A pilot randomised trial of home-based rehabilitation compared to usual care in patients with heart failure with preserved ejection fraction and their caregiver’s (REACH-HFpEF study): a process evaluation.

Background: Whilst heart failure (HF) with preserved ejection fraction (HFpEF) affects almost 50 percent of the HF population, evidence-based treatment options remain limited. However, there is emerging growing evidence of the potential value of exercise-based cardiac rehabilitation. This study reports the process evaluation of the Rehabilitation Enablement in Chronic Heart Failure (REACH-HF) intervention for HFpEF patients and their caregivers conducted as part of the REACH-HFpEF pilot trial. This study reports the process evaluation conducted as part of the REACH-HFpEF trial. Methods: Mixed methods Pprocess evaluation sub study parallel to a single centre (Tayside, Scotland) pilot randomised controlled trial with qualitativequantitative assessment of both intervention delity delivery and a qualitative exploration of HFpEF patients’ and caregivers’ experiences. The Rehabilitation EnAblement in Chronic Heart Failure (REACH-HF) intervention consisted of self-help manual for patients and caregivers, facilitated over 12 weeks by trained healthcare professionals. Data included audio-recorded intervention sessions; demographic information; intervention delity scores; and qualitative interviews conducted following completion of intervention in a purposeful sample of 15 HFpEF patients and 7 caregivers., Results: Intervention delity analysis was indicative of the adequate delivery of the majority of the REACH-HF components, with three items relating caregiver involvement (addressing emotional consequences of being a caregiver, caregiver health and well-being, and closure of the session) scoring poorly. Qualitative information from the facilitator interactions and interviews identied three key themes and a number of sub-themes for patients and caregivers: (1) understanding their condition (2) emotional consequences of HF, and (3) patients’ and caregivers’ responses to the REACH-HF intervention were uncovered by qualitative interviews. Different professional backgrounds and typical models of service delivery of facilitators demonstrate to possibility of delivering this through either our HF or CR services of a combination of the two. Conclusions: The REACH-HF home-based facilitated intervention for HFpEF appears feasible and well accepted model for delivery of a CR intervention, with the potential to address key unmet needs of patients and their caregivers who are often

EnAblement in CHronic Heart Failure (REACH-HF) is a healthcare professional facilitated home-based rehabilitation intervention designed to improve self-care and health-related quality of life in people with HF. Using qualitative research methods, this pilot trial process evaluation sought to address the uncertainties of: (1) whether the REACH-HF intervention could be delivered with acceptable delity and (2) patients' and caregivers' experiences of participation in the intervention What are the key feasibility ndings?
Results of this process evaluation sub study of this single centre pilot trial showed the REACH-HF intervention was largely successfully delivered and well received by participants. However, whilst the delity analysis found the interventions to be delivered adequately over many of its components, we also found there was scope for improvement -particularly in relation to engagement with caregivers.
What are the implications of the feasibility ndings for the design of the main study?
This study highlights the genuine need for support for HFpEF patients and their caregivers. Results will guide the research team in the design and delivery of a recently funded multicentre trial of REACH-HF in this population. Emphasise in the trial recruitment patient-facing documentation, the value of coinvolvement of a caregiver (such as a spouse, family member, or friend) who can actively support the patient and caregiver with their engagement in intervention. Enhancing the training of healthcare professions who will facilitation the intervention to recognise the challenges and opportunities of engaging caregivers and the key role that caregivers can bring as agents of sustainable patient behaviour change. Undertake an analysis (qualitative and quantitative) of the delity of intervention delivery in the main trial to check if this caregiver engagement is achieved and to explore how it impacts on HFpEF patient and caregiver outcomes.

