Remote Physiotherapy as Effective as Face-to-face Consultations?

Introduction

In Covid-times, many healthcare professionals were forced to deliver remote consultations, including physiotherapy. Therefore, the effectiveness and applicability of doing so have been frequently examined since then. Evidence is available, particularly for osteoarthritis and back pain, but is less abundant for musculoskeletal injuries. Therefore, the current study ought to examine delivering remote physiotherapy or face-to-face usual physiotherapy and whether the first was as good or potentially better than the latter.

Methods

This randomized controlled trial was conducted across five hospitals in Australia. Patients were eligible to participate when they were on an outpatient waiting list and had any musculoskeletal condition or injury. The aim was to compare usual face-to-face physiotherapy to delivering remote physiotherapy to determine whether remote care can be as good or better than usual care.

People in the usual physiotherapy care group were seen in an outpatient department. The treating physiotherapist was able to select exercises for a home exercise program if deemed necessary. The content, number, and duration of the physiotherapy sessions were individualized and determined by the treating physiotherapist.

All participants included in the remote physiotherapy group were invited to an initial face-to-face appointment. During this session, the participant’s key problems and goals were determined and this formed the basis for delivering the remote physiotherapy program. The treating physiotherapist was free to select exercises for a home exercise program. This was adapted to the materials the participants had available at home. The exercises were selected from a large database (physiotherapyexercises.com) and provided through an application or printed on paper. Guidance was provided on the number of repetitions of each exercise to perform. Remote follow-up consultations were foreseen to further guide and instruct the participants to progress exercises when deemed appropriate. Some participants were already taught from the beginning how to progress with exercises.

The remote physiotherapy group received a telephone call at 2 and 4 weeks into the program. This was to revise the exercises and give instructions on how to progress the exercises. These calls were typically 5-10 minutes in duration. Automated text messages were sent every week to encourage participants to continue the program and to increase adherence.

At the end of the 6-week program, both groups were told that they were free to pursue whatever treatment they wished for their musculoskeletal injury, but it was not encouraged or facilitated.

The primary outcome was the Patient-Specific Functioning Scale obtained at the end of the intervention period of 6 weeks. A long-term follow-up was scheduled at 26 weeks, but this was a secondary outcome measure. All measurements were obtained by a blinded assessor at baseline and over a telephone call at 6 and 26 weeks. The Patient-Specific Functioning Scale is described on our website. The participants were able to select up to five functional activities that were important to them and for which they experienced difficulties in execution. Every activity is rated from 0 (inability to perform the activity) to 10 (ability to perform the activity at pre-injury level).

Results

A total of 210 participants were included in the study and 104 participants were randomly allocated to the remote physiotherapy program. The remaining 106 received usual in-person physiotherapy. Sixty-five percent of the participants were women and the median age was 53 years. More than 80% of the participants suffered from their injury for more than 12 weeks.

The injury locations included most were the knee, shoulder, and back, each accounting for roughly 20-30% of all locations. Diagnoses that came forward most often were

Back or leg pain (post-surgery or not),
Osteoarthritis,
Fractures,
Frozen shoulder, capsulitis, impingements or muscle tears
ACL or meniscus injury (with or without repair), patellofemoral pain or knee pain, hamstring injuries
Neck pain
Ankle pain, tendonitis or sprain

The between-group difference for the Patient-Specific Functioning Scale at 6 weeks was 2.7 points (95% CI -3.5 to 8.8). This means that no statistically significant difference was observed between both groups and that delivering remote physiotherapy may be as good as usual face-to-face physiotherapy care.

Questions and thoughts

The absence of a between-group difference led the authors to conclude that delivering remote physiotherapy is as good as delivering traditional face-to-face physiotherapy. But is that really the case? I wouldn’t say so as the average scores on the Patient-Specific Functioning Scale were only about 50% for both groups, both in the 6th and 26th week. Would you call it a success when your patient has only improved to half of their objectives?

At 6 weeks, you could still call it a relevant improvement, but the fact that the scores stabilized hereafter isn’t to be called improvements in my opinion. An important side note to this is explained by the fact that at 6 weeks the remote physiotherapy and the usual care were ceased.

