Eliciting required and desired diagnostic test characteristics to develop patient-focused target product profiles (TPPs) for diagnosing infection and exacerbations in cystic fibrosis

Patient and Public Involvement

Following on from the research priorities set out by the James Lind Alliance (https://www.jla.nihr.ac.uk/priority-setting-partnerships/cystic-fibrosis) who consulted with pwCF, family members and clinical care teams globally, the project team wanted to further understand the diagnostic unmet needs and priority areas for clinicians and pwCF, and investigate where, and what type of new diagnostics tests might have the most patient and NHS benefits.

For the focus groups informed consent forms and participant information sheets were drafted and revised following consultation with a patient and public involvement and engagement (PPIE) manager and the NIHR Newcastle MIC insight public panel members. Insight panel members were also given an overview of the project and it’s aims and invited to comment on the methodology and later the first draft of the TPP document, which was revised accordingly.

This project also involved an expert advisory group comprised of people with lived experience of CF, clinicians, methodologists expert in TPP development, CF charity members and academics with an interest in CF and microbiology. The panel met formally approximately 7 times throughout the duration of the project, with supplementary ad-hoc meetings and online document review conducted as required. Output from work at each stage of the project was shared with the group including drafts of the TPP document, which was revised accordingly. The expert advisory group for this project contained a person with CF, and output was shared with the group at each stage. The focus groups described here contained people with CF and were also attended by the PPIE Coordinator from the CF Trust who provided contributions and facilitation.

A virtual symposium held after the focus groups and not described here had a pwCF as a keynote speaker and pwCF in all breakout room and sessions.

Participant selection

18 participants were recruited through existing networks; through the CF trust, or personal networks of the authors (convenience sampling) and invited to take part in the focus groups via email. Participants were based across different regions of the UK. Anonymised details of the focus group members with CF are outlined in Table 1 and details of clinical care team focus group members are given in Table 2 and Table 3.

Focus group format

Informed consent forms and participant information sheets were drafted and revised following consultation with a patient and public involvement and engagement (PPIE) manager at the NIHR Newcastle MIC. Consent was taken verbally prior to the focus group commencing by one researcher, and witnessed and recorded on the consent form by a second researcher. All participants consented to the recording of the interviews.

Separate draft focus group topic guides were designed for clinical staff and pwCF by members of the NIHR Newcastle MIC and revised iteratively through consultation with the wider project team (extended data).

Three focus groups were conducted on 13^th July 2022 and 20^th and 24^th October 2022, using Microsoft Teams. For each focus group, in addition to the lead interviewer (KG), an additional methodologist (RN) was present to act as facilitator and note taker/support. One 90-minute focus group was conducted with a total of six participants with lived experience of CF. LA and KG conducted the focus group, with support from CT, and PS. Two 120-minute focus groups were conducted with six clinical participants in each, totalling 12 clinical participants. KG conducted these focus groups with support from LA, CT, and RN. Six participants per focus group was selected to allow a range of views but allow time for participation by all present.

A lay summary explaining TPPs was produced for pwCF participating in the focus group and distributed prior to and during the focus group (extended data). The clinical participants were provided with an agenda and briefing document prior to the clinical focus groups (extended data).

Focus group topic guides ensured the focus groups addressed the main objectives including:

Analysis

Audio recordings of each focus group were auto-transcribed initially by the online web-service otter.ai (https://otter.ai/), before auto-generated transcripts were cleaned and checked for accuracy by author KG. Notes were taken by authors RN & CT throughout the session and used to assist the transcription and analysis. Accuracy of the interview content was confirmed by the authors, and any outstanding issues or clarifications were resolved with the focus group participants. Summary content and themes were extracted from the transcript to aid in early project decision making and to inform the subsequent analysis.

A thematic analysis as described by Braun and Clarke (Braun & Clarke, 2006) was performed based on the recordings, transcripts and notes taken during the interviews. Transcripts and notes were coded by hand by author KG and annotated to identify themes, after which an electronic thematic summary document was produced with relevant data extracted under each key theme. Themes were discussed and agreed with authors CT and RN. Themes broadly reflected the key questions asked during the focus group. One thematic summary document was produced for the clinician focus groups and a separate document was produced for the pwCF focus group. The thematic summary documents were checked for accuracy and completeness by authors KG and RN and agreed by all researchers who facilitated the focus groups. Results are presented in a narrative synthesis.

