Importance and Assessment of Quality of Life in Symptomatic Permanent Atrial Fibrillation: Patient Focus Groups from the RATE-AF Trial

Atrial fibrillation (AF) is a major burden on patients and healthcare services. These effects are projected to increase exponentially as our communities grow older and the incidence of AF in older people increases [1]. Although adverse outcomes in AF such as stroke rightly receive attention from clinicians due to their preventable nature, poor patient quality of life (QoL) often lacks consideration in clinical practice, despite also being amenable to treatment.

Patients with AF have significantly poorer health-related QoL [2], which includes comparison with both healthy individuals and those with other cardiovascular diseases [3]. This has been attributed to the variety of symptoms that AF patients can suffer, including lethargy, palpitations, dyspnoea, sleeping difficulties, chest discomfort, and psychosocial distress, as well as anxiety related to treatments and potential complications [4]. Although QoL is significantly related to mortality, AF-related symptoms do not necessarily track with clinical outcomes such as stroke, heart failure, or myocardial infarction, making extraction of QoL information important for routine clinical management [5]. The majority of published studies on QoL in AF relate to the response to rhythm control therapies such as antiarrhythmic drugs or ablation. In contrast, patients with permanent AF have even worse QoL [6], account for around 50% of all patients, and yet we lack an adequate description of underlying factors [7].

The current unknowns about QoL in patients with permanent AF limit the scope of how effective clinicians can be in addressing patient concerns. We designed a qualitative study, led by a Patient and Public Involvement (PPI) team, and embedded within a clinical trial; the RAte control Therapy Evaluation in permanent Atrial Fibrillation (RATE-AF) trial [8]. Our aim was to explore 3 broad domains: (1) The perspective of patients on core components of health-related QoL and how this is influenced by AF; (2) The measurement of QoL in AF and what tools were felt useful by patients to measure the response to treatment; and (3) Whether QoL was an important outcome that clinicians should address.

This mixed methods study was part of the RATE-AF trial programme, a prospective, open-label, blinded endpoint, randomised controlled trial of 160 patients with symptomatic permanent AF. The trial is the first to directly compare longer-term heart rate control using digoxin and beta-blocker therapy in this patient group (clinicaltrials.gov: NCT02391337, ISRCTN: 95259705, and EudraCT: 2015-005043-13). The rationale and design of the study have previously been described [7]; in brief, the trial was embedded within the UK National Health Service (NHS), with minimal selection criteria to reflect routine clinical care. Patients were aged 60 years or older, with permanent AF, in need of rate control, and breathlessness equivalent to at least New York Heart Association (NYHA) class II. As per guidelines, permanent AF was characterised as a physician decision for rate control with no plans for cardioversion, antiarrhythmic drugs, or ablation [9]. We only excluded patients with either clear requirements or contraindications for either drug, for example myocardial infarction in the last 6 months, a history of severe bronchospasm, bradycardia, or previous intolerance. The RATE-AF trial and the qualitative aspects were sponsored by the University of Birmingham and funded by the UK National Institute for Health Research (NIHR).

A team of 3 PPI members helped to design and manage the trial, including positions on the Trial Steering Committee. The design of the focus groups was led by J.J. and M.S. from the PPI team, with the support of cardiology, patient-reported outcomes research, and qualitative research teams at the University of Birmingham.

Patients for the RATE-AF trial were recruited from referrals to 3 hospital sites in Birmingham (Queen Elizabeth Hospital, City Hospital, and Heartlands Hospital), and also directly from General Practices across the West Midlands region in the UK. As part of the consent procedure, all participants were asked if they could be contacted to contribute to the focus groups. From this cohort, 20 patients were consecutively invited to attend based on completion of all drug uptitration at that time (i.e., beyond the first 2 months of their trial participation), purposely sampled by gender and randomised group. No clinical variables were used to decide on focus group composition, but the participants had to be able to attend on the specified date (5 patients refused/were unable to attend on the date and were replaced with the next available patients). One patient who accepted was unable to attend on the day due to illness. Focus groups were held at the NIHR/Wellcome Trust Clinical Research Facility at the Queen Elizabeth Hospital Birmingham in 2019 and split into 2 meetings for each arm of the trial. The first meeting focused on building rapport within the group and the impact of AF on their lives, with the second meeting focused on assessment and tools to measure that impact. Each meeting lasted approximately 2.5 h separated by two weeks (with a maximum of 10 participants in each meeting). Focus groups were led by the patient and public representatives (J.J. and M.S.) with D.K. also in attendance to address any medical issues.

