Exercise-based telerehabilitation for the management of chronic pain in people with severe haemophilia: a mixed-methods feasibility study

Abstract

Background

Chronic pain is reported by between 30 and 71% of people with haemophilia (PWH). Exercise is shown to be effective for pain management in other arthritides, but it remains unclear if such an approach is effective or acceptable to PWH. The aim of this study was to evaluate the feasibility and acceptability of a telerehabilitation exercise intervention for PWH living with chronic pain.

Methods

This was a multisite, non-randomised, pre-post feasibility design, with a nested qualitative study. People with severe haemophilia > 18 years, living with chronic pain, were recruited. The intervention comprised 12 low-impact/moderate intensity, individualised exercise sessions and 3 knowledge-sharing and discussion sessions. Primary objectives assessed according to predefined progression criteria were as follows: (a) recruitment rate (5 participants enrolled per site over 8 weeks), (b) adherence (≥ 75% participants would adhere to ≥ 75% of sessions), (c) follow-up rate (≥ 75% completion of self-reported measures), (d) fidelity (intervention delivered as described in protocol) and (e) safety (≤ 30% participants would report adverse events). Acceptability was evaluated from thematic analysis of post-intervention participant interviews. Preliminary evaluation of self-reported pain, function and quality of life (QoL) was a secondary objective. Results were reported using descriptive statistics integrated with qualitative findings.

Results

Ten PWH were recruited and completed the intervention. Nine agreed to be interviewed post intervention. Attendance at individual sessions was 84.5% compared to 52.1% for the group sessions. Outcome measures were successfully completed for 100% at baseline, 70% at intervention end and 60% at 3-month follow-up. No serious adverse events were recorded. Group median values in outcome measures (pain, function, QoL) showed minimal change post intervention. Participant interviews highlighted high levels of enjoyment, confidence in continuing exercises independently and positive views of virtual delivery and condition-specific exercise.

Conclusions

Recruitment rate and safety met the predefined progression criteria. Fidelity partially met the progression criteria, but the follow-up rate for self-reported measures did not. The study was acceptable to both participants and physiotherapists. Further intervention development is needed to review approaches to outcome measure collection and refine the usefulness of the knowledge-sharing sessions.

Trial registration

The study was prospectively registered on 9 July 2021: International Standard Randomised Controlled Trial Number ISRCTN 17454597.

Key messages regarding feasibility

• What uncertainties existed regarding the feasibility?

Chronic pain associated with haemophilia arthropathy is a significant clinical issue. Current haemophilia treatment guidelines recommend a primarily pharmacological approach to chronic pain management with the use and approach of physiotherapy being discrepant. Whilst general exercise is now accepted as safe for PWH, the acceptability and feasibility of an exercise-based approach in managing chronic pain are unknown. Furthermore, there is a dearth of knowledge on the applicability and usefulness of delivering such an intervention using virtual communication technology.

• What are the key feasibility findings?

This study confirmed that a telerehabilitation intervention was feasible and acceptable to PWH living with chronic pain. There were no serious adverse events reported. Whilst the virtual delivery was highly acceptable to participants, the physiotherapists reported an increased administrative burden in delivering the study. Participant-reported outcome measures did not fully capture change experienced by participants, but post-intervention interviews did. Inclusion of knowledges and discussion sessions did not appear to provide added value to the intervention.

• What are the implications of the feasibility findings for the design of the main study?

Virtual delivery was highly regarded by participants in terms of time and convenience, although both participants and physiotherapists felt that having an option of face to face as well as virtual may be beneficial after more than 6 weeks. Further work is needed to establish which outcome measures may be more acceptable to monitor change as well as be meaningful to those taking part. Given the burden on the physiotherapists, further evaluation of feasibility is required for delivery of this study in services with less than a full-time physiotherapist.

Background

Haemophilia is the umbrella term for the most common of the rare lifelong bleeding disorders — haemophilia A (deficiency of clotting factor protein VIII) and haemophilia B (deficiency of clotting factor protein IX). Both disorders occur due to a mutated or absent genetic code on the X chromosome, and therefore, it almost exclusively affects males [1, 2]. Disease severity is based against normal values for clotting factors of 50–150%: mild (levels between 5 and 40% of normal), moderate (between 1 and 4% of normal) and severe (levels of < 1% of normal) [3, 4]. In its untreated state, spontaneous musculoskeletal bleeding is a hallmark of the condition in almost all people with severe and some with moderate haemophilia. The mainstay of current treatment is to raise the factor levels in the blood or balance haemostasis enough so as to limit the possibility of spontaneous bleeding (prophylaxis) or treat if a bleed is suspected (on-demand) [3].