Background
In the United Kingdom (UK), approximately one million people live with heart failure (HF) -a condition which negatively affect cardiovascular functioning, often presenting with debilitating symptoms of fatigue, shortness of breath (dyspnoea), reduced exercise capacity, and a potentially dangerous accumulation of uid in bodily tissues [1]. Almost 50 percent of heart failure (HF) patients have preserved ejection fraction (HFpEF) and its prevalence is predicted to grow [1][2][3]. Although these patients are more often women, generally older, with a higher prevalence of co-morbidities (hypertension, diabetes and atrial brillation) and are less likely to have coronary artery disease than those with HF with reduced ejection fraction (HFrEF), their prognosis, associated morbidity, mortality, health-related quality of life (HRQoL) and healthcare costs are comparable [1][2][3].
The health burden of HFpEF on patients, caregivers, the health system and the broader economy, is substantial -with markedly reduced ability to perform activities of daily living, poor health-related quality of life, high rates of unplanned hospitalisations, high costs and premature mortality [4,5]. In contrast to HFrEF, where evidence-based therapies result in improved life expectancy and health-related quality of life, there is an absence of evidence-based treatment options for HFpEF [6]. Drugs or devices shown in trials to be effective for HFrEF, have not successfully altered prognosis in individuals living with HFpEF [7][8][9].
In contrast to the neutral outcomes of device and drug trials in the HFpEF population, there is promising evidence suggesting that exercise-based cardiac rehabilitation (CR) can bene t people with HFpEF [10]. CR is traditionally delivered in supervised group hospital-based programmes. However, given the suboptimal uptake of CR in people with HF, there is a need for alternative models of CR delivery, including home-based interventions [11]. The Rehabilitation EnAblement in CHronic Heart Failure (REACH-HF) is a healthcare professional facilitated home-based rehabilitation intervention designed to improve self-care and health-related quality of life in people with HF and their caregivers and to improve access to CR [12].
The HFpEF trial was a single centre study with the aim of assessing the feasibility of undertaking a de nitive randomised trial to assess the clinical effectiveness and cost-effectiveness of the REACH-HF intervention in patients with HFpEF and their caregivers [13]. The ndings and the patient, caregiver and cost results of the REACH-HFpEF trial are reported elsewhere [14]. This paper reports on a process evaluation substudy of the pilot REACH-HFpEF trial that aimed to assess the delity of intervention delivery, and patients' and caregivers' experiences of participation in the REACH-HF intervention.

Design
Details of the REACH-HFpEF single centre (Tayside, Scotland) randomised pilot trial have been published elsewhere [13,14]. In brief, 25 HFpEF patients and 11 caregivers were allocated to either the REACH-HF intervention plus usual care (intervention group) and 25 patients and 10 caregivers to usual care alone (control group). Participating patients were aged 18 years or older and had a con rmed diagnosis of HFpEF on echocardiography, radionuclide ventriculography or angiography (ie, left ventricular ejection fraction ≥45% within the last 6 months prior to randomisation).
A process evaluation was undertaken that included a qualitative and quantitative assessment of both the delity, i.e. the quality and consistency of the facilitators delivery of the REACH -HF intervention, and a qualitative exploration of both HFpEF patient and caregiver experiences of the REACH-HF intervention through semi-structured interviews. The quantitative delity results are reported in our main pilot trial publication [14]. The intervention group formed the sampling frame for the process evaluation. Patients were sampled for maximum variation techniques based on age, gender, presence of a caregiver, and psychological well-being (assessed by Hospital Anxiety Depression Scale [15]) providing a purposive subsample of 15 patients and 7 with caregivers who agreed to participate in the qualitative interviews. In accordance with the pilot trial protocol [13], 6 of these patients and caregivers agreed to participate in the delity analysis to assess the quality and consistency of what was delivered with the planned intervention.

