“Pacing does help you get your life back”: The acceptability of a newly developed activity pacing framework for chronic pain/fatigue

Objectives: We have developed and feasibility tested an activity pacing framework for clinicians to standardise their recommendations of activity pacing to patients with chronic pain/fatigue. This study aimed to explore the acceptability and fidelity to this framework in preparation for a future trial of activity pacing. Design: Acceptability and fidelity were explored using semi ‐ structured interviews. Data were analysed using framework analysis. Participants: Patients who attended a rehabilitation programme for chronic pain/ fatigue underpinned by the framework, and clinicians (physiotherapists and psychological wellbeing practitioners) who led the programmes. Results: Seventeen interviews were conducted, involving 12 patients with chronic pain/fatigue and five clinicians. The framework analysis revealed four deductive themes: (1) Acceptability of the activity pacing framework, (2) Acceptability of the feasibility study methods, (3) Processes of change and (4) Barriers and facilitators to activity pacing; and one inductive theme: (5) Perspectives of patients and clinicians. Conclusions: The activity pacing framework appeared acceptable to patients and clinicians, and adherence to the framework was demonstrated. Processes of behaviour change included patients' regulation of activities through activity pacing. Barriers to pacing included work/social commitments and facilitators included identifying the benefits of pacing on symptoms. Different perspectives emerged between clinicians and patients regarding interpretations of symptom ‐ contingent and quota ‐ contingent strategies. The framework recognises fluctuations in symptoms of chronic pain/fatigue and encourages a quota ‐ contingent approach with flexibility. Future work will develop a patient friendly guide ahead of a clinical trial to explore the effects of pacing.


| INTRODUCTION
The management of complex conditions of chronic pain/fatigue includes varying and individualised strategies to facilitate behaviour changes and improve physical and psychological wellbeing (Bourke et al., 2014;British Pain Society, 2013;Pearson et al., 2020). Activity pacing is one such strategy that is commonly advised to increase individuals' participation in meaningful activities while managing symptoms (Abonie et al., 2020;Antcliff et al., 2019b;Nielson et al., 2013). Activity pacing involves modifying behaviours such as overactivity-underactivity (boom-bust) cycling, avoidance and excessive persistence. Such behaviours may be adopted in an attempt to prevent symptoms, as a reaction to symptoms or to distract from symptoms (Birkholtz et al., 2004). Behaviours that are driven by symptom severity (symptom-contingency) may leave individuals feeling out of control. Although activity pacing aims to regulate activities, the approach to activity pacing varies across clinicians, researchers and patients. As such, there is confusion regarding how pacing is instructed in a manner that optimises this self-care strategy (Andrews & Deen, 2016;Nielson et al., 2013).
We have developed an activity pacing framework using multistaged mixed methodology, in accordance with the Medical Research Council (MRC) guidelines for developing complex interventions (Craig et al., 2008). With an inclusive approach to widen its relevance and usability, the pacing framework was developed for complex conditions with predominant symptoms of chronic pain and fatigue. Such conditions commonly present with overlapping symptoms, including anxiety, depression and reduced function, may coexist and even predict the likelihood of further somatic comorbidities (Aggarwal et al., 2006;Bourke et al., 2014;McBeth et al., 2015;Warren et al., 2013). Based on the findings from a national online survey across healthcare professionals (Antcliff et al., 2019b) and a nominal group technique (consensus meeting between patients and clinicians) (Antcliff et al., 2019a), the framework comprises of two booklets: 'Theory and Overview', and 'Appendices and Teaching Guide'. The framework contains a conceptual model and definition of pacing, the aims and components of pacing, and incorporates activity diaries and goal setting to practise pacing. The model of pacing is underpinned by a quota-contingent operant approach (setting meaningful and realistic goals), with components of flexibility, choice and control. The framework moves away from principles of symptom-contingency due to the known multifactorial processes that can impact on individuals' experience of symptoms of chronic pain/fatigue, not just pathophysiological processes (Moseley, 2003;Nijs et al., 2012;Raja et al., 2020). The framework was developed to be relevant to people with a range of abilities, and apply to a variety of activities including work, housework, exercise, relaxation, socialising and hobbies.
The feasibility of using the activity pacing framework in a rehabilitation programme for chronic pain/fatigue, and the suitability of the questionnaires to measure activity pacing and symptoms were assessed in a repeated measures study (manuscript under review).
Exploring acceptability is a key component of feasibility testing, and greater acceptability is considered to improve adherence to complex interventions by clinicians and patients (Sekhon et al., 2017).
Acceptability interviews are a useful method of process evaluation of fidelity and contextual factors that may influence the development and testing of a complex intervention (Craig et al., 2008;Moore et al., 2015;Sekhon et al., 2017). Therefore, the next stage in the development of the activity pacing framework involved interviewing patients and clinicians who were involved in the feasibility study.

