C-reactive protein-guided antibiotic prescribing for COPD exacerbations: a qualitative evaluation

Rhiannon Phillips; Helen Stanton; Amina Singh-Mehta; David Gillespie; Janine Bates; Micaela Gal; Emma Thomas-Jones; Rachel Lowe; Kerenza Hood; Carl Llor; Hasse Melbye; Jochen Cals; Patrick White; Christopher Butler; Nick Francis

doi:10.3399/bjgp20X709865

Abstract

Background Antibiotics are prescribed to >70% of patients presenting in primary care with an acute exacerbation of chronic obstructive pulmonary disease (AECOPD). The PACE randomised controlled trial found that a C-reactive protein point-of-care test (CRP-POCT) management strategy for AECOPD in primary care resulted in a 20% reduction in patient-reported antibiotic consumption over 4 weeks.

Aim To understand perceptions of the value of CRP-POCT for guiding antibiotic prescribing for AECOPD; explore possible mechanisms, mediators, and pathways to effects; and identify potential barriers and facilitators to implementation from the perspectives of patients and clinicians.

Design and setting Qualitative process evaluation in UK general practices.

Method Semi-structured telephone interviews with 20 patients presenting with an AECOPD and 20 primary care staff, purposively sampled from the PACE study. Interviews were audio-recorded, transcribed, and analysed using framework analysis.

Results Patients and clinicians felt that CRP-POCT was useful in guiding clinicians’ antibiotic prescribing decisions for AECOPD, and were positive about introduction of the test in routine care. The CRP-POCT enhanced clinician confidence in antibiotic prescribing decisions, reduced decisional ambiguity, and facilitated communication with patients. Some clinicians thought the CRP-POCT should be routinely used in consultations for AECOPD; others favoured use only when there was decisional uncertainty. CRP-POCT cartridge preparation time and cost were potential barriers to implementation.

Conclusion CRP-POCT-guided antibiotic prescribing for AECOPD had high acceptability, but commissioning arrangements and further simplification of the CRP-POCT need attention to facilitate implementation in routine practice.

INTRODUCTION

A multinational study found that 79% of patients presenting in primary care with an acute exacerbation of chronic obstructive pulmonary disease (AECOPD) were prescribed an antibiotic, ranging from 49% of patients in Denmark to 93% of patients in Russia.¹ However, antibiotics are unlikely to provide meaningful benefit for most outpatients with AECOPD.² Unnecessary use of antibiotics increases the threat to society from antibiotic resistance, and increases individual patients’ risk of side effects and carriage of resistant organisms in their lungs that may in turn increase risk of subsequent exacerbations and disease progression.³^–⁶

Current guidelines recommend using symptoms as the main guide to making antibiotic prescribing decisions for AECOPD.⁷^,⁸ C-reactive protein (CRP) is an acute-phase reactant and raised levels are associated with AECOPD.¹^,⁹ The PACE study randomised 653 participants at 86 general practices in the UK to establish whether CRP point-of-care testing (CRP-POCT) could safely reduce antibiotic consumption for AECOPD.¹⁰ The PACE study found that use of CRP-POCT and guidance on interpretation of the test in primary care led to a 20% reduction in the consumption of antibiotics over the 4 weeks following consultation for an AECOPD, without compromising recovery at 2 weeks post-consultation compared with usual care.¹¹

Process evaluations are an important part of evaluating complex interventions, providing a better understanding of the implementation and receipt of interventions, and the context in which they were delivered.¹²^,¹³ This qualitative process evaluation was embedded within the PACE study¹⁰^,¹¹ to better understand the acceptability of CRP-POCT-guided antibiotic prescribing for AECOPD; explore possible mechanisms, mediators, and pathways through which the intervention had the effects observed in the quantitative component of the study;¹¹ and identify potential barriers and facilitators to implementation from the perspectives of patients and clinicians.

The PACE randomised controlled trial found that a C-reactive protein point-of care test (CRP-POCT) management strategy resulted in a 20% reduction in patient-reported antibiotic consumption over 4 weeks following consultations for acute exacerbation of chronic obstructive pulmonary disease (AECOPD) in primary care. Understanding the perceived value of CRP-POCT to clinicians and patients, potential mechanisms, and identifying barriers and facilitators to its use is vital in informing implementation plans. This study indicated that the CRP-POCT had high acceptability for use in the management of AECOPD in general practice, increasing clinician confidence, reducing decisional uncertainty, and as a tool to facilitate communication and patient education. GPs should consider adopting CRP-POCT in the routine management of acute exacerbations of COPD, but commissioning arrangements and further simplification of the point-of-care test need attention to facilitate this.

How this fits in

METHOD

Setting and participants

A purposeful sampling method using predefined criteria¹⁴^,¹⁵ was used to select 20 patients in the CRP-POCT trial arm of the PACE study, and 20 members of primary care teams who had carried out the CRP-POCT with patients or had used the CRP-POCT result during patient consultations to guide their prescribing decision (that is, where the CRP-POCT had been carried out by another member of the primary care team). The sampling framework was designed to ensure representation from patients and primary care teams in each of the regions where the PACE study centres were located (Wales, Oxford, London, and Norfolk), and from both patients who had been prescribed antibiotics at their initial consultation and patients who had not. Patients who had provided consent to be contacted about an interview at the beginning of the study were contacted by telephone.

As part of the PACE study, primary care staff were provided with brief training in use of the CRP-POCT, and guidance on interpretation and use of the CRP-POCT test result. The CRP-POCT was used in addition to usual best practice, with the test result being used to guide (but not to mandate) antibiotic prescribing. The guidance stated that people with a CRP-POCT reading of <20 mg/L would probably not benefit from antibiotics, people between 20 mg/L and 40 mg/L may benefit from antibiotics, and that antibiotics are likely to be beneficial for those with a reading of >40 mg/L. Full details are provided in the trial protocol.¹⁰

The PACE study adopted a pragmatic approach to implementation of the CRP-POCT, allowing primary care practices to arrange for the test to be conducted in a way that fitted with their own structure and processes.¹⁰ This meant that staff other than the doctor, including nurses, healthcare assistants, and research assistants with appropriate training, sometimes undertook the CRP-POCT testing. Some of these individuals were included in the interviews to see how the test was used in these different contexts.