The phenomenon of intra-articular joint bleeding in haemophilia is proposed to initiate the process of synovial joint destruction in three interrelated stages of acute haemarthrosis, synovitis and degenerative joint arthritis [5]. Haemophilic arthropathy (HA) is the term given to this process and is characterised by chronic synovitis, cartilage destruction, epiphyseal enlargement and bony deformity [6] and has been shown to have predominantly degenerative, rather than inflammatory, characteristics [7]. A recent UK study of data from the National Haemophilia Database highlighted that those younger PWH (< 19 years old) had little or no joint damage due to having treatment since infancy, whereas those over 40 years old had significantly higher levels of joint disease [8]. Increasing severity of joint damage alongside the increased number of joints affected in older adults with severe haemophilia has also been shown to be strongly correlated to poor perception of function and moderately correlated with pain [9, 10].

Current data indicates that the experience of pain is an unavoidable reality for many PWH, with figures suggesting between 49 and 61% of PWH experience pain on a daily basis [11, 12]. Episodic acute pain is reported in 20–68% of adults [12, 13] with chronic pain experienced by 30–71% of adults [12, 14] and in 19% of children [15]. Living with haemophilia and chronic pain brings with it constraints in mobility and independence, increased anxiety, poor quality of life and frustration due to restrictions in activities of daily living [16, 17].

Figures vary from 21 to 50% of PWH reporting that they did not believe their pain was well managed [12, 18] and may reflect a lack of standardised management pathways. Interestingly, physiotherapy as an option in pain management is discrepant, reportedly used by between 12 and 46% of people [19, 20]. However, what that physiotherapy may entail and in what context (acute or chronic pain) is poorly described, as is the effectiveness of such physiotherapy intervention. Whilst effectiveness of rehabilitation for primary management of pain in osteoarthritis and rheumatoid arthritis is well established [21,22,23], there has been no structured scientific research to evaluate its effectiveness for management of chronic pain in PWH. Whilst exercise in general has been shown to be safe for PWH [24], its use as an option in pain management remains unknown.

Physiotherapy for PWH has traditionally been delivered in person in haemophilia centres. However, even prior to the Covid-19 pandemic, the use of a telemedicine approach was being highlighted as an opportunity for specialist haemophilia care to be delivered locally to those living large distances from their specialist centres [25]. Telemedicine approaches for multiple aspects of haemophilia healthcare delivery have been shown to be acceptable to PWH throughout the Covid pandemic relating mostly to the delivery of routine care, e.g. review appointments over the telephone or on webcam [26]. In relation to physiotherapy specifically, there has been tentative feasibility demonstrated in the virtual delivery of general exercise classes for PWH [27, 28], as well as co-developed hybrid interventions (mix of face to face and virtual) for those living with haemophilic arthropathy [29] and to increase confidence in being physically active [30]. Feldberg and colleagues evaluated the use of an asynchronous exercise and pain education intervention (videos) for chronic pain in PWH, reporting positive improvements in pain intensity and function [31]. However, to date, no intervention has been developed or evaluated that uses a real-time (synchronous) telerehabilitation approach specifically developed for use in the management of chronic joint pain in PWH.

The UK Medical Research Council (MRC) framework defines the importance of feasibility testing in the development of new complex interventions [32]. In considering the level of complexity presented by PWH living with painful haemophilic arthropathy, the design and effective components of a rehabilitation intervention and its potential use for pain management in PWH have not previously been evaluated. Furthermore, the potential feasibility of a telerehabilitation approach for delivering an exercise-based intervention for PWH with chronic pain remains unknown.

The overall aim of this study was to evaluate the feasibility and acceptability of a physiotherapy-led, low-impact, moderate intensity telerehabilitation intervention in PWH who have chronic joint pain related to haemarthropathy, termed the REMAP-Haemophilia study (REhabilitation for the Management of Arthritic Pain in haemophilia). Evaluation of objectives were carried out using quantitative and qualitative approaches.