| Aim
The aim of this study was to explore the acceptability of using the newly developed activity pacing framework in a rehabilitation programme for chronic pain/fatigue. Our specific objectives were to explore: 1. Patients' and clinicians' opinions on the acceptability of the activity pacing framework 2. Practical issues regarding the feasibility study methods to prepare for a future pacing trial 3. Processes of behaviour change (patients and clinicians) 4. Barriers and facilitators to activity pacing.

| Study design
Acceptability of the activity pacing framework was explored via semistructured interviews with patients and clinicians. The theoretical qualitative methodology that underpinned this study was framework analysis since this was a pragmatic study to explore participants' opinions on the acceptability of the activity pacing framework.
Framework analysis is widely used in policy and healthcare research, and holds advantages of deductive and inductive approaches (Gale et al., 2013;Ritchie et al., 2003). This study is reported in accordance with the consolidated criteria for reporting qualitative research (COREQ) (Tong et al., 2007).

| Participants
Participants included patients who consented to the feasibility study, completed the pre-treatment questionnaire booklet and attended both sessions on activity pacing (weeks 2 and 3) during the six-week rehabilitation programme. Eligible patients included those with an initial GP/hospital consultant referral with diagnoses of chronic low back pain, chronic widespread pain, fibromyalgia or myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), with symptoms ≥3 months. Patients were required to read and write in English.
Ineligible patients were those with evidence of a serious underlying pathology (e.g., current cancer diagnosis), or patients with severe mental health or cognitive functioning issues. Patients were recruited via purposeful sampling to include those with varying conditions of chronic pain/fatigue, and patients who completed/did not complete the rehabilitation programme.
Purposeful sampling recruited clinicians who were qualified physiotherapists (Physio) and psychological wellbeing practitioners (PWP), who had attended a half-day training on using the activity pacing framework and implemented it in the rehabilitation programmes. Clinicians were observed delivering the pacing content of the rehabilitation programme for fidelity (by DA) and completed a fidelity checklist each programme to ensure their inclusion of key components of the framework.

| Recruitment
Patients were contacted via the telephone to invite them to take part in one acceptability interview after they attended the programme.
Clinicians were invited to participate in person or via email/telephone. Patients/clinicians were sent written information regarding the study and provided written consent in advance of the interview and verbal consent at the start of the interview.

| Data collection
Participants were interviewed via the telephone or face-to-face in a healthcare setting, according to their preferences. Interview questions were developed by the research team that were informed by current literature on pacing and the findings from the previous stages of development of the activity pacing framework; and the questions were developed to facilitate an exploration of the acceptability of using the activity pacing framework in the clinical setting. Specifically, patients' interviews included discussions on how activity pacing was instructed, their opinions/challenges of pacing, using activity diaries/ goal setting, and the ease of completing the questionnaire booklets.
No further guides or prompts were administered to patients since the purpose of the interviews was to discuss their experiences. Clinicians' interviews discussed the content and clarity of the framework and its usability in the clinical setting; their opinions/experiences of activity pacing and any challenges of the feasibility study (see Figure 1). All interviews were digitally recorded and transcribed verbatim. Interview data were anonymised using unique study codes. Patients and clinicians were sent their interview transcription to check for accuracy. Field-notes were made during the interviews regarding any contextual factors and as prompts for data analysis (Morse & Field, 1996).

| Data analysis
The qualitative data from the transcriptions were analysed using framework analysis. Framework analysis is suitable for analysing semi-structured interviews due enabling deductive processes to answer specific questions, while allowing for inductive processes to develop new ideas. The framework stages include: familiarisation with the data, identifying a thematic framework, indexing, charting and interpretation (Ritchie et al., 2003). The NVivo program (Version 12) was used to manage the qualitative data. Data analysis was undertaken by the Chief Investigator (DA) working alongside a researcher (LMc) with an expertise in qualitative research methods. A patient representative independently indexed three interviews which were compared and discussed to reach an agreement and develop the indexing codes.

| RESULTS
Interviews occurred throughout the feasibility study (September 2018-November 2019) and patients were interviewed within approximately three months of starting the programme. The duration of interviews was 18-41 (mean = 32) minutes for patients and 20-39 (mean = 28) minutes for clinicians.