Written informed consent was obtained from all patients. Primary care staff provided verbal consent, which was audio-recorded. Patient participants were sent a 10 GBP gift voucher after their interview as a gesture of appreciation for their time. Primary care staff were not provided with incentives for participation in qualitative interviews, as they completed these as part of their wider involvement in the PACE study research activities.

Approach

Flexible topic guides were used to guide interviews. The topic guides were piloted with patients (n = 10) and primary care staff (n = 9), and refined prior to use in this study. Interviews were audio-recorded and field notes were made. Transcripts were not returned to participants for comment and no repeat interviews were carried out. Interviews were carried out by experienced qualitative interviewers employed on the PACE study. Their role as non-clinical researchers was explained to interview participants. Interviewers had no prior relationship with participants. Researchers were provided with study-specific training and supervision by an experienced qualitative researcher.

Analysis

Interviews were transcribed verbatim. NVivo (version 11) qualitative analysis software was used to assist coding and facilitate analysis. Data were analysed using framework analysis: a systematic approach to a thematic analysis that allows for easy comparisons between and within cases, facilitates sharing and discussion of data, and allows for clear linking of developed themes to original data.¹⁶^–¹⁸ The data were analysed using a hybrid inductive and deductive approach, based primarily on social phenomenology.¹⁹ The framework analysis (familiarisation, development of framework, and charting) took place before the trial outcomes were known, in line with the Medical Research Council guidance on process evaluation.¹² The protocol did not include dual coding of the data. Instead, regular qualitative research team meetings were held with the Trial Management Group at key junctures in the analysis to discuss data production, the development of the coding framework, and data analysis. This approach has been identified as appropriate for qualitative research.²⁰ The definition of data saturation used in this study was the point at which the ability to obtain additional new information had been attained, and when further coding was not feasible.²¹ The qualitative researchers assessed whether the last five interviews with primary care staff and patients provided new information that would add to the thematic framework being developed. On this basis, the judgement was made that data saturation had been achieved.

RESULTS

Interview participant characteristics are shown in Table 1. Semi-structured interviews were conducted with 20 clinicians and other primary care staff who undertook CRP-POCT testing across 19 practices. Of the 47 primary care practices that were invited to participate, two declined, two were unable to take part as they had not randomised any participants to the CRP-POCT arm, and 24 did not respond. Antibiotic prescribing rates were similar in practices that did (56.1%) and did not (60.5%) take part in the qualitative interviews (Supplementary Table S1).

View this table:

Table 1.

Characteristics of qualitative process evaluation participants

Interviews were carried out with 20 patients across four regions (Wales, Oxford, London, and Norfolk). Of the 40 patients invited to take part in an interview, 16 declined, one was in hospital when telephoned, and two were interested but unable to arrange a suitable time for an interview. A participant from the control arm of the trial was recruited for the interviews erroneously and their data were not included in this analysis. One-to-one interviews were conducted between October 2015 and March 2017. A practice manager briefly joined one discussion part of the way through an interview with a clinician. Patient interviews lasted between 15 and 35 minutes; primary care staff interviews lasted between 20 and 45 minutes.

Framework analysis

Key themes identified through the framework analysis related to perceptions of the value of the CRP-POCT, perceived mechanisms of impact of the CRP-POCT, and implementation of the CRP-POCT in routine practice. A summary of key findings is provided in Box 1.

Main theme	Subtheme	Patient views	Primary care staff views
Perception of the value of the CRP-POCT	General views of the CRP-POCT	Many felt that the CRP-POCT was a useful addition to the consultation that would help guide their doctor’s antibiotic prescribing decision.	Most thought the CRP-POCT was a useful addition to the consultation, particularly where there was diagnostic uncertainty. Clinicians emphasised the importance of using the CRP-POCT in addition to, not in place of, a thorough clinical assessment.

Perceived mechanisms of impact of the CRP-POCT	Objective sign of illness	Patients felt that the CRP-POCT provided an objective sign of illness severity that could help guide treatment.	Prescribers felt that having this additional piece of objective evidence increased their confidence in their antibiotic prescribing decisions.
	Enhancing patient–clinician communication	CRP-POCT is useful in understanding whether antibiotics are needed, but there are some misconceptions about when antibiotics might or might not be helpful (for example, for viral infections).	Primary care staff felt that the test provided an opportunity to open discussions with patients about antibiotic use and antimicrobial resistance.
	Reinforcing prescribers’ decisions	Patients were generally passive in terms of making decisions about antibiotic treatment, with clinicians explaining their decision to/not to prescribe antibiotics to them.	Primary care staff perceived the CRP-POCT result as being useful in reinforcing their decision about antibiotic prescribing when communicating with patients.

Implementation of the CRP-POCT	Views about implementation in routine practice	Many patients expressed positive attitudes towards the use of the CRP-POCT in routine NHS care for the management of AECOPD.	Positive attitudes towards the use of the CRP-POCT in routine NHS care, but there were differences of opinion about whether the CRP-POCT would be used for all patients with AECOPD, or only those where there was clinical uncertainty.
	Technical aspects of the test	Patients did not report any difficulties with the use of the CRP-POCT by clinicians.	Found the CRP-POCT easy to use, but felt that the need for test cartridges to be refrigerated during storage and returned to room temperature before use, need for regular calibration of the machine, and lack of portability of the device were potential barriers to widespread use in primary care.
	Time and resources	Patients felt that use of the test was quick and did not report any problems with the administration of the test.	Acknowledged the impact on consultation length that use of the CRP-POCT had, but felt that it was worthwhile. Felt that the cost of the CRP-POCT machine and cartridges was prohibitive under current funding arrangements.

Contextual factors	Non-medical factors that influenced prescribing	Patient attitudes with regard to antibiotic use for AECOPD were varied, but many did not want to take antibiotics for AECOPD unless they were required.	Patient anxiety, a strong patient preference for antibiotics, and individual circumstances (for example, recent death of a spouse) were cited by primary care staff as reasons for still prescribing antibiotics despite a low CRP-POCT result.

AECOPD = acute exacerbation of chronic obstructive pulmonary disease. CRP-POCT = C-reactive protein point-of-care test.

Box 1.