REACH-HF Intervention
The REACH-HF intervention is a comprehensive 12-week practitioner facilitated self-care support programme co-designed with HF patients, caregivers, and healthcare professionals [12]. It comprises: (1) a patient 'Heart Failure Manual' that provides information and interactive elements which target patients understanding of and adaption to living with HF, their medications, the rationale for engaging in exercise and how to monitor and manage HF associated symptoms and stress. The manual content was modi ed for this study to re ect relevant medications, causes, and treatment of HFpEF [13]; (2) a 'Progress Tracker' to record, review and monitor symptoms, well-being, physical activity and other self-management behaviours; (3) a choice of exercise training programme (chair based and walking), (4) information of stress managing stress and anxiety (including a relaxation CD); and (5) a 'Family and Friends Resource' for caregivers that provides information to support the HF patients, as well as manage their own health and well-being.
REACH-HF participants were supported by one of two nurse facilitators (one with experience in CR and the other in HF) who had undergone a 3-day training course on the intervention. During the 4 to 6 anticipated facilitator contacts (an initial 60-90 minutes face-to-face consultation at their home, up to three further 30-minutes visits, plus 2 to 4 telephone contacts). Facilitators sought to increase the patient and caregiver understanding of living with and self-managing their HF.

Data collection and analysis
Data was captured from multiple sources. This included: Fidelity of intervention delivery: Facilitator interactions with participants were audio recorded and the quality of delivery was assessed by an experienced researcher (KS) . A sub-sample (from 3 patients) were independently checked by a second experienced qualitative researcher (JC). Scoring was discussed and compared to facilitate consistency. Listening to the detailed facilitated interactions provided additional rich data which would not have been illuminated through patients and caregiver interviews alone.
Semi-structured qualitative interviews: as de ned in the study protocol 15 Patients and their caregivers (n=7) were interviewed by JC in participant's homes or by telephone where a visit was not possible, using a pre-de ned topic guide (see e-supplement) after the completion of their intervention. Participants' understanding of their condition, engagement with and response to the REACH-HF intervention in supporting their adjustment to daily living with HFpEF, the perceived bene t of the intervention, maintenance of self-care behaviours and coping skills were explored. Participants were encouraged to openly express their views by the researcher and through further probing. Interviews were audio-recorded and transcribed verbatim. Field notes were also completed (JC) to enable re ection on the process, the interviewer's performance and participants responses to questions and to provide contextual information to the analysis where relevant. Re exive memo notes were kept to assure transparency and trustworthiness of the analysis. Thematic analysis [16] of the transcripts (led by JC and supported by KS) included data familiarisation through repeated listening to the audio recordings and review of interview transcripts. Initial codes which summarised the content either descriptively or interpretively were created. Codes with common features were then grouped together into emergent themes, before nally being assigned to three interpretive overarching themes. The themes are illustrated using participant quotes. A sample of 3 transcripts was independently analysed by KS with a high level of agreement. All participants were asked if they wanted a copy of the interview transcript to review and add comments if desired, but none requested this.
Both facilitators were also interviewed about their experience of delivering the intervention (by KS) using the same process described above.

Study participants
The ow of study participants is shown in Figure 1. Between April 2015 and June 2016, 50 patient participants were randomised (intervention group 25; control group 25). The characteristics of the intervention group patients and caregivers who participated in the process evaluation as summarised in Table 1.
Patients were equally likely to be male or female with an mean age of 71 years (standard deviation 10.7 years). Caregivers were typically a spouse or partner, younger (mean age 66 years, standard deviation 10.6 years) and female. The process evaluation sample was representative of the wider intervention group [14].

Fidelity of intervention delivery
The six patients and caregivers included in the delity analysis contributed a total of 41 facilitator interactions. Of these, 34 of which were face to face (average duration: 63 minutes, range 10 to 154 minutes) and 7 telephone contacts (average duration: 6 minutes, range 5 to 13 minutes). All patient/caregiver and facilitator face to face contacts were recorded with one exception due to audiotape malfunction. In contrast not all telephone interactions were recorded and in those which were, the recording quality of the patient's conversation was poor. The content of the telephone interactions identi ed that they were often used to check in with people and con rm the next scheduled face to face appointment rather than for goal setting or discuss health issues.
The audio-recording analysis of the interactions highlighted some excellent examples of the skilled facilitation using active listening skills. Facilitators listened and responded to concerns, addressed health issues, corrected misconceptions, and provided education, reasurance and support. They facilitated goal setting and pacing within daily living and behaviour change by enabling and empowering participants to better manage their condition and engage in the REACH-HF intervention. They instilled con dence in patients and care givers through supportive interactions. In addition to many examples of excellent practice, this data also highlighted areas for improvement for example when important cues were missed by facilitators and the absence of relatives in the interactions, or even when present the lack of caregiver interaction.