Quotes from focus group participants have been anonymised and appear in bold italicised text below.

Detection and diagnosis of infections and exacerbations in pwCF

The impact of HEMT

Elexacaftor/Tezacaftor/Ivacaftor (ETI) is licensed for use in pwCF with at least one F508del mutation in the CFTR gene. There remains a population of pwCF who are unable to benefit from HEMT like ETI. Disease causing CFTR gene mutations are responsible for the respiratory symptoms associated with CF, namely the production of thick, sticky mucus in the airways which is difficult to clear and leads to recurrent respiratory infections. ETI works to increase the activity of the CFTR proteins on the cell surface, leading to a reduction in pulmonary exacerbations and an improvement in quality of life During this elicitation exercise, both pwCF and the clinical care teams stated noticeable changes in the frequency of pulmonary exacerbations, as well as changes in the symptom associated with acute illness following the roll out of ETI.

When considering detection and diagnosis of pulmonary exacerbations in the era of HEMT, many pwCF have experienced changes to their symptom profile. Sputum production has decreased for many following initiation of HEMT, resulting in fewer expectorated sputum samples. This in turn has led to a greater reliance on obtaining induced sputum and cough swabs for detecting and diagnosing infections

Symptoms of infection and exacerbation

Participants in the pwCF and clinical team focus groups described subjective and objective signs and symptoms used to diagnose infection and exacerbation. Some of these were common, such as an increased cough, body pains, fever and inflammation, but others were person-specific, including post-nasal drip, migraines and dry, cracking skin on the lips and knuckles. With other reported symptoms including pleuritic pain, exhaustion, loss of taste (Ageusia) and loss of smell (Anosmia), diagnosis of infection based on signs and symptoms alone can be challenging. In addition, the importance of distinguishing a chronic infection from an acute exacerbation was flagged.

Other indicators of possible exacerbations are linked to monitoring tools. A decline in lung function noted by pwCF based on home spirometry can be an early indicator of a potential exacerbation, alongside changes to sputum colour, volume and consistency. Some participants in the pwCF group described noticing changes to their resting heart rate in the days leading up to the onset of symptoms. Continuous home monitoring of heart rate is part of ongoing CF trials such as Project Breathe, although no data has been released on the correlation between resting heart rate and pulmonary exacerbations, it is used by pwCF and their clinical teams as an early indicator of acute illness. Participants in the pwCF focus group also reported having a ‘gut feeling’ of impending illness that would lead them to seek antibiotic treatment from their care team.

Current diagnostic tests

There are a range of diagnostic tests currently in use in the UK NHS for diagnosing infection or exacerbation in pwCF. For example, sampling of airway secretions is considered ‘routine’ and will be performed during clinic visits in both symptomatic and asymptomatic cases. Others, including imaging techniques and more invasive procedures such as bronchoscopy, are performed ad-hoc, when additional information is required to aid decision making. Blood tests, i.e. whole blood counts, and spirometry, while not diagnostic tests for infection per se, are used as part of a series of tests which when viewed collectively assist in diagnosing acute illness alongside culturing techniques.

In this section the tests elicited through the focus groups are considered in turn, highlighting features and characteristics of the tests and opinions from pwCF and clinicians regarding their use, accuracy and acceptability.

Cough swabs

Diagnostic type: Culture

Population: Adults and children unable to expectorate sputum. Predominantly used in paediatrics.

Indication: first-line diagnostic (in those unable to expectorate sputum), routine, symptomatic and asymptomatic patients.

Setting: at-home, GP/primary care, community, secondary care.

Time to results: minimum ~3 days, (considered less sensitive than sputum for detection of different pathogens).

Elicitation - Use: Cough swabs were generally performed by a specialist nurse or physiotherapist, and were more commonly used in paediatrics, however since the roll out of HEMT, use of cough swabs in the adult population has risen as many are no longer able to expectorate sputum on demand. The samples would be sent via courier or internal post to a microbiology laboratory for analysis. Cough swabs are taken regularly both during routine clinic visits and during periods of infection.

Elicitation - Accuracy: Participants in all focus groups believed cough swabs were not accurate or reliable and would not normally be used in isolation. One clinical representative (CFG1_2) stated that cough swabs had a good positive predictive value, but a poor negative predictive value, indicating that a positive result usually indicated an infection, but with a negative result, clinicians would not be confident enough to rule out an infection based on a cough swab alone.