The primary objectives identified for the study were as follows:

• 1. Determine the safety of an exercise-based telerehabilitation intervention for PWH.

• 2. Evaluate the feasibility and acceptability of the intervention delivery and content.

• 3. Determine the acceptability of the overall intervention (recruitment rate, adherence to the intervention, attrition and study completion rate).

• 4. Determine the acceptability of chosen outcome measures.

The secondary objective identified for the study was as follows:

• 1. Collection of preliminary efficacy data (before and after) with patient-reported outcome measures (PROMs) evaluating pain, quality of life and function.

Methods

Study design

This was a multisite, non-randomised, pre-post feasibility study with an explanatory-sequential nested qualitative study.

The study was given ethical approval by the East Midland-Nottingham 2 Research Ethics Committee (rec. reference number: 21/EMI/0161). The study was sponsored by the Royal Free London NHS Foundation Trust (reference number: 141604) and was prospectively registered (ISRCTN 17454597).

Participants

Participants were identified by the physiotherapist in advance of attendance at routine haemophilia clinic reviews. Following eligibility screening, participants were given a study information sheet. Written consent was obtained prior to completion of baseline assessments. Baseline demographic data was collected from the medical notes.

The study inclusion criteria were as follows:

• People with severe haemophilia A or B (with or without an inhibitor)

• Aged 18 years and over

• Self-reported symptoms of chronic pain associated with haemophilic arthropathy in any joint

• Willing and able to give informed consent for participation in this study

• Able to follow instructions

• Have a good command of written and spoken English

• Registered at a UK-located haemophilia comprehensive care centre with a named physiotherapist

• Have access to a laptop/tablet with webcam at home and sufficient Internet connection

The study exclusion criteria were as follows:

• People with mild or moderate haemophilia A or B

• Any other inherited bleeding disorder

• A diagnosis of chronic pain that is not associated with HA

• Severe and/or unstable cardiovascular disease

• Severe and/or unstable pulmonary disease

Intervention

In keeping with the MRC guidelines for the development of complex interventions, REMAP-Haemophilia used stakeholder participation to develop the theory underpinning the intervention. Stakeholders included people with haemophilia, specialist haemophilia physiotherapists and clinical academics with experience in intervention development [33]. Prior qualitative studies also informed aspects of the intervention and outcome measure choice [34, 35]. The theory development process also informed the identification of behavioural change techniques (BCT’s) to include in the overall design and delivery of the intervention (the full list of BCTs is provided in Supplementary File 1).

The overall design and delivery of the study are described in Table 1 according to the Template for Intervention Description and Replication (TIDier) and Consensus in Exercise Reporting Template (CERT) checklists [36, 37].

Table 1 Exercise intervention summarised as per the CERT and TIDier checklists

The REMAP-Haemophilia study was a 12-session (6 week), low-impact, moderate intensity exercise-based intervention delivered virtually using the Microsoft Teams digital platform. One individual exercise session and one group exercise session were planned each week, with appointment times agreed between the physiotherapist and participants. An initial face-to-face assessment provided each participant with their starting point for each exercise. Exercises targeted upper and lower limbs, as well as comfortably challenging cardiovascular effort. The lower impact approach aimed to limit mechanical stress on those with haemophilic arthropathy of the ankle.

A physiotherapist led every session, gave the instructions for each activity, monitored effort and participation ability, provided feedback and encouragement and kept time. Each exercise was repeated three times per set and timed at 30 s of moderate exertion (as per the rated perceived exertion score card that each participant had at home), 30-s rest and a 2-min break in between each set. Exercises included resistance (body weight or additional) and cardiovascular with an additional exercise being added to the overall session plan every 2 weeks. The total time needed per exercise session was designed not to exceed a total session time of 40 min by the end of week 6. Participants were not restricted from participating in their normal routines, nor were they prevented from commencing new physical activities whilst taking part in this study (detail of the exercises can be found in the Supplementary File 2).

Three ‘knowledge-sharing and discussion’ sessions were delivered by the physiotherapist before the group exercise session at weeks 1, 3 and 5. Delivered over MS Teams, the sessions focussed on the following: (1) why we experience pain and what is means, (2) physical activity (benefits and struggles) and (3) pacing and finding your own level. After a short presentation, the aim was to encourage a forum for participants to discuss shared experiences of pain and activity and any actions or activities they had found to be helpful for them (detail of the sessions can be found Supplementary File 3).