Semi-structured qualitative interviews
All 15 patients and 7 caregivers completed the interviews lasting an average 42 minutes (range 7 to 70 minutes) and were undertaken either in participant's homes (21) or by telephone (1). Three overarching themes and their related subthemes emerged from the analyses: (1) understanding their condition, (2) emotional consequences of HF, and (3) response to the intervention.

Theme 1: Understanding their condition
Many participants were unaware of their HF diagnosis and its potential severity. Participants often described a protracted and uncertain path to diagnosis of their HF, with symptoms being masked by other underlying conditions and con icting diagnostic information being provided by clinicians.

Reaction to diagnosis
For a few participants, particularly those who perceived themselves as ' t' and healthy, their HF diagnosis was a 'shock' which challenged their current identity.
"… I was shocked, I couldn't believe it. I just couldn't believe, because I've always been very t" [Patient interview 3] Following diagnosis others reported that they were 'too frightened to do anything'. However, the majority (13 patients) did not regard their HF as fatal, and believed they could accommodate it in the way they had with their other long term conditions seeing HF a continuation of a biography of adaptation to illness and disability. For many patients, their diagnosis came as a relief because it normalised and explained their symptoms e.g. tiredness and breathlessness, making them less anxious and enabled them to explain their symptoms and condition to others. The majority felt the description of HF in the manual, combined with the facilitator's explanation, aided their understanding of HF thus equipping them better to untangle, identify and act on HF symptoms. Some avoided reading or asking about HF believing this was 'morbid', and a source of stress which reminded them of the possibility of death.
Similar extremes in perspectives were reported by caregivers with some viewing HF as ' nal' (a kind of death sentence), requiring constant surveillance and for this group the role of caregiving was extremely stressful.
"I think when you get diagnosed with heart failure, from my point of view, the very word of heart failure is absolutely terrifying...And the word 'heart failure' is so completely nal… you're sort of…you're never relaxing. You're always watching to see he's okay. You're waking up in the middle of night, if you've got to get up and you're looking at him to make sure he's breathing… And then you think, good, he's just sleeping.…if he's not moving, you think, why hasn't he moved?" [Caregiver Interview 20] In contrast, others perceived that minor changes to health behaviours such as healthy eating and exercise were su cient to maintain a good health-related quality of life. and another that "Everything's been taken away." [Patient Interview 13] They often compared their lives now to those before their illness for example related to their occupational roles or physical tness. They were frustrated by how others (e.g. family, health care staff) now perceived them i.e. as individuals struggling with the constraints of their condition. Some yearned for the opportunity to demonstrate their more positive 'former selves' e.g., con dent people with a purpose in life. One participant even expressed it may be better for himself and his family if he were dead "I don't want to be here, and everybody says: That's not fair to your wife or your kids. Wait a minute, I says, Really? This is unfair to my wife and my kids. My wife deserves to be taken away for the weekend. I can't do that." [Patient interview 13] Caregivers con rmed this loss of social and professional roles in HF patients and acknowledged their personal challenges in managing such strong negative emotion. Caregivers highlighted the importance of regaining 'a sense of purpose "… because I think he feels worthless. Sometimes I think he wishes he wasn't here." [Caregiver interview 13b] Instead, when patients felt useful (e.g. helping other people) or socialised and interacted with others it lifted them emotionally and motivated them to care for themselves.
