Article Text
PDF
Abstract
Objectives Pulmonary rehabilitation (PR) is a highly effective, recommended intervention for patients with chronic obstructive pulmonary disease (COPD). Using behavioural theory within mixed-methods research to understand why referral remains low enables the development of targeted interventions in order to improve future PR referral.
Design A multiphase sequential mixed-methods study.
Setting United Kingdom (UK).
Participants 252 multiprofessional primary healthcare practitioners (PHCPs).
Measures Phase 1: semistructured interviews. Phase 2: a 54-item paper and online questionnaire, based on the Theoretical Domains Framework (TDF). Content and descriptive analysis utilised. Data mixed at two points: instrument design and interpretation.
Results 19 PHCPs took part in interviews and 233 responded to the survey. Integrated results revealed that PHCPs with a post qualifying respiratory qualification (154/241; 63.9%) referred more frequently (91/154; 59.1%) than those without (28/87; 32.2%). There were more barriers than enablers for referral in all 13 TDF domains. Key barriers included: infrequent engagement from PR provider to referrer, concern around patient’s physical ability and access to PR (particularly for those in work), assumed poor patient motivation, no clear practice referrer and few referral opportunities. These mapped to domains: belief about capabilities, social influences, environment, optimism, skills and social and professional role. Enablers to referral were observed in knowledge, social influences memory and environment domains. Many PHCPs believed in the physical and psychological value of PR. Helpful enablers were out-of-practice support from respiratory interested colleagues, dedicated referral time (annual review) and on-screen referral prompts.
Conclusions Referral to PR is complex. Barriers outweighed enablers. Aligning these findings to behaviour change techniques will identify interventions to overcome barriers and strengthen enablers, thereby increasing referral of patients with COPD to PR.
- chronic obstructive pulmonary disease (COPD)
- pulmonary rehabilitation (PR)
- primary care
- theoretical domains framework (TDF). mixed methods research
Data availability statement
The datasets during and/or analysed during the current study available from the corresponding author on reasonable request. All data requests should be sent to R.E.Jordan@bham.ac.uk.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
- chronic obstructive pulmonary disease (COPD)
- pulmonary rehabilitation (PR)
- primary care
- theoretical domains framework (TDF). mixed methods research
Strengths and limitations of this study
This is the first mixed-methods study to use the Theoretical Domains Framework to identify barriers and enablers to pulmonary rehabilitation referral from a primary healthcare practitioner perspective.
The utilisation and combination of two differing research paradigms in this exploratory sequential approach offers novel and detailed insights through combined research lenses which encompass multiple perspectives.
Many geographical regions across the UK are represented and include a diverse range of primary healthcare practitioners.
A combination of participant recruitment approaches have been used to reduce potential sample and selection biases.
Generalisability of the overall findings are limited by the inability to calculate distribution and therefore response rates.
Background
Pulmonary rehabilitation (PR) is a low cost, high value, internationally recommended intervention for patients with chronic obstructive pulmonary disease (COPD) which is effective in improving exercise capacity, reducing the impact of symptoms and improving prognosis.1 2 It is a structured multidisciplinary intervention combining individualised exercise with disease-related education.3 Despite the clear evidence of its effectiveness, the proportion of patients with COPD receiving PR is persistently low worldwide.4 5 Our previously published inductive qualitative paper presented the experiences of primary healthcare practitioners (PHCPs) as key potential referrers to PR.6 We found that there was a generalised awareness of PR, but little detailed knowledge of either the programme or the clinical benefits. Relationships with PR providers were limited, but considered important. Patient characteristics, rather than clinical need, influenced referral offers and referrers frequently believed patients to be poorly motivated. PR was most commonly offered during times of disease stability (usually at COPD annual review) and ease of the referral process and financial incentives positively influenced referral. In summary, referrers reported many barriers but few enablers, which collectively resulted in infrequent discussions about PR and associated referrals.