Every patient received a fixed sum of GBP 50 for their contribution to each focus group, in addition to appropriate compensation for travel and subsistence costs. PPI members received funding according to NIHR INVOLVE guidance (https://www.invo.org.uk/). There was no industry funding for any part of this trial.

Three validated QoL questionnaires were used in the RATE-AF trial [7]. AF-specific QoL was assessed using the Atrial Fibrillation Effect on QualiTy-of-life (AFEQT) questionnaire [10]. Generic QoL was assessed using both the EuroQol EQ-5D-5L questionnaire [11] and the Short Form 36 Health Survey (SF36) [12]. The SF36 survey is comprised of eight domains; normalised UK values were taken from 8,889 respondents of a large-scale social survey, the Third Oxford Health and Lifestyles Survey (OHLS-III), sampled from primary care in the UK [13]. t tests adjusted for unequal variance were used to compare mean values with data from the baseline visit of the RATE-AF trial. All participants of the focus groups had experience of completing all 3 questionnaires on at least 2 occasions during their trial visits. At the end of the first meeting, copies of the questionnaires were also provided to the participants for discussion at the next meeting.

A topic guide was developed and finalised by the PPI team prior to the focus groups, and this was used as a roadmap for discussions in each meeting. The topic guide included specific questions relating to the three study domains and corresponding probe questions to explore these issues in more detail (online suppl. file; see www.karger.com/doi/10.1159/000511048 for all online suppl. material). Audio recordings of the meetings were made with the consent of all participants. A professional service transcribed the interviews (clean verbatim), and the transcription was reviewed for consistency and accuracy by J.J. and M.S. The data (recordings and transcripts) were analysed using thematic approaches [14]. To organise the patient comments into the 3 key domains of interest, we developed a set of codes after the authors had familiarised themselves with the content of the transcripts. For the components of health-related QoL in AF, the codes related to any comment on: Physical impact from AF; Emotional impact from AF; Daily life impact from AF; Impact on carers; Interaction with comorbidities; Prior knowledge of AF or lack thereof; and Importance of QoL to the patient. For measurement of QoL in AF: Value of the QoL questionnaire; and Comparison of QoL questionnaires. For the importance of addressing QoL: Treatment expectations; Impact of the trial medications on physical function; and Impact of the trial medications on QoL. An iterative process was performed of coding each transcript (J.J. and M.S.). The codes were compiled in a data spreadsheet with extracted statements given primary codes (main issue raised by the patient in that comment) and secondary codes (where, if required, an additional issue was raised in the same comment). J.J. and M.S. also manually extracted specific quotes from patients that were relevant or particularly significant to the group as a whole or may not have fitted into the specified codes.

RATE-AF trial data at baseline from 160 patients with symptomatic permanent AF confirmed a substantial reduction in QoL in all domains of the SF36 questionnaire, except for mental health. This was observed compared to the UK norm, as well as compared to those with a longstanding illness in population survey data (Fig. 1).

Nineteen participants took part in the focus groups, comprising 8 women and 11 men, all from a white British background. Mean age was 74 years (SD 7), with a range of 62–86 years. Patients had been diagnosed with AF for a mean of 5 years, and 10 (53%) had a known diagnosis of co-existing heart failure or signs of heart failure at baseline. Other common comorbidities were hypertension in 11 patients (58%), type 2 diabetes in 5 (26%), airways disease in 4 (21%), and previous stroke or transient ischaemic attack in 3 (16%). The demographics of the focus group participants were all comparable with the main trial population [8]. All patients had completed uptitration of their trial medication (10 randomised to digoxin and 9 to beta-blockers) and were on stable dosage with adequate control of heart rate. A summary of key findings, determined by the Patient and Public leads, is presented in Figure 2 and a summary for patients in Figure 3.

Key issues raised with regards to determinants of QoL and how AF affects the lives of patients were:

1There was consistent feedback about a lack of information from healthcare professionals about AF and a lack of focus on QoL in healthcare consultations. This led to a sense of frustration, isolation, and reduced confidence which contributed to the overall impact of AF on patient wellbeing.