Summary of key themes extracted from the qualitative interviews

Perceptions of the value of the CRP-POCT

Although clinicians felt the CRP-POCT provided useful information, several felt this only affected their decision when there was uncertainty about whether antibiotics were needed. They emphasised the importance of using clinical findings to guide antibiotic prescribing decisions, and did not view the CRP-POCT as a replacement for clinical skills: ‘It’s shown that we’re not always right when we listen in, you know. There is a possibility that this may just be a viral crackle, as opposed to bacterial, but again it’s very difficult without the reassurance of the, the CRP, to let the patient go away.’(Nurse practitioner [NP]1)

Clinicians talked about the added value of the test, as demonstrated by this clinician who described a case where a CRP reading had been unexpectedly high: ‘I told my partner, who had seen this gentleman first this morning and I told him how high the CRP was. He was, he was as shocked as I was. Now it may be that this man has another reason for having a high CRP, you know, there may be something else going on other than infection and we’re going to follow that up. But, but I would say that it would be, you know, the point-of-care testing would be an excellent thing to have in the surgery, because it can, you know, it can give you some information which, which you would not have on a clinical examination.’(GP1)

Primary care staff felt that the CRP-POCT reassured patients, and that the test demonstrated to patients that a thorough examination had taken place: ‘They [patients] feel reassured that no antibiotics have been given and the doctor’s actually checked that this was not necessary before he said “no” to the antibiotics, rather than just saying “no you don’t need it ”.’(GP2)

Clinicians were aware of the need to reduce antibiotic prescribing, and felt that the perceived risk of under-treatment was a driver for prescribing unnecessary antibiotics for AECOPD: ‘There’s so much pressure not to refer patients to hospital, so if you, the view is, if you treat them early, you know, when their symptoms are relatively mild, maybe we’ll be able to stop someone going to hospital unnecessarily.’(GP3)

The perception that early prescribing can reduce hospitalisation is at odds with evidence from a Cochrane review,² which did not find evidence that antibiotic prescribing for AECOPD in outpatient settings has an effect on hospital admissions or mortality. A GP also raised the issue of fear of litigation, where the CRP-POCT was seen as providing objective evidence to help justify prescribing decisions: ‘I can only speak for myself, but every patient I see, when I’m writing down, I’m thinking that somebody’s going to be suing me as a result of it, which is very sad but it’s just the way the world’s going, and I think every GP is probably very similar, and I know that if I write down “CRP less than five” then anyone taking me to court over that is going to have one hell of a hard time of it to prove that that patient was ill at that point.’(GP5)

Patients felt the CRP-POCT could ‘help’ doctors with their decisions, and did not report any anxiety about having the test. Patients felt that the CRP-POCT was useful in rapidly deducing the severity of illness and/or need for antibiotics: ‘I think it’s a great idea to measure really sort of how ill you are and whether you really need more treatment or not.’(patient [P]1, female [F], CRP <20 mg/L, no antibiotics prescribed)

Perceived mechanisms of impact of the CRP-POCT

Three subthemes were identified relating to perceptions about how the use of the CRP-POCT might achieve the desired aim of safely reducing antibiotic use: the CRP-POCT provided an objective sign of illness severity; the CRP-POCT enhanced physician–patient communication; and use of the CRP-POCT reinforced prescribers’ decision.

The CRP-POCT provided an objective sign of illness severity

Prescribers reported that the CRP-POCT reading provided objective evidence to support clinical decision making and reduce decisional uncertainty: ‘I think the clinical decision was, was probably there anyway without needing the CRP test, but obviously there are some instances where, you know, if you’re not too sure, then obviously that CRP test could’ve maybe made that difference as to whether you gave the antibiotics or not.’(Non-prescriber 1)

Being able to share the reading with patients helped to provide objective evidence to provide support for treatment decisions when communicating with patients: ‘Because I think if it’s just you face-to-face and you have no objective evidence, it’s just your opinion and they sometimes question that.’(GP4)

Clinicians felt that the CRP-POCT enhanced their confidence and reassured both prescribers and patients about their decision with regard to antibiotic treatment: ‘I found writing down “CRP normal”, I found that that was a very powerful way of reassuring me and the patient actually, it seemed to place a great deal of, you know, faith on, on blood testing.’(GP5)

Many patients viewed the CRP-POCT as a useful way of objectively measuring the severity of their illness: ‘I thought it [CRP-POCT] was excellent because it was just proving what I already knew if you know what I mean.’(P2, F, CRP 20–40 mg/L, prescribed antibiotics)

However, one patient viewed the CRP-POCT negatively as they felt that the test result was not consistent with their subjective experience: ‘I wasn’t happy to be honest, because, simply because they said the test that was OK and [I had] an ever [so] slight inflammation which they took because of this blood test she found and she gave me five days of the steroids, but after the five days I was back to square one.’(P3, male [M], CRP <20 mg/L, no antibiotics prescribed)

The CRP-POCT enhanced physician–patient communication

Clinicians felt that patients had greater involvement in the consultation through discussion of the test outcome, and that it provided them with an opportunity to talk to patients about antibiotic stewardship: ‘It allows you to talk a little bit about antibiotics, you can then, you can, we can then add and refer people to an information sheet about the duration of common symptoms for example.’(GP3)

From the patient perspective, there was a reasonable level of understanding of the purpose of the CRP-POCT in terms of guiding doctors’ antibiotic prescribing decisions: ‘Yes, it was to see if I had an infection on my chest and the count of it was I think five, so they decided I didn’t have an infection but that the steroids would help me, which they did.’(P4, F, CRP <20 mg/L, no antibiotics prescribed)

Nonetheless, some patients were uncertain about what the CRP-POCT was testing, and there were some misconceptions about the type of infection that would require antibiotic treatment: ‘They need to confirm, which is what I thought this test and that was doing, that it is, it is a proper viral infection.’(P5, M, CRP 20–40 mg/L, prescribed antibiotics)

CRP-POCT reinforced prescribers’ decisions

The CRP-POCT reading was generally used by clinicians to articulate and justify their prescribing decisions: ‘It gives something to justify to the patient that it’s not just your clinical judgement on the signs and things. That you have actually done a test and that has, you know, given even more back-up that the fact that you confidently don’t need antibiotics.’(GP6)

Patients felt that their prescribers were, and should be, the decision makers with regard to antibiotic treatment: ‘Well I don’t think it comes under what the patients want, it’s the patient is ill enough to need antibiotics, you know then they should be given. Other than that I don’t think they should be given, if the patient isn’t ill enough for them.’(P6, F, CRP <20 mg/L, no antibiotics prescribed)