However, in order to aid the development of appropriate interventions to improve referral rates it is important to establish the generalisability and relative importance of these findings within a broader population of PHCPs. Furthermore, applying theory to identify the psychological and structural drivers that influence behaviour7 8 may offer new insights to shape interventions.9
The Theoretical Domains Framework (TDF) is a well-recognised approach which was derived from a synthesis of behaviour change theories,10 and examines the processes that influence behaviour.11 When applied, it offers explanations for behaviours, highlighting reasons that may inhibit or promote12 13 implementation of practice-based change.12
Using mixed methods and applying the TDF, we sought to assess and explain the reasons for low PR referral by PHCPs for patients with COPD. The aim of our multiphase design was to inform the development of theory informed interventions to improve PR referral rates from primary care in future.
Methods
We used a multiphase sequential design defined by two separate phases (figure 1). The cognitive and practical experiences of PHCP when considering and undertaking referral for patients with COPD were initially explored using a deductive approach by applying the TDF to data from our previously collected qualitative interviews. These findings informed a second quantitative phase, where we tested themes for generalisability using a nationwide survey of PHCP, to highlight the most relevant factors influencing referral.14–16
Multiphase sequential research design.
Both datasets retained independent value and meaning, but were connected at two time points: (1) where the qualitative data were used to construct the questionnaire and (2) where phase 1 and 2 results were integrated to inform interpretation. The multiphase sequential mixed-methods design therefore achieves both methodological and content integration.15 16
Patient and public involvement
There has been no public and/or patient involvement in this study.
Phase 1: application of TDF to qualitative interview data
We reanalysed data from our previously published inductive qualitative study6 in which 19 PHCPs from two differing geographical regions across Central and East of England were recruited and interviewed to thematic saturation using a predesigned topic guide. A deductive approach using content analysis17 was used for re-analysis of the data in order to align the results to the TDF and to offer new insights.
The interview topic guide (online supplemental additional file 1) was mapped to the Capability Opportunity Motivation-Behaviour model (COM-B), a model that highlights three critical prerequisites for behaviour change.18 This model was adopted rather than the TDF to guide interviews primarily because of the practical need to reduce interview length without compromising its aim. COM-B is very closely aligned to the TDF and has been utilised as a topic guide and mapped to the TDF in a similar healthcare professional study.19
Supplemental material
Analysis
All interview transcripts were managed using NVivo V.12. Barriers and enablers emerging from the interviews via content analysis were mapped to the relevant TDF domain, initially using construct labelling10 20 (online supplemental additional file 2). Utterances were coded once to the key TDF construct which then determined TDF domain alignment. JSW undertook the initial coding, then five transcripts were randomly allocated and distributed throughout the team (REJ, PA and SG) and independent TDF coding occurred, followed by frequent collaborative team discussion to ensure agreement with the coding. Queries were discussed with a behavioural expert (IV).
Supplemental material
Phase 2: quantitative methodology
Study design: cross-sectional survey
PHCPs were recruited via two main methods. Initially an invitation was included in a fortnightly newsletter emailed to members of the Primary Care Respiratory Society (PCRS). The survey was additionally distributed and shared by PCRS via their organisational Twitter and Facebook accounts. Social media distribution of the survey was further increased by individual and other organisational sharing, including the Facebook accounts of Advanced Practice UK and General Practice Nurse UK. A link for questionnaire completion was provided to the platform ‘Online Survey’.21 This was open between April and December 2019. To increase participation, responders were invited to opt in to a prize draw to win an I-pad.
Simultaneously, paper versions of the questionnaire were distributed at six UK conferences between March and November 2019 to attending PHCPs (predominately by hand by JSW, and using ‘in-conference bag’ distribution at one event). On self-completion, questionnaires were placed by participants in a locked ballot box and an optional token of appreciation was offered. Paper questionnaires were manually entered onto ‘Online survey’ by JSW.