Recognising and responding to emotion Reports of anger or low mood often related to their feelings of frustration associated with the limitation's HF imposed on their lifestyle, were provided by patients and caregivers. For six patients the manual helped them to recognise altered mood. Working with the facilitators enabled better management of these emotions sometimes drawing on existing strategies e.g. mindfulness or using new techniques e.g. relaxation. Enabling patients to understand that that these feelings were 'normal' under the circumstances, allowed caregivers to support the patient's psychological adjustment to their HF. Caregivers suggested that the intervention had reduced anxiety and improved mood particularly in patients with elevated HADS scores. As one caregiver described "I just feel once he started to understand more about heart failure, with the manual, that yes, he sort of -I don't know, sort of maybe accepted it more…. I think sometimes he sort of panics, thinking oh you know, should I be feeling this way? Whereas having the manual has, I think, sort of made him realise yes, this is normal for me to feel like this and be like this." [Caregiver Interview 18] Caregivers also reported how the intervention positively addressed their own personal anxieties, thus allowing them to be more supportive. As one wife said "someone like myself who needs the con dence to know how to understand heart failure, to know how be less anxious... because your stress goes on to the patient … and can make them more anxious. So, if you understand maybe a little bit more about it. …. you can sort of be more of a support. I think that's what I'm trying to say." [Caregiver Interview 20] Theme 3: Response to REACH-HF Intervention Engagement with the REACH-HF Intervention While all participants engaged with the intervention at some level, this varied across the components. Participants con rmed that the manual provided information and reassurance, 'offering something for everyone'. In combination with the Progress Tracker this aided symptom monitoring and supported selfmanagement. Patients and caregivers accounts again reinforced their need to understand how to manage HF by knowing what to look for in case of deterioration and what to do in an emergency.
Through improved understanding caregivers felt more con dent in supporting the patients.
Most patients said that they followed the exercise recommendations within the manual, and progressed through the chair based exercise programme (delivered by DVD) as advised by the facilitator, either alone or together with their caregiver. A few needed guidance to prevent inappropriately rapid progression through the exercise levels. Facilitators proactively encouraged engagement in exercise, at times completing the chair or walking programme with resistant individuals. One participant, on the verge of giving up, described how the facilitator had supported him to identify and complete an alternative activity.
[She said] "No, if you can't do that what do you love doing?" I says, "I love walking." She went, "Right, if you want to go out for a walk, let's go out for a walk." [Patient Interview 13] Such support provided participants with the knowledge and con dence to continue this themselves.
Several caregivers who completed the chair exercise or walked with the patient valued this opportunity for social interaction and felt better emotionally afterwards. Those with a positive perception of the exercises were more strongly motivated to maintain them and integrate this into their lives. The biggest barrier to exercise was concurrent illness e.g. chest pain which either delayed exercise initiation or progression or episodes of acute illness (e.g. chest infection) which affected six participants and resulted in stopping exercise for several weeks then restarting this at a reduced level. Of four patients with co-ordination and balance problems, two adapted by holding the back of chairs for balance and slowly progressing through the chair exercises. In contrast, the other two discontinued their exercise programme.
When progress through the exercise levels did not match patients' expectations they were disappointed. The facilitators played an extremely important role, encouraging them, a rming progress and suggesting more appropriate alternative exercise. Five other patients were also disheartened feeling the chair exercises had not increased in intensity su ciently nor challenged them enough even at level seven. Caregivers con rmed how the combined manual information and advice from the facilitator increased their con dence to gauge and support more appropriate levels of exercise for the patient.
The negative impact of HF on participants' normal lifestyle and abilities could be profound as illustrated by one man who loved gardening.
'I had just a little bit of turf about that long and I dug it up and whether it was the bending up or down or just pushing the shovel in that made me very unwell for I think it was about a week to recover. That's one of the things I've found now that if I push myself and overdo it it's the aftereffects that last longer' [Patient interview 7] Again, facilitators helped patients to reframe their thinking, set more realistic goals and breakdown their activities to make them manageable through goal setting and pacing, which maintained a sense of achievement.