Patients who perceived being involved in decision making about their antibiotic prescription described this in terms of their agreeing with the doctor’s decision and having confidence in their expertise or because they felt that the doctors had explained their decision to them, rather than being actively involved in the decision-making process as such: ‘I would say my doctors give me sound advice about what to do, because at the end of the day I know they are very busy people and their range of knowledge is quite astounding, and at the end of the day I’m relying on him to give me the correct information to make an educated decision.’(P7, M, CRP <20 mg/L, prescribed antibiotics)

Implementation of the CRP-POCT in routine practice

Views about implementation in routine practice

Patients and primary care staff had a positive view about whether the CRP-POCT should be introduced into routine NHS care for patients with AECOPD: ‘I think it’s an important test and if we, it’s something I’d certainly want to explore in the future after the trial is finished, getting a CRP machine for the practice.’(GP7)

Primary care staff discussed the advantages of using the test in routine care mainly in terms of antibiotic stewardship and achieving more consistent prescribing decisions: ‘So I think it may help to standardise the treatments that we offer, I definitely think it’s a good idea, I think it’s something that we should be doing more of, because I think we probably would end up prescribing less antibiotics because of it.’(NP2)

Patients discussed the benefits of the test mainly in terms of reducing antibiotic use and saving money. From the patient’s perspective, their priority when they had an AECOPD was to resolve their symptoms. There were mixed feelings about when antibiotics should be prescribed. Mostly, patients recognised how valuable antibiotics were when they were needed, but did not want to take them if they were not required: ‘It’s not good taking antibiotics just for a minor complaint, you know, you should have it being really bad with your chest before taking antibiotics.’(P6, F, CRP <20 mg/L, no antibiotics prescribed)

Within this context, they were receptive to the use of the CRP-POCT in routine care: ‘I think they’re [GPs] doing their best, and I do think that the pinprick test is absolutely amazing and I should […] I would like it to be done as a regular thing if you get a flare up.’(P4, F, CRP <20 mg/L, not prescribed antibiotics)

Clinicians and other primary care staff had mixed views on how the test should be implemented. Some clinicians felt that using the CRP-POCT for all patients presenting with AECOPD to increase their data provided a learning tool to improve their ability to detect patients who need antibiotic treatment. Others felt they would only use the CRP-POCT when there was decisional uncertainty about the need for antibiotics.

Technical aspects of the CRP-POCT

Patients did not report any difficulties with the use of the CRP-POCT by clinicians. Primary care staff reported being able to use the CRP-POCT with all patients randomised to the intervention arm, and in general the CRP-POCT was easy to use. The need to refrigerate cartridges and allow time for them to return to room temperature before use, and the need to regularly carry out control testing, were seen as burdensome and were potential barriers to implementation. Clinicians felt that some modifications to the technology would facilitate implementation: ‘I think that, you know, in theory that [using the CRP-POCT in routine care] could be very good, but the only thing I would say is that because it’s so cumbersome within the consultation clinicians won’t use it, I’m just being honest with you, it takes, you know, 10 minutes to go and sort the machine and calibrate it, you know, how easy is that going to be?’(GP8) ‘I think it would be nicer if it was, you know in and ideal world, if it was a hand held machine, so you could take it with you on a, on a home visit for instance, would be a useful.’(GP7)

Time and resources

Patients felt that use of the test was quick. The primary care staff felt that using the CRP-POCT made consultations slightly longer, but felt that this was a good investment of their time: ‘I think where there was a great degree of uncertainty about what the right thing was to do, yeah there are definitely times when you’d be willing to invest that extra bit of time to do it.’(GP9)

Primary care staff felt that the cost of the CRP-POCT machine and cartridges was prohibitive under their current funding arrangements, and it would not be widely adopted unless additional funding was provided to cover these costs.

Contextual factors that could influence the way the CRP-POCT is implemented

Patient attitudes with regard to antibiotic use for AECOPD were varied, but many did not want to take antibiotics for AECOPD unless they were required. Patient anxiety, a strong patient preference for antibiotics, and individual circumstances (for example, recent death of a spouse) were cited by primary care staff as reasons for still prescribing antibiotics despite a low CRP-POCT result, indicating that non-medical factors continued to influence antibiotic prescribing.

DISCUSSION

Summary

It was found that patients and clinicians considered CRP-POCT useful in guiding management of AECOPD by providing an indication of disease severity, facilitating communication with patients and managing their expectations, and increasing clinician and patient confidence in antibiotic prescribing decisions. Previous research identified difficulties with interpreting the implications of CRP results and concerns about distracting from clinical reasoning as perceived barriers to implementing CRP-POCTs.²² In the PACE study, clinicians were given guidance on the interpretation of the CRP-POCT result and were asked to use the CRP-POCT with all patients randomised to the CRP-POCT trial arm. The clinician responders in this study did not report difficulties in interpreting the CRP-POCT or any negative impact on their clinical judgement. Likewise, the patient responders felt the CRP-POCT was useful in guiding their doctors’ antibiotic prescribing decisions and felt that it would be acceptable for use in routine care.

Strengths and limitations

GPs and patients who agreed to participate in the trial may have more favourable views about this technology than those who did not agree to participate. Nonetheless, the interview participants were purposively sampled using a maximum variation approach to ensure that a range of views were captured; therefore, the findings are likely to be representative of those who would be willing to consider use of the test. Approximately 50% of practices and 40% of patients invited for interviews in the present study declined or did not respond; thus self-selection to the interviews may also have introduced a sampling bias. An implementation study involving a wider roll-out of the intervention would be required to investigate to what extent the findings this study generalise to a broader population.

Comparison with existing literature

Patients with COPD are at risk of developing frequent and severe respiratory infections that can have a long-term impact on their lung function,²³ and in this sense are a higher risk group than those with uncomplicated acute cough presenting in primary care. Nonetheless, clinicians and patients in this study expressed similar views on the use of CRP-POCT to guide antibiotic prescribing as those that have been reported in studies of acute cough.²²^,²⁴^–²⁸ Potential benefits of using CRP-POCT routinely include improved opportunity for early intervention, reduced hospital admissions, and reducing unnecessary use of antibiotics by improving diagnostic confidence and improving communication with patients.²⁴^,²⁹^,³⁰

CRP-POCTs are widely used in a number of European countries in the management of lower respiratory tract infections, but have not yet become routinely used in the UK.²⁹ The absence of a funding and reimbursement model has been identified as the primary barrier for widespread adoption of the CRP-POCT for acute cough in the NHS.²⁹ This concern was shared by the clinicians interviewed in the PACE study and should be addressed by commissioners and policymakers to enable implementation of the CRP-POCT in routine care for AECOPD.