2The impact of AF was not just felt by patients, but also by their primary caregivers and the wider family.

3There was marked variability in AF symptoms in individual patients, with some describing severe impact on physical capacity and activities of daily living due to breathlessness, fatigue, and dizziness, whereas others had few symptoms.

4In group discussion, the patients concluded that the impact of comorbidities (especially large-joint arthritis) was the most important determinant of the effect of AF on physical activity. There was consistent feedback that the effect and treatment of comorbidities was poorly considered in interactions with healthcare professionals.

5Adaptation to a new style of living was required after the diagnosis and treatment of AF, and there were considerable emotional impacts.

A summary of quotes from patients relevant to this domain are presented in Table 1.

The key points discussed about tools to measure QoL and treatment response were:

1Most patients found it difficult to complete questionnaires due to considerable day-to-day variation in their AF-related symptoms. There were challenges for questionnaires like EQ-5D-5L (which asks about impact today), as well as SF36 and AFEQT (4-week recall period).

2Separation of emotional from physical wellbeing (for example in SF36) was confusing for many patients who felt that AF impacts were inter-related.

3AFEQT had the most relevance to AF symptom burden, but attribution to AF or other comorbidities was challenging.

4There was a difficult balance to strike between content and time taken; for example, comparing the brevity of the EQ-5D-5L questionnaire with the more nuanced but also potentially duplicating questions in SF36.

5Consensus that collecting both generic and AF-spe-cific QoL would be useful in clinical practice to help clinicians understand the patient’s perspective; on balance, the group favoured using EQ-5D-5L with AFEQT.

A summary of relevant quotes from patients relating to each of the 3 QoL tools are presented in Table 2.

The major discussion points raised in the focus groups as to whether QoL was an important outcome that clinicians should address were:

1Improvement in QoL was the most important consideration for this patient group, ahead of mortality or the need for hospital visits.

2Healthcare professionals in prior consultations often prioritised issues that were important to them rather than the patient, such as deciding on stroke prevention or rhythm control treatments, or on a specific target for heart rate control.

3A lack of medication review in patients on longstanding treatments contributed to worse QoL and emotional/physical impact from AF.

4A focus on QoL improvement could engage patients leading to empowerment. The integrated approach in the RATE-AF trial of providing patient education and support was unanimously well received by patients in the focus groups who expressed a sense of confidence and better ability to self-manage their AF.

5Participation in the focus group itself was beneficial for the patients, who recommended a similar process of peer-support for patients newly receiving treatment for AF, for example, after discharge from hospital.

A summary of relevant quotes from patients are presented in Table 3; in general, these reflect a mixture of the psychosocial and physical impacts of AF, coping mechanisms, and outcomes related to these effects. Whilst this study cannot address whether attention to QoL would improve clinical outcomes, there was consensus from patients that it could improve the care pathway in patients with AF.

This study confirmed that patients with symptomatic permanent AF have considerably worse QoL than the general population or those with a longstanding illness. In focus groups, QoL and symptom management were the predominant concerns of these patients. The wide variation in symptoms experienced, both generic and AF-specific, were underscored by a substantial emotional burden on patients and their families. A reported lack of focus on QoL in prior healthcare consultations contributed to a loss of confidence and a sense of isolation that hindered medical management of their condition. Further, patients felt that the extent of their comorbidities was often neglected. To improve patient wellbeing, management protocols need to consider other conditions such as large-joint arthritis, and not just concentrate on anticoagulation therapies for AF.

There is increasing recognition that QoL and symptom data should be collected in routine clinical practice for patients with AF [15]; however, measurement of QoL in patients with AF is associated with a number of methodological challenges. In a systematic review of measurement properties for AF-specific QoL tools, we identified substantial validity concerns for commonly used questionnaires [16]. Hence, many research studies have used generic QoL tools, such as SF36 and EQ-5D-5L. Whilst generic questionnaires allow for comparison across different diseases, they lack attention on the types of symptoms that can impair QoL in patients with AF. In our focus groups, there was consensus amongst patients that a short generic tool, such as EQ-5D-5L, in addition to an AF-specific tool, such as AFEQT, provided the best balance of time taken versus information gained. Although SF36 is comprehensive, and from a technical standpoint appears to cover many of the concepts discussed by patients, there was concern around difficulty to separate emotional and physical impacts from AF, the need for questions to be explained by clinical staff, and duplication in responses. An added advantage of EQ-5D-5L is the use within health economics such as quality-adjusted life-year (QALY) analyses. Whichever QoL tool is chosen, our patient groups were clear that they could see a potential benefit in rolling-out these questionnaires into routine clinical practice. Not only would this give patients a framework for their consultations with healthcare professionals, but it would also allow staff to monitor changes in QoL in response to treatment.