As this was exploratory research, no a priori sample size calculations were performed. A pragmatic approach to study closure was adopted, this being online availability for a period of 8 months, distribution of the questionnaire at several appropriate PHCP targeted events, and that a reasonable range of PHCP had responded.
Methodology: instrument design
The cross-sectional survey (online supplemental additional file 3), collected (1) individual sociodemographic data, (2) current referral experiences, using TDF-based Likert scale questions (n=54) and (3) any new or complementary issues which may not have been previously mentioned, using an optional open question.22
Supplemental material
Socio-demographic data
These included questions on geographical location of practice, job title, postqualifying respiratory education and estimated frequency of PR referrals, using questions with pre-specified options.
Psychometric data
Barriers and enablers for PR referral identified from the phase 1 qualitative findings were converted into belief statements,20 including some that sought to test direct understanding. All questions were generated and aligned to the TDF by the coder (JSW) and validated by other team coders (REJ), including a TDF expert (IV). Fifty-four closed, fully labelled 5-point Likert scale questions/belief statements were included with responses ranging from ‘strongly disagree’ to ‘strongly agree’ and a midpoint rating. Some statements were reversed as an opposite belief to that frequently reported in the phase 1 data. These design elements were purposely selected to improve reliability and validity.23
The final survey mapped the 54 belief statements and open question section to 12 out of 14 theoretical domains (‘emotion’ and ‘behavioural regulation’ was excluded, given its low mapping in phase 1 results). Two rounds of survey piloting were undertaken with five practice nurses and the questionnaire refined to ensure question clarity and clearer completion instructions.
Analysis
All data were exported into an Excel V.16spreadsheet and STATA V.16 used to conduct simple descriptive statistics (frequencies and percentages), dichotomising into Agree/Strongly Agree versus the remaining options. Free text that directly related to barriers and enablers of referral practice was content-mapped to the TDF and thematic analysis applied.24
Results: phase 2
Response rates
Paper surveys (>1100) were distributed across six UK primary care focused events which were attended by a variety of PHCPs. A total of 154 (~14%) were returned and 134 (83%)/154 completed the survey sufficiently and were included. Online, it is unknown how many potential practitioners read the survey invitation, therefore participation rates could not be calculated. One hundred and twenty three participants started the online survey, but only 99 (80.5%) completed it and were included in the analysis.
Full details of the paper survey distribution and return rates can be found in online supplemental additional file 1.
Description of participants
Table 1 presents the sociodemographic characteristics for participants in the phase 2 quantitative (n=233) studies. Participants characteristics for phase 1 (qualitative) are available in the previously published paper.6
Baseline demographics of phase 2 participants
In contrast to the qualitative study where 6 (32%)/19 were GPs, the survey respondents were predominantly female nurses. Nurse respondents were similarly distributed across both conference and online groups (110/134, 82.1%; and 76/99, 76.9%, respectively) and responders from both sources had similar time working in practice. However, respondents recruited through conferences, compared with those who responded online, tended to be younger (28% <40 years of age), more likely to be practice nurses rather than other types of professionals, but were less likely to have respiratory qualifications, to see patients with COPD or to refer them to PR.
Referral to PR by type of healthcare professional
Overall, 109 (49.1%) reported being frequent referrers to PR, with GPs being less likely to refer and other professions including emergency care practitioners, nurse practitioners and advanced nurse practitioner (ANPs) more likely to refer. Referral was also higher among those with one or more continuous practice development (CPD) respiratory qualifications. However, this may be partly related to such qualification being higher among ANPs (82.5% (47/57)) and other grouped professions (58.8% (10/17)) than among GPs (17.9% (5/28)). More than 10 years spent in general practice appeared to marginally increase referral frequency (60.7%; 51.8%). Table 2 presents PHCP reported referral practice.