Previous research has suggested that risk aversion and the perception that the CRP-POCT is time consuming are potential barriers to its adoption.²⁹ In this study, clinicians felt that the CRP-POCT would reduce risk to patients through better targeting of antibiotics, and as such the time invested in carrying out the test was worthwhile. In European countries where CRP-POCT is routinely used for acute cough, typical patient pathways for CRP-POCT include GPs or practice nurses carrying out the CRP-POCT during their consultations with patients.²⁹ Other appropriately trained members of primary care teams can complete the CRP-POCT and pass the information on to prescribing clinicians.²⁹ In the PACE study, practices could use any of these options to enable them to adapt the implementation of the CRP-POCT to fit with their practice routines.¹⁰ Clinicians had mixed views on how the CRP-POCT should be implemented in terms of whether it should be used with all patients, or only in cases where there was clinical uncertainty. Guidance for clinicians will be required in introducing CRP-POCT in routine practice for AECOPD to facilitate its implementation in a consistent and effective way.

The PACE study found reduced antibiotic prescribing for AECOPD from CRP-POCT use. Nevertheless, antibiotics were prescribed for 33% of patients who had a CRP-POCT level of <20 mg/L, where guidance indicated that antibiotics were unlikely to be of benefit.¹¹ Non-clinical contextual factors, such as patient expectations for antibiotics, access to antibiotics before consulting with a clinician, and a lack of clear guidelines, can influence clinicians’ antibiotic prescribing behaviour for acute cough.³¹ Being able to effectively elicit patient ideas, concerns, expectations, and beliefs is an important skill for clinicians managing acute cough.²⁷^,³²^,³³ There was a perception among some patients that the CRP-POCT provided an accurate indication on ‘how ill’ they were, as well as some misconceptions around the necessity of antibiotics for bacterial and viral infections. Providing patients with more information about the function of the test and effective management of exacerbations may, therefore, be of benefit. In acute cough, a combination of training in the use of a CRP-POCT and enhanced communication-skills training has a larger effect than either form of training alone.³³ Additional training for clinicians in integrating the CRP-POCT into consultations in a patient-centred way, and enhanced information for patients about the purpose and potential benefits of testing, might facilitate adoption.

Implications for practice

Patients and clinicians reported that the CRP-POCT led to less clinical uncertainty, increased prescribing confidence, and enhanced communication. These were all potential mechanisms for a safe reduction in overall antibiotic use for AECOPD. Both patients and clinicians emphasised the need to use the CRP-POCT as part of, not in place of, a high-quality consultation that includes a clinical examination, elicitation of patient views and preferences, the application of the prescriber’s clinical judgement, and information on how and when patients should re-consult if they are not recovering as expected. Taken together with the quantitative findings from the PACE study,¹¹ the findings of the present study suggest that healthcare practitioners should consider adopting CRP-POCT in the routine management of AECOPD. Implementation planning for the CRP-POCT should include consideration of funding arrangements, simplifying the CRP-POCT technology so that it is quicker and easier to use, guidelines on implementation of the test for clinicians, and better information for patients.

Acknowledgments

The authors would like to thank all of those who have supported the PACE study, including the members of the PACE study research team, Trial Management Group, Trial Steering Committee, and Independent Data Monitoring and Ethics Committee, as well as the patients and primary healthcare staff who took part in these interviews for their time and for sharing their views. The authors would also like to acknowledge and thank the Health and Care Research Wales Workforce, the Thames Valley and South Midlands, Eastern, and West of England Primary Care Research Networks, and Comprehensive Local Clinical Research Networks for their support in recruiting sites for the PACE study.

Notes

Funding

The PACE study was funded by the National Institute for Health Research (NIHR) Health Technology Assessment Programme (project number: 12/33/12). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, or the Department of Health. The work was undertaken with the support of the UK Clinical Research Collaboration (UKCRC)-registered Centre for Trials Research, who receive infrastructure funding from Health and Care Research Wales and Cancer Research UK, and the UKCRC-registered University of Oxford Primary Care and Vaccines Clinical Trials Consortium. Neither the NIHR nor Health and Care Research Wales had a role in study design, data collection, data analysis, data interpretation, and report writing.

Ethical approval

The PACE study protocol was approved on 15 September 2014 by the Research Ethics Committee (REC) For Wales (Wales REC 6) recognised by the United Kingdom Ethics Committee Authority (REC ref: 14/WA/1106). Written informed consent was obtained from all patients who took part in an interview. Primary care staff offered their consent to take part in the interview verbally over the telephone. Clinical Trial Registration ref: ISRCTN24346473.

Provenance

Freely submitted; externally peer reviewed.

Competing interests

Rhiannon Phillips and Micaela Gal were research fellows at the Wales Centre for Primary and Emergency Research, Cardiff University, supported by a Research Centre Grant from Health and Care Research Wales during the completion of this study. Christopher Butler is an NIHR senior investigator, and clinical director of the University of Oxford Primary Care and Vaccines Clinical Trials Collaboration, and the NIHR Oxford Community Medical Technology and In Vitro Diagnostics Cooperative. Christopher Butler has received fees from Roche Diagnostics for participating in an Advisory Board about point-of-care testing; holds a grant from Roche Diagnostics to evaluate the analytic performance of a point-of-care testing device; and is part of publicly funded research consortia that include industrial partners. Jochen Cals is professor and primary care physician at the Department of Family Medicine at Maastricht University, and is supported by a Veni-grant (ref: 91614078) of the Netherlands Organization for Health Research and Development (ZonMw). Kerenza Hood is a Health and Care Research Wales senior investigator and director of the Centre for Trials Research.

Discuss this article

Contribute and read comments about this article: bjgp.org/letters

Received August 27, 2019.
Revision requested October 15, 2019.
Accepted December 23, 2019.

©The Authors

http://creativecommons.org/licenses/by-nc/4.0/

This article is Open Access: CC BY-NC 4.0 licence (http://creativecommons.org/licences/by-nc/4.0/).