Consideration of patient-reported outcomes is now part of international practice guidelines on AF management; clinicians are asked to assess and re-assess QoL, with symptomatic improvement a major treatment objective in patients with AF [9]. Aside from oral anticoagulation for stroke prevention, nearly all other treatments (including rate and rhythm control) are based on evaluation of symptoms. More research on assessment of QoL in clinical practice is clearly warranted to inform clinical decision making and to tailor care to the needs of individual patients. Electronic capture of these data in routine practice could allow for real-time monitoring, flexible and responsive scheduling of hospital appointments, early detection of problems, and prompt intervention to prevent AF-related adverse outcomes [17]. Attention on comorbidities is also an essential component for optimal management of AF. In the RACE III trial, 245 patients with early persistent AF and mild-to-moderate heart failure were randomised to either targeted therapy of underlying conditions or normal care including rhythm control [18]. Comorbidities were better treated in the intervention group and recurrence of AF was lower (75% of all patients were in sinus rhythm at 1 year, compared to 63% in the conventional management group; p = 0.042).

Similarly, integrated treatment programmes that improve patient and healthcare staff education, along with nurse-led and lifestyle interventions, have demonstrated improvements in clinical outcomes. In a randomised trial of 712 patients with AF, integrated care led to better adherence to guidelines [19]. The composite of cardiovascular hospitalisation and cardiovascular death was significantly lower in the integrated care arm, 14.3% compared with 20.8% with usual care (hazard ratio 0.65; 95% CI 0.45–0.93; p = 0.017). Bringing together patients and their healthcare professionals in order to make shared decisions is a key element in patient empowerment [9], especially in this digital era [20], and has the potential to address the persistently poor outcomes seen in older patients with AF [1]. Randomised trials to clarify the impact of AF education for healthcare professionals are ongoing [21]. As demonstrated in a mixed-methods study of 101 patients and 15 clinicians, a lack of appropriate education also hinders the ability of patients to self-manage their AF [22]. The importance of a good knowledge-base has been studied previously, mostly in the context of anticoagulation for stroke prevention [23]. Our impression is that education on other aspects of care, for example, heart rate control, is often neglected in clinical practice. This is particularly the case for patients with permanent AF, who often receive no other therapy and are, therefore, left without the support that patients with paroxysmal AF typically receive when considered for rhythm control.

An unexpected outcome from the focus groups was the benefit the groups themselves had for individual patients. Having a safe space to talk to other patients led to amelioration of the sense of isolation and lack of knowledge. As with other chronic diseases, many of the impacts of AF are psychosocial in nature, and so better adaptation can be a powerful tool (a clear, cross-cutting theme across the domains we investigated). More widespread use of patient support groups within secondary care could have profound benefit for patients and lead to reductions in healthcare utilisation. For example, in hospitalised patients, accredited information could be provided to all patients before, during, and after discharge, along with details of a national or local patient organisation for information and emotional support. Local patient support groups are run by charities such as the Arrhythmia Alliance and AF Association (https://www.heartrhythmalliance.org/), who also moderate online forums with thousands of members, organise Patient Educational Days, and provide dedicated helplines. Close partnership with these organisations has the potential to support patients and their carers, and further enhance health-related QoL in those with AF [24].