Physiotherapists received training in advance of the study commencement which included delivery of protocol, study delivery/management requirements and delivery of the BCTs.

Participant-reported outcome measures

The PROMs were collected at three time points — pre intervention (T0), on intervention completion (T1) and at 12-week post-intervention completion (T2). The pre-intervention measures were collected in person at the initial face-to-face session, with the remaining two time point outcome measures being posted to participants with a prepaid, addressed envelope.

Measures of pain

Brief Pain Inventory (BPI-SF)

This 9-item self-administered questionnaire evaluates the severity of a person’s pain and its impact on their daily functioning and is widely used in a range of non-malignant pain conditions [38]. Test–retest reliability construct validity is good when used in PWH [39, 40], but responsiveness is as yet unknown.

Pain Self-Efficacy Questionnaire (PSEQ)

This is a 10-item questionnaire that assesses the confidence of people (with any type of chronic pain) in activity despite pain. Each item’s response is on a 7-point scale and is scored 0–6. It is an additive score between 0 and 60, whereby a higher score indicates higher self-efficacy beliefs [41]. Validity, reliability and responsiveness in people with musculoskeletal disorders are excellent [42], but its use has not previously reported in PWH.

Measures of quality of life

EQ5D-5L

This is a 5-item questionnaire evaluating generic health-related quality of life over five dimensions (mobility, self-care, usual activities, pain/discomfort and anxiety/depression). It is reported as an overall utility score, alongside an overall health report with a visual analogue scale [43]. It has been shown to have satisfactory construct validity in PWH [39, 44].

Musculoskeletal Health Questionnaire (MSK-HQ)

This 14-item questionnaire allows people with MSK conditions to report their symptoms with questions relating to pain/stiffness in the day and night, problems with activities of daily living, sleep disturbance, emotional wellbeing and confidence in managing symptoms. It is scored additively from 0 to 56, whereby a higher number indicates better musculoskeletal status. Whilst its use has not previously reported in PWH, it has been shown to be responsive across a range of musculoskeletal conditions [45].

Measures of function

Haemophilia Activities List (HAL) Questionnaire

This measures the impact of haemophilia on self-perceived functional abilities in adults with haemophilia. It has 42 multiple-choice questions across seven domains. The total score ranges between 0 and 100, where a higher score indicates less perceived functional impairment. It has been shown to have good internal consistency and convergent validity in PWH [46].

Patient-specific functional score (PSFS)

This is a self-reported measure that aims to assess functional change in people presenting with predominantly musculoskeletal disorders. Participants identify up to five activities that they are having difficulty with as a result of their problem, rating the current level of difficulty associated with each activity on an 11-point scale (0–10). They then rescore at the end of the intervention.

Measuring overall change

Patient global impression of change (PGIC)

This is a single question completed at the end of the intervention that measures a change in an individual’s clinical status. People rate the change in their own clinical status on a 7-point scale, from very much improved to very much worse.

Diaries

Participants completed a short weekly diary reflecting on their experience that week and any change they noticed in themselves. They were asked to send their completed diaries back in the same envelope as the outcome measures at T1.

The physiotherapists delivering the study also completed a weekly diary to record their thoughts and reflections on practicalities of delivery of the study, feedback or comments they had received from participants, technical issues and any changes they made to how they delivered the study.

Qualitative evaluation

All those completing the exercise session component of the study were contacted by email or telephone to confirm if they still wished to be interviewed, with an interview arranged at a time convenient for them.

A topic guide developed with a PWH was used for the post-intervention interviews. Semi-structured interviews were conducted over Microsoft Teams or on the telephone. Questions were open ended and aimed to gain an insight into each person’s experience of taking part in the exercise intervention, as well as drawing focus to the objectives relating to the feasibility and acceptability of the intervention.

Each participating physiotherapist was interviewed over MS Teams on completion of the intervention with a focus on feasibility, acceptability and fidelity of the delivery of the study, as well as general feedback and overall views of having taken part.

All interviews were recorded, and the audio files transcribed verbatim by a professional transcription service.

Feasibility outcomes and progression criteria

Safety of the intervention was evaluated by the number of reported adverse events/serious adverse events recorded at each site. Perception of safety was also evaluated in the participant post-intervention interviews.