REFERENCES

1.↵
1. Llor C,
2. Bjerrum L,
3. Munck A,
4. et al.
Predictors for antibiotic prescribing in patients with exacerbations of COPD in General PracticeTher Adv Respir Dis201373131137
OpenUrl CrossRef PubMed
2.↵
1. Vollenweider DJ,
2. Jarrett H,
3. Steurer-Stey CA,
4. et al.
Antibiotics for exacerbations of chronic obstructive pulmonary disease.Cochrane Database Syst Rev201212CD010257
OpenUrl CrossRef PubMed
3.↵
1. Soler N,
2. Ewig S,
3. Torres A,
4. et al.
Airway inflammation and bronchial microbial patterns in patients with stable chronic obstructive pulmonary diseaseEur Respir J199914510151022
OpenUrl Abstract
4.
1. Cosby JL,
2. Francis N,
3. Butler CC
The role of evidence in the decline of antibiotic use for common respiratory infections in primary careLancet Infect Dis2007711749756
OpenUrl CrossRef PubMed
5.
1. Costelloe C,
2. Metcalfe C,
3. Lovering A,
4. et al.
Effect of antibiotic prescribing in primary care on antimicrobial resistance in individual patients: systematic review and meta-analysisBMJ2010340c2096
OpenUrl Abstract/FREE Full Text
6.↵
1. Goossens H,
2. Ferech M,
3. Vander Stichele R,
4. Elseviers M
Outpatient antibiotic use in Europe and association with resistance: a cross-national database studyLancet20053659459579587
OpenUrl CrossRef PubMed
7.↵
1. Global Initiative for Chronic Obstructive Lung Disease (GOLD)
GOLD 2017 global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease 2017 reportAm J Respir Crit Care Med20171955557582
OpenUrl CrossRef PubMed
8.↵
1. National Institute for Health and Care Excellence
Chronic obstructive pulmonary disease in over 16s: diagnosis and management NG115LondonNICE2019https://www.nice.org.uk/guidance/ng115 (accessed 20 Apr 2020).
9.↵
1. Hurst JR,
2. Donaldson GC,
3. Perera WR,
4. et al.
Use of plasma biomarkers at exacerbation of chronic obstructive pulmonaryAm J Respir Crit Care Med20061748867874
OpenUrl CrossRef PubMed
10.↵
1. Bates J,
2. Francis NA,
3. White P,
4. et al.
GP use of a C-reactive protein point-of-care test to help target antibiotic prescribing in patients with acute exacerbations of chronic obstructive pulmonary disease (the PACE study): study protocol for a randomised controlled trialTrials2017181442
OpenUrl
11.↵
1. Butler CC,
2. Gillespie D,
3. White P,
4. et al.
C-reactive protein testing to guide antibiotic prescribing for COPD exacerbationsNew Engl J Med20193812111120
OpenUrl
12.↵
1. Moore G,
2. Audrey S,
3. Barker M,
4. et al.
Process evaluation of complex interventions: Medical Research Council guidanceBMJ2015350h1258
OpenUrl FREE Full Text
13.↵
1. Oakley A,
2. Strange V,
3. Bonell C,
4. et al.
Health services research: process evaluation in randomised controlled trials of complex interventionsBMJ20063327538413
OpenUrl FREE Full Text
14.↵
1. Cresswell JW,
2. Plano Clark VL
Designing and conducting mixed method research2nd ednThousand Oaks, CASage Publications2011
15.↵
1. Palinkas LA,
2. Horwitz SM,
3. Green CA,
4. et al.
Purposeful sampling for qualitative data collection and analysis in mixed method implementation researchAdm Policy Ment Health2015425533544
OpenUrl CrossRef PubMed
16.↵
1. Gale N,
2. Heath G,
3. Cameron E,
4. et al.
Using the framework method for the analysis of qualitative data in multi-disciplinary health researchBMC Med Res Methodol201313117
OpenUrl CrossRef PubMed
17.
1. Srivastava A,
2. Thomson SB
Framework analysis: a qualitative methodology for applied policy researchJournal of Administration and Governance2009427278
OpenUrl
18.↵
1. Bryman A,
2. Burgess R
1. Ritchie J,
2. Spencer L
Qualitative data analysis for applied policy researchBryman A, Burgess RAnalyzing qualitative dataLondonRoutledge1994
19.↵
1. Fereday J,
2. Muir-Cochrane E
Demonstrating rigor using thematic analysis: a hybrid approach of inductive and deductive coding and theme developmentInt J Qual Methods2006518092
OpenUrl
20.↵
1. Barbour RS
Checklists for improving rigour in qualitative research: a case of the tail wagging the dog?BMJ2001322729411151117
OpenUrl FREE Full Text
21.↵
1. Fusch P,
2. Ness L
Are we there yet? Data saturation in qualitative researchQual Rep201520914081416
OpenUrl
22.↵
1. Wood F,
2. Brookes-Howell L,
3. Hood K,
4. et al.
A multi-country qualitative study of clinicians’ and patients’ views on point of care tests for lower respiratory tract infectionFam Pract2011286661669
OpenUrl CrossRef PubMed
23.↵
1. Sethi S
Infection as a comorbidity of COPDEur Respir J201035612091215
OpenUrl Abstract/FREE Full Text
24.↵
1. Cals JWL,
2. Chappin FHF,
3. Hopstaken RM,
4. et al.
C-reactive protein point-of-care testing for lower respiratory tract infections: a qualitative evaluation of experiences by GPsFam Pract2010272212218
OpenUrl CrossRef PubMed
25.
1. Hardy V,
2. Thompson M,
3. Keppel GA,
4. et al.
Qualitative study of primary care clinicians’ views on point-of-care testing for C-reactive protein for acute respiratory tract infections in family medicineBMJ Open201771
OpenUrl CrossRef
26.
1. Anthierens S,
2. Tonkin-Crine S,
3. Cals JW,
4. et al.
Clinicians’ views and experiences of interventions to enhance the quality of antibiotic prescribing for acute respiratory tract infectionsJ Gen Intern Med2015304408416
OpenUrl CrossRef PubMed
27.↵
1. Tonkin-Crine S,
2. Anthierens S,
3. Francis NA,
4. et al.
Exploring patients’ views of primary care consultations with contrasting interventions for acute cough: a six-country European qualitative study.NPJ Prim Care Respir Med20142414026
OpenUrl PubMed
28.↵
1. Cals JW,
2. Butler CC,
3. Dinant G-J
‘Experience talks’: physician prioritisation of contrasting interventions to optimise management of acute cough in general practiceImplement Sci20094157
OpenUrl CrossRef PubMed
29.↵
1. Huddy JR,
2. Ni MZ,
3. Barlow J,
4. et al.
Point-of-care C reactive protein for the diagnosis of lower respiratory tract infection in NHS primary care: a qualitative study of barriers and facilitators to adoptionBMJ Open201663e009959
OpenUrl Abstract/FREE Full Text
30.↵
1. Anthierens S,
2. Tonkin-Crine S,
3. Douglas E,
4. et al.
GPs’ views on the acceptability and applicability of a web-based intervention to reduce antibiotic prescribing for acute cough in multiple European countries: a qualitative study prior to a randomised trial.BMC Fam Pract201213101
OpenUrl CrossRef PubMed
31.↵
1. Brookes-Howell L,
2. Hood K,
3. Cooper L,
4. et al.
Clinical influences on antibiotic prescribing decisions for lower respiratory tract infection: a nine country qualitative study of variation in careBMJ Open201223e000795
OpenUrl Abstract/FREE Full Text
32.↵
1. Coenen S,
2. Francis N,
3. Kelly M,
4. et al.
Are patient views about antibiotics related to clinician perceptions, management and outcome? A multi-country study in outpatients with acute coughPloS One2013810e76691
OpenUrl CrossRef PubMed
33.↵
1. Little P,
2. Stuart B,
3. Francis N,
4. et al.
Effects of internet-based training on antibiotic prescribing rates for acute respiratory-tract infections: a multinational, cluster, randomised, factorial, controlled trialLancet2013382989911751182
OpenUrl CrossRef PubMed

