfaction with the technology was high [18,47]. Organisational barriers to using remote physiotherapy were lack of clarity about the medico-legal/liability situation; licensing and intellectual property ownership; and charging processes and other policies [23,50]. While physiotherapists highlighted the potential impact of using remote physiotherapy on their workload as a barrier if physiotherapists were expected to provide it in additional to in-person care [42]. Lack of familiarity with remote physiotherapy systems and difficulty undertaking patient assessments remotely were further issues [42,54].

The effects of remote physiotherapy

The effects or effectiveness of remote physiotherapy was the objective in 21 studies. Six were systematic reviews [12,24,25,33,34,45], seven were randomised controlled trials [13,14,26,34,36,37] eight were cohort studies [16,29,38–41,45,46]. The outcomes measured tended to reflect the priorities of the clinical group involved. For musculoskeletal conditions, the focus was on joint and muscle impairments; pain and function. For stroke, this was activities of daily living, balance, upper limb function, and health-related quality of life for caregivers and stroke survivors, while for cardiac and pulmonary conditions exercise capacity, physical activity and modifiable risks factors were measured. Regardless of condition, study design, or measures used, remote physiotherapy was found to be effective in that outcome improved, but to a comparable extent to traditional delivery whether measured immediately after the end of treatment or after longer-term follow-up.

Seven studies including five systematic review [13,14,16,19,23,24,55] assessed costs, or perceived costs (rather than cost-effectiveness) as a secondary outcome, and reported that remote physiotherapy was less costly than traditional care due to lower therapist input; less travelling and/or fewer hospital re-admissions.

Discussion

This scoping review has indicated that remote physiotherapy is feasible, acceptable and safe, with comparable effectiveness to traditional in-person physiotherapy. Patient satisfaction was high and adherence was mixed. Phase III randomised controlled trials with adequately powered samples are needed to more fully understand the clinical effectiveness of remote physiotherapy. It is noteworthy that the selected studies investigated remote physiotherapy delivered by physiotherapists working specifically to deliver it as part of the research study, rather than alongside all the other aspects of everyday physiotherapy practice. Furthermore, although the care was labelled as “remotely delivered,” only nine studies had no in-person contact at all [13,15,18,21,24–26,28,31]. The remainder involved some combination of in-person assessment, treatment demonstration/initiation or monitoring. These might be considered a “blended” rather than a purely remote approach to care delivery. Therefore, future trials need to use a cluster controlled design and process evaluations to investigate the effectiveness and implementation of remote physiotherapy within everyday practice and to consider the optimal combination(s) of in-person and remote contact for different conditions and types of treatment. The participants in the selected studies were patients and therapists who were sufficiently motivated and invested in remote physiotherapy to participate as there is some suggestion that uptake, satisfaction, and adherence may differ when delivered as part of everyday practice [57–59].

All studies that assessed costs, found remote physiotherapy less costly with less health resource utilisation than traditional physiotherapy. However, the cost-effectiveness has not been addressed which should also be included in future trials.

Patient satisfaction with remote physiotherapy was high. Patients valued the flexibility of remote physiotherapy which could be fitted around other commitments. None of the selected studies directly compared satisfaction with remote physiotherapy with traditional delivery but other studies (which were not selected in this review) have suggested that satisfaction is comparable or higher for remote delivery than for in-person care [58,59]. Future trials should include patients’ satisfaction and experience of traditional physiotherapy delivery as well as remote physiotherapy using mixed methods, to enable a richer, more detailed understanding of the advantages and disadvantages and how services can be optimised.