PHCP referral practice*
Forty (17.2%) of 233 responding PHCPs reported never referring to PR, with the largest group being practice nurses (29/40; 72.5%). Thirty-three of 40 PHCPs offered a variety of reasons for non-referral including; not considering it to be part of their role, not seeing patients with COPD or not knowing they could refer (12/33; 36.4%). Others reported it was undertaken by other respiratory specialist/interested healthcare professionals across primary and secondary care settings (12/33; 36.4%). Further reported reasons were unsure how to and/or a lack of training (5/33; 15.1%), uncertainty about local service provision (3/33; 9.1%) and 1/33 (3.0%) reported belief that patients were not interested.
Phase 1 results: TDF analysis of the qualitative interviews
Table 3 shows the referral behaviour of PHCPs mapped to all 14 TDF domains. The most frequently mapped domain was social and professional role (n=287 times) while the least mapped was behavioural regulation (n=4).
Phase 1: mapping of barriers and enablers for referral to TDF domains
Phase 2: questionnaire results—referral practice beliefs
Table 4 presents the number and proportion of PHCPs that agreed or strongly agreed with each belief statement by frequency of referral.
Results of TDF belief statements by referral frequency
In general, most PHCPs had some PR knowledge (especially the frequent referrers) and understood the beneficial consequences of PR. However, resources, social influences (such as relationship with PR providers) and pessimism about patient motivations were perceived barriers by a high proportion of PHCPs, irrespective of their referral practice.
There were however, differences in domains between frequent and infrequent PR referrers.
The greatest differences were within the ‘Knowledge’ domain. Frequent referrers most commonly reported agreement with all seven statements, when compared with the infrequent referrers. For example, 97.3% reported knowing when to refer to PR and 80.7% being able to answer patients’ questions versus 65.5% and 53.1% of infrequent referrers.
Further group differences were demonstrated in the ‘Skills’ domain and ‘Beliefs about (PHCP) capabilities’, which showed that infrequent referrers were less confident in encouraging unmotivated patients to attend PR (67.6% vs 83.5% of frequent referrers). Reduced confidence among infrequent referrers was further reflected within the ‘Optimism’ domain and belief statement ‘I am confident my local provider offers a good service’ (46% against 74.3% of frequent referrers). However, over half (56.9%) of frequent referrers felt that patients in work were not able to attend PR, compared with less than a third (31%) of those who referred infrequently.
The remaining belief statements demonstrated greater group similarities than differences.
Environment, social and professional role
Most respondents felt that there was enough time in practice to refer (84.7%) and believed in encouraging PR attendance (96.4%). Yet promotional information on PR was rarely available in practices (29%). There was no clearly identified PR referrer; less than half (48.6%) felt it was the practice nurse’s role and (51.8%) reported other practice staff refer.
Social influences
Frequent referrers were slightly more likely to agree with three of the four domain belief statements than infrequent referrers. Although, collectively the groups reported both PR provider engagement and referral outcome reporting as low at only 22.6% and 29%, respectively. PHCPs also reported patients rarely request referral to PR (5%).
Belief about consequences and optimism
Most PHCPs agreed that PR offers physical health benefits, including improving breathlessness and reducing hospital admissions (91.9%, 89.6%), respectively. Yet far fewer PHCPs believed patients would attend and complete PR (46.2%), with fewer still agreeing that patients are PR motivated (24.2%).
Memory (decision-making)
Only a small number of PHCPs reported forgetting to refer patients to PR (11.7%). COPD annual review templates were reported as helpful referral reminders (63.8%) and 25.8% reported the best time to discuss referral with patients was during COPD stability. Patient characteristics such as disease stability and smoking status do not appear to impede PHCP referral decisions as 98.2% reported referring smokers.
Goals, reinforcement and intention
In-practice review of eligible patients was not commonly reported (41%) and only 19.8% reported in-practice targets to improve referral rates. Practice financial reward for referral (pre April 2019) was rarely reported (5%); indeed the implementation of financial reward via national QoF incentives (post April 2019) was considered unlikely to greatly improve referral behaviours, with less than a third (32.6%) stating they would refer more. However, there was general agreement that this incentive would increase practice awareness of PR (70.1%).