| Participant demographics
Of the 16 patients invited to participate, 12 consented and were interviewed (recruitment rate = 75.0%). Nine patients completed the programme and three did not. Reasons for noncompletion included: unavailability, a family bereavement, and feeling younger and therefore disconnected from the group. All five clinicians (three physiotherapists and two PWPs) who were invited to participate, consented and were interviewed (see Table 1).

| Framework matrix
The Framework matrix contained five main themes. Four deductive themes addressed the objectives of the study: (1) Acceptability of the activity pacing framework, (2) Acceptability of the feasibility study methods, (3) Processes of change and (4) Barriers and facilitators to activity pacing. A fifth theme emerged inductively: (5) Perspectives of patients and clinicians. Initially, 48 codes were indexed. Following the cross-check with the patient advisor, four further codes were added ('other coping strategies', 'effects of pacing on others', 'understanding of condition' and 'mind-set').
When charting the data, codes were grouped together and summarised (see Table 2). ANTCLIFF ET AL. -3 (1) Acceptability of the activity pacing framework Patients found the instructions on pacing to be clear and thorough. Patients benefitted from pacing being introduced in the context of unhelpful behaviours, including illustrations of boom-bust, excessive persistence and avoidance behaviours: "It was talked about and explained on a white board… showing a chart of the boom-bust cycles, and how to recognise when you're doing that and how to try and avoid it" (Patient3) All clinicians reported the framework provided a comprehensive guide from which they could instruct patients, and there was nothing either missing or redundant. They felt able to deliver the framework in their own style and in a meaningful way to patients, while maintaining fidelity to the framework. The conceptual model of the framework supported a rehabilitative approach which encouraged progression, together with aligning with other psychological approaches: "It's progression too, so once you've established good pacing strategies, the idea is you want to progress in the future, which I think is always good at keeping people motivated to keep up with it." (Physio1) "I think it's [pacing] also a good 'in-road' from the psychological type of things to just maybe challenging behaviours." (PWP2)

Examples of interview questions for patients
Understanding of activity pacing activity pacing Barriers to activity pacing Activity pacing tools and study practicalities

Examples of interview questions for clinicians
Opinions on pacing

Content and clarity of the activity pacing framework
Barriers to adherence with the pacing framework The training process and study practicalities

F I G U R E 1 Examples of interview questions
Clinicians and some patients found the activity diaries a useful tool to recognise behaviours. Three patients did not find activity diaries helpful due to completing other paperwork, or self-reported low motivation or depression. Clinicians thought goal setting was helpful for patients to practise pacing, to individually tailor pacing and to facilitate patients to monitor their own progress: "Because of the way we do the goal setting, we do take time…pacing isn't just an abstract concept in session, it's about how are you going to go away and apply it." Patients set goals to socialise with friends, try varying exercises, protect time for hobbies and relaxation, and gradually try activities they had been avoiding due to symptoms. Seven patients reported benefits of goals to embed the pacing theory into practice:  (40) Profession: Psychological wellbeing practitioner (PWP) 2 Duration of specialism in chronic pain/fatigue (in years): range (mean) 2.0-14.0 (6) ANTCLIFF ET AL.
-5 T A B L E 2 Overview of themes and subthemes from the framework analysis  All patients believed that pacing was relevant to them, including patients across a range of ages and working statuses. Patients believed that pacing could be relevant to various medical conditions and also for those without a medical condition.

(5) Perspectives of patients and clinicians
Some differing opinions emerged inductively between clinicians and patients, for example, regarding interpretations of quotacontingent or symptom-contingent pacing. Clinicians recommended quota-contingency as an appropriate approach to activity pacing for chronic pain/fatigue:

| DISCUSSION
This study explored the acceptability of a new activity pacing framework, fidelity to the framework and contextual factors as a process evaluation of this complex intervention development.
Acceptability relates to clinicians' and patients' perception of the appropriateness of an intervention based on their expectations or experiences (Sekhon et al., 2017). The framework appeared acceptable for clinicians in terms of its conceptual model, and it had clinical utility due to the applied approach (e.g., goal-setting). The framework provided a structure and standardised guide for what is considered an ambiguous coping strategy (Gill & Brown, 2009;Nielson et al., 2013). Clinicians' feedback on the key themes of activity pacing added evidence of their fidelity to the framework. They commended the framework on including psychological approaches and experiential learning. Such components are crucial, since patients' symptoms encompass a personal experience, that is, underpinned by complex bio-psychosocial factors and a learned response to pain (Raja et al., 2020). Patients' feedback alluded to their adherence to the framework, confidence to implement pacing, autonomy to manage their health conditions and perceived benefits. Such findings link with the Theoretical Framework of Acceptability (TFA) constructs of: affective attitude, intervention coherence, perceived effectiveness and self-efficacy (Sekhon et al., 2017).
Clinicians' process of behaviour change through using the framework included leading thorough discussions with patients on avoidance and excessive persistence, not just boom-bust behaviour.
This context to pacing facilitated the individualised tailoring of pacing. The benefits of tailored pacing may include lower fatigue and more consistency of activity levels (Murphy et al., 2010(Murphy et al., , 2012 (Nielson et al., 2013).
Participants' perceptions of barriers and facilitators to pacing relate to the TFA constructs of 'burden' and 'opportunity costs' (Sekhon et al., 2017). These comprised of environmental and social influences, together with psychological and emotional factors. Other barriers included changing habitual overactive/underactive behaviours or personality traits, similarly to other studies (Andrews et al., 2015;Cane et al., 2016). Facilitators to pacing included patients' and their families' recognition of improved emotional, cognitive and physical wellbeing.
Some differing opinions emerged between clinicians' and patients' perspectives of symptom-and quota-contingency. Quotacontingency supports the principles of pain education, including explanations that symptom severity may not always be explained by tissue damage (Raja et al., 2020); and endorses the aim of rehabilitation programmes to increase function rather than directly reduce symptoms (British Pain Society, 2013). Furthermore, a reduction in pain may not be a feasible expectation of pacing (Guy et al., 2019).
However, the impact of symptoms cannot be ignored while pacing. To the contrary, symptom severity plays a role in pacing when patients identify baselines of manageable activities. Baselines are centred on individuals' tolerance levels and undertaking activities in a manner that does not trigger a set-back (Nielson et al., 2013). Symptoms of chronic pain/fatigue can also fluctuate. Therefore, the framework encourages flexibility within a quota-contingent approach; to acknowledge symptoms while ensuring pacing remains relevant and achievable. However, confusion may arise between clinicians' instructions of quota-contingency with flexibility, and patients' interpretation as symptom-contingency.
Clinicians and patients had convergent opinions that flexibility was an important component of setting goals and when making considered decisions whether or not to pace. Similarly, principles of psychological flexibility include active decision-making to change or persist with an action with consideration of individuals' goals, emotions and situation (McCracken, 2013). Participants were aligned in understanding that pacing involved acceptance and self-compassion.
Components of acceptance contained in the framework, such as enabling appropriate adaptations of activities and encouraging satisfaction with achievements are important components of pacing (Andrews et al., 2015;Cane et al., 2016), and rehabilitation programmes as a whole (Kallhed & Mårtensson, 2018). Furthermore, participants agreed that pacing facilitated a sense of choice and control, as recognised in other pacing literature (Birkholtz et al., 2004;Pearson et al., 2020).
Pacing may have been confused by the minority of patients as resting after over-activity, or considered to be a natural response.
However, 'naturalist pacing' may be underpinned by symptomcontingent and reactive behaviours such as going slow and steady or taking breaks after activity rather than pre-planned pacing as a learned strategy (Murphy & Kratz, 2014). Naturalist pacing has previously been associated with reduced function and increased symptoms (Andrews et al., 2012;Guy et al., 2019;Murphy & Kratz, 2014). Furthermore, we believe that pacing comprises of more than simply resting after activity, slowing down or taking breaks (Antcliff et al., 2018).

| Strengths and limitations
Whilst the sample size was relatively small, it was in keeping with qualitative approaches. Purposeful sampling enabled a variety of participants to be involved including clinicians of different professions and patients of varying ages, conditions and abilities, together with those who did/did not complete the programme. Through including diverse opinions, rich data were collected; and purposeful samples may reduce sampling bias (Ayres, 2007;Tong et al., 2007).
Bias may have arisen during the data collection and analysis since the lead author undertook all of the interviews and led the framework analysis. Participants were informed that the lead author is a physiotherapist and researcher in chronic pain/fatigue. The author did not routinely deliver the rehabilitation programmes, but may have had contact with some patients during their treatment. To reduce bias and increase the patient voice, a subset of interviews were coded independently by a patient advisor. Consequently, new codes/subthemes were added which may reflect the lived experience of implementing pacing.

| CONCLUSIONS
We have developed a comprehensive activity pacing framework to facilitate the modification of behaviours to support the management of chronic pain/fatigue. The framework appears clinically usable and the conceptual model appears acceptable. The framework encourages quota-contingent goal setting with flexibility, acceptance, choice and control to create meaning and relevance for patients. Future study will assess the effects of using the activity pacing framework in a clinical trial.