Changes in health-related behaviours
The majority (14 patients) reported some change in behaviour as a result of the programme. Changes included maintaining the exercise regime beyond completion of the programme (13 patients), continued symptom tracking/monitoring (8 patients), and dietary modi cations (2 patients). Patients who perceived immediate symptomatic bene ts from the exercises were most likely to keep up their exercise regime. Others reported how this had also improved their sleep patterns. Most patients claimed they knew about and followed healthy diets, two men discussed the value of the healthy eating section. One participant described how involvement in REACH-HF motivated him to set new dietary goals which included healthier food shopping, and more home cooking. By setting goals and applying pacing techniques this enabled him to complete his weekly shopping which had previously been a challenge.
Only two participants (a 75-year-old male and 84-year-old female) continued to engage in undesirable health behaviours such as smoking, and consuming a high fat diet. These behaviours were life-long habits which the patients did not causally connect to heart disease or weight gain.
Caregivers often described facilitators as the primary motivation for behaviour change in patients and the manual was a useful resource to complement facilitator-patient interactions. Caregivers typically encouraged and supported patients to change their lifestyles, helping them manage symptoms, and engage in activities, reinforcing the facilitator's recommendations. Some also changed their own behaviours. For example one wife reported "[The facilitator] was very helpful for me in so many different ways. Helping me to understand heart failure….she encouraged me to go out walking…. Just the reassurance that things were better, that there was somebody there that was willing to, erm, say, well, okay, you're doing well. Even just the smallest amount of encouragement. And 'my husband always felt better after [the facilitator] went away. Because she felt…almost like a little security blanket, if you want to say. That somebody was there, somebody was asking." [Caregiver Interview 20] Feeling that someone 'cared', listened, answered questions and provided feedback and encouragement was important to participants.

Monitoring and symptom tracking
Use of the Progress Tracker to record weekly symptom monitoring and exercise progress was variable.
Some showed no desire to complete this, expressing that "Filling it all in…. that is a bit annoying you know what I mean" [Patient Interview 1] Others only completed speci c sections, most commonly the weight and symptoms section. For those who found the tracker helpful, this daily symptom assessment was translated into long term behaviour change extending beyond the REACH-HF intervention. Proactive symptom monitoring also improved patients' abilities to communicate with doctors to allow prescribing of appropriate treatment.
In contrast, a few participants disliked the repetitiveness of the tracker, even suggesting that this at times became the focus of their interaction with the facilitator "Every time she would come out she would start to look back through the stuff but she would go right to the front of the manual, not the manual the chart you call it, and would go through preceding weeks that she'd already covered" [Patient Interview 7] Nearly all caregivers believed that monitoring and assessing the physical and mental health of the patient was the most important but also a very di cult aspect of their role. Identifying signs and symptoms, deciding on their seriousness, and whether they related to HF or another condition then initiating appropriate action was challenging. As one wife described "I suppose his breathing and I know like he's been quite concerned about sort of circulation in his legs. I don't even know if that's connected to the heart failure or if that's something else, because he has got quite a few health problems." [Caregiver Interview 18] Proximity to the patients and frequency of contact also in uenced caregivers' perceptions of their engagement in REACH-HF. One caregiver despite living over 50 miles away provided an excellent example of monitoring her relative's physical and emotional state and adopting a 'virtual caregivers role' providing encouragement and emotional support through mobile technology using texts or more often FaceTime.
She described the bene t of how "you can see it on him, to be honest. Sometimes he doesn't look too good… his breathing isn't good and he looks kind of grey… I suppose FaceTime is a different way of doing things and…. It's lets you be involved." [Caregiver Interview 18] Although using technology allowed her to assess his appearance, body language and suggest interventions in a similar way to face-to-face caregivers, what differed was her limited ability to provide physical assistance. The patient could also choose not to converse over visual media (especially if they are feeling particularly unwell). This obviously undermines the virtual caregiver's ability to assess the situation, and can cause them stress, worry and sleepless nights Unique Caregiver views and experiences Within this study there was a strong reluctance to be identi ed as 'caregivers', even when the 'caregiver' assisted the patient in activities such as washing and dressing. Caregivers regarded their caring role as ' uid'. Most described providing minimal physical assistance on a day-to-day basis with increased help when away from home or during episodes of patient's illness. Caregivers also highlighted how balancing competing demands on their time (e.g., caring roles for other family members), or their own health status, could affect the support they were able to give. Despite these challenges, caregivers did report examples of acting as an enabler and motivator especially in encouraging patients to exercise, often by doing this together.
Use of the friends and family resource also varied. Some read this from cover to cover, then used it as a reference (to review the meaning of symptoms or reinforce the facilitator's advice by referring the patient to that section of the manual) and a guide to action. However, the majority were intermittent engagers often reading the information explaining HF or quickly glancing through it. Caregivers with no or intermittent engagement believed that the manual was primarily for the patient.