The biggest advantages of remote physiotherapy from both the patients’ and the therapists’ perspective were that it increased the scope and flexibility of access to therapy. However, the digitally excluded would not have been recruited for these studies. Digital exclusion is linked to social and economic deprivation [60,61] not only in that some patients would be unable to afford the technology or Internet connection but also Internet coverage is often poor in social deprived areas such as remote and inner city locations [62,63]. Furthermore, for physiotherapy, limited language skills, physical, and/or other disabilities may limit patients’ ability to use technologies for remote treatment [62–65]. It is therefore important that traditional delivery remains available for those for whom remote treatment cannot be accessed. Remote physiotherapy systems also need to be suitable for a range of technologies. For example, a mobile (or even landline) phone may be suitable for patients who cannot access or use a web-based system. Patients preferred remote physiotherapy to supplement, rather than replace in-person contact with a therapist [15,21,31,43,45,47]. Thus, a hybrid or blended system, involving both may be most effective and acceptable. Future research needs to consider the exclusion rates for remote physiotherapy in different patient populations and explore ways to overcome them so that physiotherapy services are suitable and acceptable for as many patients as possible.

The other advantage of remote physiotherapy was that patients found systems which demonstrated home exercise plans useful to remind them when and how to exercise, which in turn enhanced adherence and confidence. This is to be welcomed as adherence to physiotherapy home exercise programmes is notoriously patchy and resistant to change [66–68]. Although the mechanisms of increased adherence have not been addressed, patients noted that access to on-going physiotherapy remotely enhanced the patient–therapist relationship, which is important [68].

All the studies that assessed adverse events [24–26,28,32,40,46] reported that no adverse events occurred and concluded that remote physiotherapy is safe. However, it should be noted that few details were provided about how adverse events were defined or monitored. It appears the researchers relied on adverse event reporting processes in research protocols or “clinical incident” procedures used in practice, which focus on serious injurious incidents. Further research or clinical audit is needed to investigate the possible (non-critical/serious) side effects of remote physiotherapy in more detail and to examine the level of risk patients find acceptable.

It is reassuring that no adverse events were reported, as health care professionals have expressed concerns about patient safety if they are not physically present to assist a patient [17,25], of “missing something” if they do not have physical contact [17,26] and of risks to patient privacy and confidentiality [51,53]. Interestingly physiotherapists concerns about the safety of remote physiotherapy appeared to be less of a problem for physiotherapists with experience of working remotely [25,67,68]. These concerns are likely due to the need for therapists to adapt existing skills (such as using observation and patient questioning rather than “hands on” skills) to assess and treat remotely, and to develop new ones (such as mastering the technology). These needs can increase workload, change established routines, and challenge professionals’ confidence and identity [43,55], all of which may contribute to resistance to implementation. To effectively implement and sustain remote physiotherapy, there needs to be a recognition of the complex and often implicit, personal processes involved as well as sufficient resources (in terms of equipment and infrastructure), specific training and on-going support to problem-solve and build expertise [68,69].

The other main barriers to remote physiotherapy highlighted in this review and echoed elsewhere [67–69] focussed on organisational and administrative factors. However, these may be expected to be expedited by the need to rapidly introduce remote working practice in response to the COVID pandemic [69].

Limitations

This was a scoping review so methodological quality of the selected papers has not been formally assessed. Some of the findings may therefore be at risk of bias and therefore should be treated with caution. This is particularly true for the randomised controlled trials (and their systematic reviews) which were usually underpowered and the randomisation, concealment of allocation and blinding processes were not completed or poorly reported.

We also limited the search to the main databases and papers published since 2015 to ensure currency but this may have excluded relevant or important papers published beforehand. Furthermore, we excluded papers which were not specifically about remote physiotherapy so there may be relevant information from studies of tele-rehabilitation involving a multi-disciplinary team which is relevant to physiotherapy which we missed. However, the selected studies were extensive and findings were consistent so it is unlikely that a wider scope of the search would have changed the results, but merely added to their strength.

Conclusions

Remote physiotherapy is safe, feasible, and acceptable to patients with comparable effects to in-person care. Remote delivery increases access to physiotherapy especially for those who cannot travel to a treatment facility whether doe to distance or disability. Remote physiotherapy may increase adherence to exercise by reminding patients when and how to exercise. However, it does not suit everyone, thus a hybrid system with both in-person and remote delivery may be most effective.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Chartered Society of Physiotherapy Charitable Trust.