Progression criteria for feasibility outcomes were identified a priori and are detailed in Table 2. Failure of an outcome to meet progression criteria would be evaluated against the need for modifications for inclusion in a future RCT, enhanced monitoring of that domain within an RCT or to decide that a full RCT would not be feasible at this stage.

Table 2 Thresholds for evaluation of study feasibility

Sample size

As this is was a feasibility study, no power calculation was required. By virtue of its status as a rare genetic disorder and acknowledging the multi-faceted impacts on daily life living with haemophilia, the potential for research waste in developing and testing novel interventions needs to be avoided. Given the rarity of this condition, coupled with the current contextual difficulties in physiotherapy access for PWH in the UK, the research team decided that 10 participants across 2 sites would be sufficient to allow preliminary evaluation of the feasibility of delivery, data collection and acceptability of this study.

Data analysis

Participant demographics and characteristics were tabulated, and a CONSORT diagram described the flow of participants through the study.

Descriptive statistics were used to assess the feasibility objectives using Excel. Due to the low numbers of participants and in keeping with the feasibility design, continuous variables (outcome measures) were summarised using median, interquartile range and range for group changes between time points.

NVivo (Release 1.6.1 version) was used to manage the qualitative dataset (transcripts and diaries). Acceptability of the intervention was evaluated from analysis of the participant and physiotherapy interviews and diaries.

The interview data and the diary entries from the participants and the physiotherapists were first analysed together using a reflexive thematic analysis approach. This is a six-phase recursive approach comprising the following: (1) familiarisation with the data; (2) coding; (3) generating initial themes; (4) reviewing and developing themes; (5) refining, defining and naming themes; and (6) writing up [47, 48].

A second stage to the data analysis reviewed the completed thematic analysis alongside the initial coding structure developed within. This enabled an analysis of the domains relating to the feasibility and acceptability objectives, helping inform the integration of the qualitative findings with the quantitative data. Quantitative and qualitative results were then tabulated and presented as a joint display.

Results

Study recruitment

Recruitment took place between November 2021 and February 2022 in two large regional haemophilia centres. Twenty-four people were screened for eligibility, 13 were eligible and 10 agreed to take part (consent rate of 77%). Recruitment and retention details are outline in Fig. 1.

Fig. 1

Participant recruitment and retention in the REMAP-Haemophilia study

Participant characteristics

An overview of participant details is presented in Table 3. Ten male participants aged between 39 and 67 (median age 57) were recruited to the study. Six participants described themselves as independently mobile, three used mobility aids intermittently (cane/crutches) and one used a mobility scooter for longer distances outside. All participants had chronic pain as defined by the International Association for the Study of Pain (IASP) [49] that was present for more than 3 months, and use of pain medication varied.

Table 3 Participant demographics

Primary outcome: feasibility

Feasibility threshold results are presented in Table 4. Target recruitment over an 8-week period was five per site. One site was over-recruited by one participant (n = 6) and the other site under-recruited by one participant (n = 4). Adherence rate for the intervention overall was 68.3%. When analysed per session type, adherence rate for the 1:1 session was 84.5% and for the group sessions was 52.1%. There were between-site differences in the attendance rates for the group sessions: Site 01 had 91.7% and Site 02 12.5%.

Table 4 Results of feasibility thresholds

The reasons given for missing individual 1:1 sessions (n = 8) included sickness (n = 2), recovery from an intra-articular ankle joint injection (n = 1), muscle injury unrelated to the study (n = 1), joint pain (n = 1) and knee haemarthrosis unrelated to the study (n = 1) and nonattendances with no reason given (n = 2). The reasons stated for nonattendance at the group sessions (n = 25) were anxiety (n = 6), other commitments (n = 3), sickness (n = 1), flank pain (n = 1) and no reason given (n = 14).

All 10 participants (100%) completed baseline PROMs (T0), and 7 (70%) completed at T1 and 6 (60%) at T2. Nine participants agreed to be interviewed at the end of study. One person declined to be interviewed due to anxiety. There were no missing data points identified on any of the outcome measures returned at T1. There were two missing outcome measures for one participant at T2.

Fidelity of the intervention delivery was 84.7%. Sixty-one of the 72 planned individual and group sessions were delivered virtually (as per protocol description) using webcams 80.4% of the time, with the remaining 12 (19.6%) being conducted over the telephone.