Phase 2: questionnaire—open questions
A third of PHCPs (33.8%) responded to the open question at the end of the survey including 5/11 PHCPs who reported referral, but did not specify frequency (answer length 3–167 words, mean 35). Non-frequent referrers reported more open comments (43/113, 38.1%) than frequent referrers (33/109, 30.3%).
This gave an additional 94 comments that related directly to PR referral. These were content mapped to all 12 relevant TDF domains. The comments predominately cited referral barriers.
Belief about capabilities had the highest number of comments 36/94 (38.3%) with many encompassing concerns about PR accessibility, particularly transport challenges for patients. For example, ‘Location of PR too far for patients to travel and too much commitment. Patients tend to be older adults on generally low incomes. A number of my patients would attend if it was close by with no expense’. A small number of PHCPs (3.2%) considered a patient’s inability to complete pre-PR spirometry as a referral barrier, and 10.6% of comments related to referral processes, which were reported to be lengthy and as such ‘easier simpler’ processes were requested.
Connected results
In order to identify the key factors that inhibit and/or enable PHCP referral to PR, Phase 1 and phase 2 results were merged to allow for data contrast and meta-inference16 (table 5).
Matrix of integrated results
Most PHCPs believed in PR and encouraged patients to attend. Referral is most likely to be considered at annual review (indeed referral is rarely offered to patients outside of this consultation). On-screen prompts are helpful reminders, but in practice material promoting PR is rare. PHCP PR knowledge is largely gained from networking with other respiratory interested health professionals and/or CPD education. PHCPs report patients have little motivation for PR, rarely ask for referral to PR and view that patients in work are unlikely to be able to attend.
Some findings of the qualitative study were not clearly replicated in the survey results. For example, phase one qualitative data highlighted that some GPs and ANPs felt the practice nurse was best placed to undertake PR referral at the time of annual review, yet respiratory interested GPs and those undertaking annual review did not share this view. The phase 2 survey data supported the latter position, where 29 (22.5%)/129 of practice nurses reported never referring. Therefore responsibility of PR referral is not based on profession, but is undertaken by PHCPs who are respiratory interested and/or conducting the patient’s annual review.
Qualitative generalisable findings were limited in a number of areas meaning clear conclusion cannot be drawn, these included; time available to undertake referral, ease of referral process, perceptions of quality of PR programme, referral of patients when COPD symptom burden is increasing and non-referral in order to protect patient relationship.
Where generalisability is clear, a summary of the key behavioural barriers and enablers by TDF domain is shown in table 5, demonstrating a greater number of barriers than enablers to referral. However, it is also important to report that barriers and enablers most commonly coexist within the same domains.
Discussion
This is the first time the TDF has been applied to a mixed-methods study to understand the key factors that determine referral to PR by PHCPs.
Results highlighted multiple intertwined barriers and few enablers across all TDF domains. Many (although not all) of the findings from the qualitative study were affirmed by the more generalisable survey and highlight that referral to PR from primary care remains poor, and that PHCPs believed that PR was beneficial for patients and wanted to refer more. They did however, request greater engagement from providers, better knowledge of local programmes and improvements in PR promotion. They also reported that in-practice goals and monitoring of referrals to address the shortfall in patients referred were rare.
However, PHCPs collectively reported low confidence in patients’ abilities and motivations to attend PR, a belief likely to be strengthened by reports of few patients requesting referral. Beliefs about low uptake may explain why referral is commonly offered at times of increasing COPD symptoms, thus acting as a lever to referral acceptance. Infrequent referrers reported reduced confidence in encouraging unmotivated patients to attend, with similar findings reported in phase 1 data as PHCPs expressed concerns around the protection of relationships with patients. Venue accessibility also appears to be a barrier and while the direct survey question (question 21) appeared not to overtly agree with this, both phase 1 and the phase 2 open question results highlighted transport as both a practical and financial barrier.