In this issue

Download PDF

Email Article

Citation Tools

Keywords

More in this TOC Section

Show more Research

Cited By...

Intended for Healthcare Professionals

Tweets by @BJGPjournal

[1] 1.↵
Llor C,
Bjerrum L,
Munck A,
et al.
Predictors for antibiotic prescribing in patients with exacerbations of COPD in General PracticeTher Adv Respir Dis201373131137
OpenUrl CrossRef PubMed

[2] Llor C,

[3] Bjerrum L,

[4] Munck A,

[5] et al.

[6] 2.↵
Vollenweider DJ,
Jarrett H,
Steurer-Stey CA,
et al.
Antibiotics for exacerbations of chronic obstructive pulmonary disease.Cochrane Database Syst Rev201212CD010257
OpenUrl CrossRef PubMed

[7] Vollenweider DJ,

[8] Jarrett H,

[9] Steurer-Stey CA,

[10] et al.

[11] 3.↵
Soler N,
Ewig S,
Torres A,
et al.
Airway inflammation and bronchial microbial patterns in patients with stable chronic obstructive pulmonary diseaseEur Respir J199914510151022
OpenUrl Abstract

[12] Soler N,

[13] Ewig S,

[14] Torres A,

[15] et al.

[16] 4.
Cosby JL,
Francis N,
Butler CC
The role of evidence in the decline of antibiotic use for common respiratory infections in primary careLancet Infect Dis2007711749756
OpenUrl CrossRef PubMed

[17] Cosby JL,

[18] Francis N,

[19] Butler CC

[20] 5.
Costelloe C,
Metcalfe C,
Lovering A,
et al.
Effect of antibiotic prescribing in primary care on antimicrobial resistance in individual patients: systematic review and meta-analysisBMJ2010340c2096
OpenUrl Abstract/FREE Full Text

[21] Costelloe C,

[22] Metcalfe C,

[23] Lovering A,

[24] et al.

[25] 6.↵
Goossens H,
Ferech M,
Vander Stichele R,
Elseviers M
Outpatient antibiotic use in Europe and association with resistance: a cross-national database studyLancet20053659459579587
OpenUrl CrossRef PubMed

[26] Goossens H,

[27] Ferech M,

[28] Vander Stichele R,

[29] Elseviers M

[30] 7.↵
Global Initiative for Chronic Obstructive Lung Disease (GOLD)
GOLD 2017 global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease 2017 reportAm J Respir Crit Care Med20171955557582
OpenUrl CrossRef PubMed

[31] Global Initiative for Chronic Obstructive Lung Disease (GOLD)

[32] 8.↵
National Institute for Health and Care Excellence
Chronic obstructive pulmonary disease in over 16s: diagnosis and management NG115LondonNICE2019https://www.nice.org.uk/guidance/ng115 (accessed 20 Apr 2020).

[33] National Institute for Health and Care Excellence

[34] 9.↵
Hurst JR,
Donaldson GC,
Perera WR,
et al.
Use of plasma biomarkers at exacerbation of chronic obstructive pulmonaryAm J Respir Crit Care Med20061748867874
OpenUrl CrossRef PubMed

[35] Hurst JR,

[36] Donaldson GC,

[37] Perera WR,

[38] et al.

[39] 10.↵
Bates J,
Francis NA,
White P,
et al.
GP use of a C-reactive protein point-of-care test to help target antibiotic prescribing in patients with acute exacerbations of chronic obstructive pulmonary disease (the PACE study): study protocol for a randomised controlled trialTrials2017181442
OpenUrl

[40] Bates J,

[41] Francis NA,

[42] White P,

[43] et al.

[44] 11.↵
Butler CC,
Gillespie D,
White P,
et al.
C-reactive protein testing to guide antibiotic prescribing for COPD exacerbationsNew Engl J Med20193812111120
OpenUrl

[45] Butler CC,

[46] Gillespie D,

[47] White P,

[48] et al.

[49] 12.↵
Moore G,
Audrey S,
Barker M,
et al.
Process evaluation of complex interventions: Medical Research Council guidanceBMJ2015350h1258
OpenUrl FREE Full Text

[50] Moore G,

[51] Audrey S,

[52] Barker M,

[53] et al.

[54] 13.↵
Oakley A,
Strange V,
Bonell C,
et al.
Health services research: process evaluation in randomised controlled trials of complex interventionsBMJ20063327538413
OpenUrl FREE Full Text

[55] Oakley A,

[56] Strange V,

[57] Bonell C,

[58] et al.