All included patients had permanent AF with symptom-related impairment of daily life at baseline (NYHA class II or above); hence the results presented reflect this patient group. The focus groups took place after control of heart rate and symptoms to better represent the broader community with managed permanent AF. We are limited by the number of patients involved and the need to provide depth over breadth of concepts. Greater numbers within each focus group would have limited discussion, and hence each meeting was restricted to 10 patients. We divided the sessions into two segments to avoid overloading the participants. Meetings were of sufficient duration to achieve saturation within each domain, and they were undertaken within a few weeks of each other to minimise any change over time. All meetings were scheduled in a comfortable and secluded area of the hospital research unit, well known to the patients to limit anxiety. Our analytical methodology used the broad concept of thematic analysis, with a pragmatic approach to reflect that patient and public representatives were leading the focus groups, coding transcripts, and providing data analysis. This provided a unique insight into patient views at many levels. Although the included patients were typical of those with permanent AF and symptom-related impairment of daily life treated in routine clinical practice, we cannot exclude distinct themes arising within patients that agreed to contribute to the focus groups. Those that participated may place a different value on the importance and subsequent assessment of QoL. However, the main discussion points raised in this research were similar in both of the randomised treatment arms, which underwent separate focus group meetings. Our findings would not necessarily apply to AF patients of other ethnic backgrounds, as there are known differences in the presentation and outcomes of AF amongst different racial groups [25].

Assessing and measuring improvement in QoL and symptoms is fundamental to better management of patients with permanent AF, who suffer from a substantial reduction in their physical wellbeing. The impact of comorbidities is poorly appreciated, with considerable variability in QoL requiring both generic and AF-specific assessment. Broader education is required beyond what is typically given around the prevention of stroke. More widespread use of questionnaires and peer support could have benefits in patients with AF seen in routine clinical practice, enabling care to be tailored to their individual needs.

We would like to thank other members of the RATE-AF team, in addition to the independent members of the trial oversight committees and Jonathan Mathers (Qualitative and Mixed Methods Applied Research, University of Birmingham). We are indebted to the patients and their families who dedicated their time to take part in this NHS research.

This study was approved by the East Midlands – Derby Research Ethics Committee (16/EM/0178) and the UK Health Research Authority (IRAS project ID: 191437). Audio recordings of the meetings were made with the consent of all participants.

None of the authors report any conflicts of interest. All authors have completed the ICMJE uniform disclosure form (www.icmje.org/coi_disclosure.pdf) and declare:

D.K. reports grants from the National Institute of Health Research (NIHR CDF-2015-08-074 and NIHR HTA-130280), the British Heart Foundation (PG/17/55/33087 and AA/18/2/34218), EU/EFPIA Innovative Medicines Initiative (BigData@Heart 116074)and IRCCS San Raffaele/Menarini and Amomed (Beta-blockers in Heart Failure Collaborative Group NCT0083244); in addition to personal fees from Bayer (Advisory Board), AtriCure (Speaker fees), Amomed (Advisory Board), and Myokardia (Advisory Board), all outside the submitted work. M.J.C. receives funding from the NIHR Birmingham Biomedical Research Centre, the NIHR Surgical Reconstruction and Microbiology Research Centre and NIHR ARC West Midlands, Innovate UK (part of UK Research and Innovation), Macmillan Cancer Support, and UCB Pharma; personal fees from Astellas, Takeda, Merck, Daiichi Sankyo, Glaukos, GSK, and the Patient-Centered Outcomes Research Institute (PCORI) outside the submitted work. A.J.C. reports that he has worked with companies that have and are developing anticoagulant drugs, antiarrhythmic therapies, and technology for the management of atrial fibrillation. J.J., M.S., S.H., K.V.B., and T.L. have nothing additional to declare.

The RATE-AF trial and this focus group work was funded by the National Institute for Health Research (NIHR) as part of a Career Development Fellowship to D.K. (CDF-2015-08-074). The Patient and Public Involvement in the design and management of the trial was supported by a grant from the West Midlands NIHR Clinical Research Network. The work is also supported by a British Heart Foundation (BHF) Accelerator Award to the University of Birmingham Institute of Cardiovascular Sciences (AA/18/2/34218). The opinions expressed in this paper are those of the authors and do not represent the BHF, NIHR, or the UK Department of Health and Social Care.

The Patient and Public Involvement team led the focus groups and all data extraction and thematic analysis (J.J. and M.S.). The concept of the focus groups was originally developed by D.K. (Chief Investigator for the RATE-AF trial) and M.J.C. (expert in patient-reported outcomes). The manuscript was drafted by J.J., M.S., K.V.B. and D.K., and all authors contributed to revision and editing for intellectual content.

Jacqueline Jones and Mary Stanbury are joint first authors.