Variability in referral rate by PHCP profession was an unexpected finding and offers insights that (1) few PNs refer and (2) where it is considered to be the ‘respiratory nurse’ role, referral opportunities may become reduced. The association between referral frequency and respiratory qualification is also a new finding. ANPs were those most likely to refer and to have respiratory qualifications.
Relation to other studies
This mixed-methods TDF-based study finds agreement with many key referral factors presented in our previous inductive qualitative study using the same data6 and Cox et al’s25 TDF-applied systematic review which included patients and HCPs views on PR barriers and enablers. However, this primary mixed-methods study reports additional findings. It disputes that the PN is the main referrer to PR within primary care, and questions the value of practice-based financial reward as a referral incentive. It also highlights that the referral process itself is not straightforward and there are no sanctions for non-referral, but that there is time in practice to refer.
Increasing the population sample and geographical reach in this study strengthens current known referral barriers including, poor patient motivation, few in-practice resources, perceived venue access difficulties and little awareness of local PR provision.26–29 Subjective patient assessments including PHCPs perceptions of patients capabilities and motivations have been described as influencing PHCP referral decisions here and previously published.6 This is a novel finding in relation to PR referral, yet similar HCP pessimistic attitudes, relating to a patient’s capability and motivation to access services and change behaviours to improve health outcomes have been reported in the primary healthcare management of reducing cardiovascular disease risk in people with serious mental illness.30 31
Phase 1 and inductive data analysis6 suggested that offering PR at COPD symptom increase was common yet this was unconfirmed in the survey results. This may demonstrate further social desirability reporting as previous analyses have demonstrated patients attending PR to have 1.24 hospitalisations per patient-year (95% CI: 0.66 to 2.34) suggesting sicker patients are those most likely to be offered PR.32 However, referral at this time supports both PHCP and patients’ concerns about patient’s capabilities,6 25 33 meaning lower acceptance and adherence to PR is probable, and negative PHCP beliefs about referral outcomes are likely to perpetuate. An alternative approach and one that appears not to be currently undertaken is to refer at the point of an acute exacerbation of COPD, which maybe a referral lever.33
In our original inductive analysis,6 we reported that financial incentives may be important, yet results in this current study are mixed and PHCPs appear uncertain of their value. It will be interesting to observe the impact of the newly implemented financial rewards for PR referral in England, but where similar practice based quality outcome framework (QoF) rewards were implemented for referral to diabetes programmes, uptake did not greatly improve.34 Given positive correlations between referral rates and CPD education, efforts to increase the number and education of the primary care workforce by Health Education England35 36 is encouraging.
The literature also supports a general consensus that for patients in employment, PR is largely considered inaccessible.6 28 This was reported as a barrier by the frequent referrers more than the infrequent referrers, which questions whether PR knowledge itself is a potential barrier as previously reported6 and that PHCP beliefs influence subsequent referral behaviours.
Strengths and limitations
Using the previously published qualitative data to inform the questionnaire offered valuable insights into PHCP referral practices and is a key strength of this research.
The range and number of PHCPs included from across the UK were broadly representative of the general practice nursing workforce, while less so for others, notably doctors and is a limitation.37 We recognise that predominately respiratory interested participants may have taken part in this study which may skew results, and it is noted that online participants reported higher referral practice and respiratory qualification(s) than their counterparts, which may be a study limitation, suggesting that more emphasis should be placed on the perspective of the infrequent referrers. Adopting additional recruitment strategies such as via general practice-based conferences is seen as a study strength which sought to capture a range of PHCPs’ views. Demographic similarities across all three recruitment streams highlight study design attempts to reduce participation and sample selection biases. Questionnaire-specific biases relating to self-reporting response is a source of potential weakness, specifically where responses maybe perceived to be ‘socially acceptable’, otherwise known as social desirability.38 This may offer some explanation around the variation observed in the belief about capabilities domain of the integrated results matrix (table 5). Grouping participants by reported referral frequency is a study strength, particularly as the aim is to understand both what supports and inhibits referral. Another limitation is that we are not sure about exact response rates where distribution was unknown.