Discussion
This process evaluation study has bene ted from a qualitative approach which enabled greater understanding of the issues surrounding HFpEF and the application of REACH-HF intervention. Through multiple data sources, we observed that the intervention was largely successfully delivered and well received by participants. This study also highlights the genuine need for support in a population often excluded from many HF and CR services [6]. The home-based nature of the REACH-HF intervention also offers an opportunity to overcome the current challenge of suboptimal uptake of CR for HF [1,11].
Whilst previously reported quantitative delity analysis found the REACH-HF intervention was adequately delivered by facilitators over the majority of its components, we also found there was scope for improvement [14]. This was particularly the case in relation to engagement with caregivers, a nding also reported with HFrEF patients [17]. Unique insights from the analysis of audio recordings of facilitator interactions provided rich data extending beyond the con nes of the previously reported delity scores, exemplifying good practice and identifying potential areas for improvement in consultation skills.
Facilitators also captured written notes of their consultations as part of a self-assessment.
Complementing these with re ection on the recordings of consultation, offered a powerful tool to enable self-re ection and professional development for practitioners. Caregivers themselves believed that REACH-HF was for the patient and not for them, suggesting a more proactive strategy for caregiver involvement may be required in future intervention delivery [17].
The need to understand about, and know how to manage their HF, reinforces earlier research in HFrEF patients [18,19] and caregivers [20,21]. Addressing participants' needs for clarity in relation to their diagnosis and the implications of this condition can increase understanding, alleviate patient and caregivers' anxiety, and allow them to accept and accommodate HF in their lives. Greater knowledge and con dence in caregivers, as shown in this study, can also enable more appropriate monitoring and support con rming previously reported ndings of optimised symptom management and self-care behaviour [21,22]. Some participants valued monitoring as a measure of their progress and stability, others seeing this as a chore which has also been reported in the use of symptom monitoring diaries [23].
Increasing patient's and caregiver's knowledge of HF, enabling them to link symptoms e.g. breathlessness to increased weight allowed earlier detection and prevention of deterioration, is a core element in HF care provision [24,25]. This was achieved through information provided in the manual and explanations by facilitators. The importance of feeling that someone listened and cared, acknowledged emotions, illness beliefs, anxiety and depression, yet provided feedback and encouragement to improve self-e cacy was highlighted by the participants. These are critical issues in empowering patients to selfmonitor and manage to optimise outcomes to improve health-related quality of life [26]. Applying goal setting and pacing techniques to break down tasks (e.g., shopping, housework and gardening) into manageable elements also allowed more proactive self-management of their condition.
The analysis from the current study suggests that HFpEF patients and their caregivers have a number of unmet needs and that the REACH-HF intervention may offer a possible solution to address this gap. The role of facilitators in implementing the REACH-HF programme is crucial. The facilitator had an important role in the prescription and support of exercise and other lifestyle change during the programme and enabled many participants to maintain exercise and dietary changes beyond completion of the programme, re ecting previous work [20]. Caregivers generally were more able to recall and describe the interactions between the facilitator and the patient or themselves better than the content of the REACH-HF manual. By employing counselling and coaching skills, listening to patients concerns, providing reassurance, reframing problems, helping them to adapt to any limitations, and motivating the patient and at times the caregivers themselves to take exercise, the facilitator had assisted the both patients and caregiver in their caring role.