Adverse events were recorded. Overall, three participants (30%) reported an adverse event related to the study. Four episodes of increased joint pain after the exercises were reported by three people (one knee, one shoulder, two elbow). One of these participants also reported one episode of hamstring pain the day after the exercise session. Another reported a muscle sprain of his left flank unrelated to the study. One knee joint bleed was reported but was found to be unrelated to the study participation. No serious adverse events were recorded for anyone participating in the study.

The burden of participating in the study was acceptable to participants; however, the burden of study-related administration was highlighted by the physiotherapists. Whilst the organisation and delivery using Microsoft Teams was viewed positively, the time required in the working week to deliver the study as described was deemed significant. The physiotherapists noted that it was just about manageable to host and deliver the sessions, but there were concerns about having time for other tasks such as note writing. Strategies to mitigate against some of these issues included trying to devote a half or whole day to all the appointments or trying to spread them out evenly through the week. The therapists highlighted that 5–6 people would probably be the maximum number of people to include in the study in its current form.

Secondary outcome: efficacy

Group changes in measures of pain, quality of life and function are presented in Table 5 as group median change and interquartile range.

Table 5 Median change in pain, function, and quality of life before and after intervention

Integrated display of quantitative and qualitative findings

Quantitative and qualitative data for feasibility (Table 6), acceptability (Table 7) and efficacy (Table 8) were collated in a side-by-side format. The level of consensus between the datasets was evaluated as follows:

• Confirmation — The findings both agree.

• Expansion — The data diverges and expands insights or describes complementary aspects of the topic at hand.

• Discrepancy — The data appear to contradict each other or are inconsistent.

Table 6 Integration of feasibility findings

Table 7 Integration of acceptability findings

Table 8 Integration of efficacy findings

Discussion

This study demonstrates the acceptability, safety and convenience of delivering an exercise-based telerehabilitation intervention for PWH with chronic pain. The protocol was feasible with respect to consent and recruitment rate, adherence to the individual exercise sessions but did not meet the progression criteria for fidelity of delivery and follow-up rates of PROM completion. Both sites recruited successfully to the study. As haemophilia service specifications require people with severe/moderate haemophilia receive biannual clinical reviews [50], it is realistic to assume that recruitment onto definitive trials, and ultimately to the intervention, is achievable. Whilst the chosen PROMs provide little quantitative evidence of change in pain, function or quality of life, the participant interviews did highlight improvements that the participants experienced.

Overall, both study participants and physiotherapists found the virtual delivery acceptable and convenient. However, the physiotherapists reported an increased burden associated with the time needed to deliver the telerehabilitation sessions. Virtual delivery of telerehabilitation has been used in a range conditions such as low back pain, post-operative orthopaedic follow-up and multiple sclerosis, where it has been shown to be comparable to in-person appointments and better than no treatment/intervention at all [51]. However, there remains limited research on the use of telerehabilitation approaches in haemophilia. A study investigating a blended approach (face-to-face physiotherapy and a smart phone application) to rehabilitation for haemophilic arthropathy found this novel approach to be feasible and showed positive effect on lower limb function [29]. Another qualitative study investigated participant experience of a haemophilia-specific exercise class delivered in real time using a smart phone application. They reported similar positive outcomes to this study in respect to convenience and access to clinicians with specialist haemophilia knowledge [28]. Together with the findings of this study, they provide important reflections when considering the use of such technology in day-to-day haemophilia care, particularly when access to specialist physiotherapy remains an issue for up to 60% of PWH [52]. Future studies should include methods to ensure equity of access to digitally delivered healthcare such as telerehabilitation, as well as health economic evaluation to determine cost-effectiveness and how this may be best used to widen access to specialist physiotherapy.

The low-impact, moderate intensity progressive exercise regimen was designed to accommodate participants with multi-joint arthropathy. A Cochrane review investigating exercise interventions and patient beliefs for people with hip and knee OA found interventions are most effective if they are tailored to an individual’s preferences, abilities and needs [53]. Whilst it was acceptable overall, some participants said they would have liked a more tailored, joint-specific programme. Future iterations of this study approach will need to consider if including more options for exercise activity based on individual ability is warranted.