Participants were not directly encouraged or facilitated to pursue physiotherapy, but they were free to do so, in whatever delivery method. Although they were free to do whatever they felt necessary for their injury, it might be possible that some stopped doing their exercises, while others possibly continued. Some people receiving remote physiotherapy might have proceeded to face-to-face physiotherapy, others might have switched to remote physiotherapy, et cetera. This gives a lot of heterogeneity in the outcome at 26 weeks (although the primary outcome was set at 6 weeks).
Also important to consider is that, unfortunately, it was not recorded what the participants did after completing the trial. The purpose was to see the difference between delivering remote physiotherapy and usual care for 6 weeks, and no difference was observed indicating that both options are effective. Yet, when looking at Table 4 it stands out that in both groups, the participants achieved only 50% of the maximum score for the Patient-Specific Functioning Scale. This means that they had not fully recovered.
It is reasonable to think they would continue receiving physiotherapy care. However, since this was not assessed, it can not be postulated that both interventions are effective in treating these musculoskeletal conditions. We can only conclude that there are no differences in outcomes when delivering remote physiotherapy or usual face-to-face physiotherapy.
Of course, given the wide nature of possible health conditions that were included, the 6 and 26-week endpoints may have contributed to this, since we know that for example an ACL tear or a frozen shoulder takes up much more time to recover from than an ankle sprain. The wide inclusion criteria (any musculoskeletal injury/condition) may have led to a very heterogeneous population, which may in turn confound the modest improvements.

Consider the context in which the current study was conducted to interpret the applicability of the results to your practice. People living in Australia may be confronted with long travel times to a healthcare facility. Some patients live in rural areas, while others may have better access to care when residing in the cities. Long waiting lists for receiving care have been reported. This may be due to several reasons, including the increased prevalence of health problems and increased awareness of the benefits that physiotherapy may exert, but can also come forth by the free consultations offered in government-funded hospitals. On the downside of these funded physiotherapy consultations are the long waiting lists. We know that many acute problems have a favorable course and are self-limiting, however, in the presence of negative prognostic factors such as fear avoidance, anxiety, or poor coping strategies for example, some of these acute problems can develop into chronic ones. The necessity of delivering remote physiotherapy may be high to prevent this long waiting period.

What can we take from this study, even if we can not directly compare care-delivery methods across different countries with each unique healthcare system? At least you can empower your patient to be as independent as possible. Your 30-minute consultation should guide them to participate in many more 30-minute (or longer) home exercise sessions. When you can find a way that works for your patient to increase their participation and self-efficacy, you have the key to avoiding making them reliant on you and still making them progress a lot. Remote follow-up consultations might be a useful adjunct. Possibly you don’t need half an hour, and thereby you might liberate some space in your calendar for others who need to be seen in real life. Maybe our blog can guide you through.

Are we ready to deliver remote physiotherapy yet? The poor adherence seen in this study where only 6 weeks of treatment were examined, may indicate that it is possibly too soon. But here again, this may be highly population-dependent.

Talk nerdy to me

This study chose a pragmatic approach to delivering remote physiotherapy as there were no standardized exercises or scheduled appointments. This was chosen because the authors aimed to mimic clinical practice as closely as possible. I tend to relate more to these kinds of studies to inform my practice since they translate more easily. On the other hand, I am also eager to learn more about how and when, so I was slightly disappointed to see that no exercise descriptions were given or summarized.

A limitation of this study might be the risk of inducing acquiescence bias. Acquiescence bias is a type of bias in which the respondents tend to say what they think the examiner prefers to hear since it is human nature to be accommodating. This is possible because the outcome assessor, although blinded, was sitting next to the participant to deliver help filling out the questionnaires on a computer. On the other hand, this way they avoided having non-response bias, where survey participants refuse or are unable to reply to a single survey item or the entire survey.

Participants were not blinded, but they were kept naive to the real purpose of the study. They weren’t told what the “gold standard” treatment was, but in stead they were only told that 2 different ways of delivering physiotherapy were compared.

Take-home messages

Delivering remote physiotherapy or participating in usual in-person care did not result in differences in patient-specific functional outcomes across various musculoskeletal conditions. Remote care is possible and feasible for physiotherapy treatment. As the improvements in patient-specific functioning are still quite modest, it is reasonable to further examine the usefulness of delivering remote physiotherapy in a more specific sample as opposed to the broad definition of musculoskeletal conditions as included in the current study.