Much of the validity of the TDF is gained from its direct application with HCPs, as utilised here. Transcript content mapping to 84 constructs is complex and time consuming as also described by others39 but was considered the most comprehensive approach in the absence of a gold standard approach to TDF application.39 The TDF offers a functional approach to behavioural data analysis, most likely to be helpful when there is little to no underlying knowledge of the investigating phenomenon. However, the inter-relations between referrer, patient and provider have previously been reported to be important factors in the referral journey.6 Yet, the TDF does not offer causal determinants of behaviour20 and alignment to predetermined domains reduces the ability to consider any phenomena falling outside those domains and the likely connecting relations, meaning the whole picture maybe missed and is a potential limitation.
All authors had different professional backgrounds, one of whom (JSW) is an experienced respiratory nurse specialist which may have altered data analysis although transparency and frequent team analysis sought to reduce potential bias.
Implications for policy and practice
While this paper highlights multiple barriers in referring patients with COPD to PR, barriers to high-quality healthcare for patients with COPD span throughout the disease trajectory and persist across health service provisions worldwide.40–42 It is interesting to note that few participants in our study thought that a financial incentive was important. It is however difficult to assess this given that face-to-face PR programmes were suspended across the country as a result of the COVID-19 pandemic. However, as previously highlighted QoF incentives for referral to diabetes programmes did not greatly improve uptake. What we need to do now is to design and test an intervention for improving referral to PR which incorporates multisystem level changes. Additional intervention considerations will also need to include post COVID-19 infection control adaptations, as well as managing increases in service demands arising from programme suspension backlogs and new referrals, including COVID-19 survivors.43
Conclusions
This is the first mixed-methods research study to examine the factors that inhibit and enable referral to PR for patients with COPD from a primary care perspective. While knowledge and respiratory qualification appear to be enablers, many barriers persist which must be overcome to increase referral opportunities for all eligible patients. The most important aspects to address are to increase PR provider engagement with referrers, increase PR awareness and support for potential patients and all PHCPs, including those with respiratory qualifications and to increase PHCP internal motivation for PR referral, particularly for those patients in work and those with less symptom burden. Mapping these TDF findings to behaviour change techniques are important next steps which will enable clear targeted interventions to be identified and tested in clinical practice, which will ultimately increase referral to PR, thereby improving health outcomes of patients with COPD and reducing health service utilisation.
Data availability statement
The datasets during and/or analysed during the current study available from the corresponding author on reasonable request. All data requests should be sent to R.E.Jordan@bham.ac.uk.
Ethics statements
Patient consent for publication
Ethics approval
Phase 1 approval granted by Health Research Authority: Project ID: 213 367. Phase 2 approval granted by University of Birmingham: ERN_19–0439. All participants in phase 1 and phase 2 studies gave consent.
Acknowledgments
The authors thank all participating primary healthcare practitioners for giving up their time, providing the data and contributing to this study.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @Nurse_JSW
Contributors JSW collected, analysed and interpreted phase 1 and phase 2 data and was a major contributor in writing the manuscript. REJ, PA, SG and AE contributed to study design, data analysis and interpretation of phase 1 and 2 data. RJ, PA and SG all contributed to the writing of the manuscript. IV supported phase 1 topic guide development, phase 1 data alignment to the TDF and the formulation of the phase 2 questionnaire where behavioural expert consensus was sought. All authors read and approved the final manuscript. REJ accepts responsibility for this work and is the guarantor.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.