The knowledge-sharing and discussion sessions appeared to have limited acceptability. Only one study site managed to conduct the session as described in the protocol, with the other study site being unable to deliver the sessions due to lack of participation with the group sessions. Other studies in haemophilia have included condition-specific education sessions alongside physiotherapy interventions such as manual therapy and exercise, although none has been evaluated for effectiveness within those studies [54,55,56]. Cochrane reviews evaluating patient education in both RA and OA have shown only small short-term effects for disability associated with RA [57] and no improvements in self-management skills, function or quality of life in OA [58]. This aspect of the REMAP-Haemophilia protocol requires further evaluation and refinement if it is to be included in future studies, in particular how and when PWH want to receive information relevant to their condition.

Evaluating the clinical efficacy of REMAP-Haemophilia was not the primary purpose of this study, but an exploratory analysis on clinical outcomes was included. Although some participants reported some meaningful improvements, these were small, and overall, there were no changes in pain, quality of life and function. Authors have highlighted the need for outcome measures that go beyond just annualised bleed rate and better reflect the improvements in medical care for PWH [59]. A recent publication presented the outcome of a consensus approach to the development a core set of measures to be used in both research and clinical settings in haemophilia. They included number and location of bleeds, health-related quality of life, treatment adherence and joint health [60]. It is clear, however, that a focus on outcomes of disease/condition modification rather than symptom management limits the usefulness of the core set for pain research in haemophilia. Whilst the participants accepted the need to collect measures, they were less accepting the applicability of the PROMs. This may be because the assessments did not fully encompass their individual view of their haemophilia and pain experience, reflected in the interviews findings (Table 6) describing ‘generic’ surveys and feeling ‘squeezed into answering in a particular way’. More work is required to establish an acceptable method of measuring impact of rehabilitation interventions in PWH and perhaps giving more consideration/weight to the overall qualitative experience.

A strength of this study was the application of mixed methods in the data collection and analysis, the inclusive approach to recruitment and the pragmatic protocol design to encourage and facilitate participation in the study activities. A mixed-methods approach enables consideration of multiple viewpoints and positions to achieve a deeper understanding of the study findings [61]. Understanding the experience of all involved in the delivery and participation of the study, as well more practical issues concerning burden of delivery and administration, means this study adds to the current quantitatively heavy evidence base of physiotherapy rehabilitation in haemophilia. A follow-up review and evaluation of the programme theory that underpins this study [33] are planned to identify changes and further refine the protocol for a more definitive future trial.

The inclusion criteria for this study were purposefully broad, reflecting real-world experience and acknowledging the highly complex nature of PWH living with multiple joint arthropathy and chronic pain. When working with people with rare disorders such as haemophilia, it is important that study design does not further marginalise those who may have most to gain from taking part. This is especially important for PWH as it remains unclear if established rehabilitation programmes addressing predominantly single joint issues such as ESCAPE-pain [62] and the GLA:D OA knee [63] would be suitable.

Another strength is the inclusion of two sites in this feasibility study. This meant the study was able to include two different groups of PWH under the care of different specialist physiotherapists. This was an important consideration for feasibility. If there were difficulties delivering this study at a local level within well-staffed haemophilia centres, then it is highly likely that it would not be at all feasible in centres with less than full-time physiotherapy input.

The main limitation of the study is the small number of participants, so no statistical inferences can be made. The study’s feasibility design aimed to see if the intervention could be delivered and if it was safe and acceptable [64]. Poor completion rates of the post-intervention PROMs at each time point impaired preliminary evaluation of efficacy of the intervention. Further work is required to ascertain what PROM’s participants consider valuable to them (e.g. generic or condition specific), when and how to collect them and the most effective way of collecting data such as digital/electronic forms.

Another limitation is that whilst almost all participants had more than three joints affected by haemophilic arthropathy, the whole-body approach of the exercise programme was not acceptable to all participants, with some wanting a more joint-specific exercise approach. Future studies will need to balance the practicalities of a high degree of individualisation, alongside evaluating feasibility and applicability of an exercise programme in a population-based representative cohort such as those here.

Conclusion

The present study demonstrated an exercise-based, telerehabilitation intervention for people with haemophilia is feasible, safe and acceptable. Further work is needed to evaluate the choice of objective outcomes used and how they are collected, as well as the value of including a more subjective, person-centric experience of taking part in studies such as these.