Strengths and Limitations
This study has a mumber of strengths. First, it bene ted from a qualitative methods approach which enabled greater understanding of the issues surrounding HFpEF Second, REACH-HF is a home-based rehabilitation intervention for HFpEF patients (and their caregivers), a high need population with currently limited access to HF and rehabilitation services. Third, we successfully recruited the target number of HFpEF patients for semi-structured interviews, strengthened by including their respective caregivers, who are often excluded in HF research studies [27]. . Fourth, we assessed adherence to intervention protocols, the quality of consultation interactions and explored HFpEF patients' and caregivers' experiences of this.
Fifth, qualitative data also captured some examples of good practice in education, engagement and support of HFpEF patients and their caregivers. Fifth, completing delity analysis may be a useful tool for self-re ection and improving professional practice for specialist nurses. Finally, we believe this process evaluation enhanced the reliability of the results and are in keeping with the ndings of the process evaluation conducted alongside our multicentre RCT in patients with HFrEF [17]. Our learning from this study will inform future optimisation of the interventions for HFpEF [28,29] and a full trial in a number of ways: (1) emphasise in need for the trial recruitment patient-facing documentation to include of coinvolvement of a caregiver (such as a spouse, family member, or friend) who can actively support the patient with their engagement in the intervention; (2) enhance the training of healthcare professions who will facilitation the intervention to recognise the challenges and opportunities of engaging caregivers and the key role that caregivers can bring as agents of sustainable patient behaviour change; and (3) undertake an analysis (qualitative and quantitative) of the delity of intervention delivery in the main trial to check if this caregiver engagement is achieved and to explore how it impacts on HFpEF patient and caregiver outcomes.
However, this study also had limitations. Our translation of complex interpersonal interactions into numerical scores within the delity analysis was unable to fully illustrate some of the excellent examples of good practice. Facilitators often demonstrated high levels of skill and competence in providing tailored educational and psychological support, enabling patients to reframe negative thoughts, engage in appropriate exercise and participate in self-management. Capturing qualitative data highlighted many unmet needs in both HFpEF patients and their caregivers, a group currently excluded from HF services, and in the main found the respective components of the intervention (the manual, exercise training, and facilitation) to be well received by patients and, when involved, their caregivers. The sample size within this study was small and the characteristics of the participants (predominantly of white ethnic origin) from a single centre limits the potential generalisability and may have failed to achieve theoretical saturation/information redundancy. Whilst the assessment of the delity of interventions by independent researchers enhanced con dence in the results, their varied professional backgrounds (nurse researcher and social scientist) may have in uenced interpretations of the delity scoring.

Conclusions
This process evaluation provides important evidence supporting the feasibility and acceptability of delivering the REACH-HF intervention that has the potential to address some key unmet needs in HFpEF patients and their caregivers. One of the most important intervention components identi ed by this study was the role of the healthcare facilitator, who acted as an educator, a source of support and reassurance, as well as a motivator and enabler. The facilitators helped to reframe participants thinking to enable engagement in activity, symptom monitoring and self-management of their HF through realistic goal setting and pacing. The study also identi ed how involving caregivers was at times challenging and a more proactive strategy may be required to optimise this part of the intervention in future applications and clinical trials. The ndings of this process evaluation will inform a future full trial.

Consent for publication
Patient consent included permission to use their anonymised data.

Availability of data
Transcripts will not be shared in their entirety to protect the anonymity of participants and healthcare staff delivering the intervention. However, requests for extracts of data will be considered on reasonable individual basis from the corresponding author. Figure 1 CONSORT diagram for REACH-HFpEF trial