[59] 14.↵
Cresswell JW,
Plano Clark VL
Designing and conducting mixed method research2nd ednThousand Oaks, CASage Publications2011

[60] Cresswell JW,

[61] Plano Clark VL

[62] 15.↵
Palinkas LA,
Horwitz SM,
Green CA,
et al.
Purposeful sampling for qualitative data collection and analysis in mixed method implementation researchAdm Policy Ment Health2015425533544
OpenUrl CrossRef PubMed

[63] Palinkas LA,

[64] Horwitz SM,

[65] Green CA,

[66] et al.

[67] 16.↵
Gale N,
Heath G,
Cameron E,
et al.
Using the framework method for the analysis of qualitative data in multi-disciplinary health researchBMC Med Res Methodol201313117
OpenUrl CrossRef PubMed

[68] Gale N,

[69] Heath G,

[70] Cameron E,

[71] et al.

[72] 17.
Srivastava A,
Thomson SB
Framework analysis: a qualitative methodology for applied policy researchJournal of Administration and Governance2009427278
OpenUrl

[73] Srivastava A,

[74] Thomson SB

[75] 18.↵
Bryman A,
Burgess R
Ritchie J,
Spencer L
Qualitative data analysis for applied policy researchBryman A, Burgess RAnalyzing qualitative dataLondonRoutledge1994

[76] Bryman A,

[77] Burgess R

[78] Ritchie J,

[79] Spencer L

[80] 19.↵
Fereday J,
Muir-Cochrane E
Demonstrating rigor using thematic analysis: a hybrid approach of inductive and deductive coding and theme developmentInt J Qual Methods2006518092
OpenUrl

[81] Fereday J,

[82] Muir-Cochrane E

[83] 20.↵
Barbour RS
Checklists for improving rigour in qualitative research: a case of the tail wagging the dog?BMJ2001322729411151117
OpenUrl FREE Full Text

[84] Barbour RS

[85] 21.↵
Fusch P,
Ness L
Are we there yet? Data saturation in qualitative researchQual Rep201520914081416
OpenUrl

[86] Fusch P,

[87] Ness L

[88] 22.↵
Wood F,
Brookes-Howell L,
Hood K,
et al.
A multi-country qualitative study of clinicians’ and patients’ views on point of care tests for lower respiratory tract infectionFam Pract2011286661669
OpenUrl CrossRef PubMed

[89] Wood F,

[90] Brookes-Howell L,

[91] Hood K,

[92] et al.

[93] 23.↵
Sethi S
Infection as a comorbidity of COPDEur Respir J201035612091215
OpenUrl Abstract/FREE Full Text

[94] Sethi S

[95] 24.↵
Cals JWL,
Chappin FHF,
Hopstaken RM,
et al.
C-reactive protein point-of-care testing for lower respiratory tract infections: a qualitative evaluation of experiences by GPsFam Pract2010272212218
OpenUrl CrossRef PubMed

[96] Cals JWL,

[97] Chappin FHF,

[98] Hopstaken RM,

[99] et al.

[100] 25.
Hardy V,
Thompson M,
Keppel GA,
et al.
Qualitative study of primary care clinicians’ views on point-of-care testing for C-reactive protein for acute respiratory tract infections in family medicineBMJ Open201771
OpenUrl CrossRef

[101] Hardy V,

[102] Thompson M,

[103] Keppel GA,

[104] et al.

[105] 26.
Anthierens S,
Tonkin-Crine S,
Cals JW,
et al.
Clinicians’ views and experiences of interventions to enhance the quality of antibiotic prescribing for acute respiratory tract infectionsJ Gen Intern Med2015304408416
OpenUrl CrossRef PubMed

[106] Anthierens S,

[107] Tonkin-Crine S,

[108] Cals JW,

[109] et al.

[110] 27.↵
Tonkin-Crine S,
Anthierens S,
Francis NA,
et al.
Exploring patients’ views of primary care consultations with contrasting interventions for acute cough: a six-country European qualitative study.NPJ Prim Care Respir Med20142414026
OpenUrl PubMed

[111] Tonkin-Crine S,

[112] Anthierens S,

[113] Francis NA,

[114] et al.

[115] 28.↵
Cals JW,
Butler CC,
Dinant G-J
‘Experience talks’: physician prioritisation of contrasting interventions to optimise management of acute cough in general practiceImplement Sci20094157
OpenUrl CrossRef PubMed

[116] Cals JW,

[117] Butler CC,

[118] Dinant G-J

[119] 29.↵
Huddy JR,
Ni MZ,
Barlow J,
et al.
Point-of-care C reactive protein for the diagnosis of lower respiratory tract infection in NHS primary care: a qualitative study of barriers and facilitators to adoptionBMJ Open201663e009959
OpenUrl Abstract/FREE Full Text

[120] Huddy JR,

[121] Ni MZ,

[122] Barlow J,

[123] et al.

[124] 30.↵
Anthierens S,
Tonkin-Crine S,
Douglas E,
et al.
GPs’ views on the acceptability and applicability of a web-based intervention to reduce antibiotic prescribing for acute cough in multiple European countries: a qualitative study prior to a randomised trial.BMC Fam Pract201213101
OpenUrl CrossRef PubMed

[125] Anthierens S,

[126] Tonkin-Crine S,

[127] Douglas E,

[128] et al.

[129] 31.↵
Brookes-Howell L,
Hood K,
Cooper L,
et al.
Clinical influences on antibiotic prescribing decisions for lower respiratory tract infection: a nine country qualitative study of variation in careBMJ Open201223e000795
OpenUrl Abstract/FREE Full Text

[130] Brookes-Howell L,

[131] Hood K,

[132] Cooper L,

[133] et al.

[134] 32.↵
Coenen S,
Francis N,
Kelly M,
et al.
Are patient views about antibiotics related to clinician perceptions, management and outcome? A multi-country study in outpatients with acute coughPloS One2013810e76691
OpenUrl CrossRef PubMed

[135] Coenen S,

[136] Francis N,

[137] Kelly M,

[138] et al.

[139] 33.↵
Little P,
Stuart B,
Francis N,
et al.
Effects of internet-based training on antibiotic prescribing rates for acute respiratory-tract infections: a multinational, cluster, randomised, factorial, controlled trialLancet2013382989911751182
OpenUrl CrossRef PubMed

[140] Little P,

[141] Stuart B,

[142] Francis N,

[143] et al.

Main menu

User menu

Search