The inaccuracy of cough swabs, particularly where there is a negative result, makes decision making around antibiotic usage more difficult. This impacts decisions such as whether or not to continue therapies for chronic infection, especially as some pwCF wish to discontinue their treatments on the basis of negative cough swabs which the clinical care team consider unreliable as a standalone result. Across the clinical care team, there was consensus around the worry of false negatives when it came to cough swabs

Elicitation - Acceptability: In regard to methods of sampling, cough swabs were the least invasive method according to the focus group attendees, they were commonly used in the paediatric settings. They were considered the most tolerable test in adults and children.

Elicitation - Other comments: Many pwCF taking HEMT are no longer able to expectorate sputum on demand, therefore cough swabs have seen more use in some clinics, despite the lack of confidence in their sensitivity compared with sample type like sputum and bronchoalveolar lavage (BAL) .

Sputum-based culture methods (sputum, induced sputum)

Diagnostic type: Culture

Population: Adults and children with CF, predominantly adult populations

Indication: first-line diagnostic, routine, symptomatic and asymptomatic patients.

Setting: at-home, GP/primary care, community, secondary care.

Time to results: minimum ~3 days, up to 6-8 weeks, depending on infection and pathogen type.

Elicitation - Use: Sputum sampling for culture is a routine method for detection and diagnosis of pulmonary infection. There are multiple indications for use including: pathogen characterisation including resistance profiling, routine surveillance, and for detecting and diagnosing infection in symptomatic people.

Elicitation - Accuracy: Sputum is considered a ‘gold standard’ specimen used for infection diagnostics in CF and culture results typically inform clinical management plans.

With the use of HEMT leading to a reduction in expectorated sputum, induced sputum samples are becoming more common. One clinical participant added that sputum induction at times yields secretions that are salivary in nature instead of the ‘classic CF sputum’ and that at times the induction procedure yields no sample altogether despite use of potent strengths of hypertonic saline

Elicitation - Acceptability: pwCF considered expectorated sputum testing to be an acceptable and necessary part of their care, although members of the clinical care team noted that children and teens do not generally like to provide sputum and do not take well to sputum induction.

A participant in a clinical team focus group commented that ‘floating the idea’ of sputum induction would usually be enough to encourage patients to attempt to provide an expectorated sample, which possibly suggests low acceptability of induced sputum, even among adults. On the other hand, it suggests that while expectorated sputum on demand is challenging to obtain, sputum samples expectorated spontaneously outside of the clinic setting can be captured and used.

Elicitation - Other comments: One clinical representative stated that following HEMT, the appearance and consistency of sputum had changed for many, with some pwCF equating these changes with being non-productive. One clinical care team participant felt that in this instance pwCF would benefit from receiving education on sputum changes associated with HEMT while being reminded of the utility of sending their samples for testing anyway as many were still expectorating useful samples.

Early detection of problematic pathogens via sinus sampling/upper airway was suggested by participants across the clinical team and pwCF focus groups. Some participants in the pwCF group described experiencing sinus symptoms in the lead up to a lung infection. This was validated by a participant in the clinical care team who said that they had pwCF on HEMT who were experiencing sinus symptoms and that at times they performed sinus sampling at their centre. They added that even where pathogens like Pseudomonas are cleared, there is a risk of reseeding via the sinuses

Bronchoscopy and BAL

Diagnostic type: Imaging / Culture

Population: Symptomatic adults and children, where infection persists despite treatment, or where initial diagnostic tests have proved inconclusive.

Indication: Second-line diagnostic, non-routine, emergency, ad-hoc

Setting: Secondary care, specialist centres, emergency department

Time to results: Typically 1–3 days for reporting; ~3 days, or up to 6-8 weeks, depending on infection and pathogen type

Elicitation - Use: Bronchoscopy and BAL was not used routinely, but would be considered where additional information was required. Bronchoscopy is a generally invasive technique, requiring specialist teams and equipment- it is not performed as a first-line diagnostic.

One clinical participant stated that bronchoscopies are more commonly used in cases of suspected or proven Non-tuberculous mycobacterium (NTM) infection due to concerns of impact on lung health. Here, bronchoscopy would be used to inform diagnosis and, or response to treatment.

Elicitation - Accuracy: No concerns were raised from participants in the pwCF group or clinical team group regarding the accuracy of bronchoscopy. One clinical team participant noted that on the occasions they do utilise bronchoscopy in adults, rarely do they gain additional information from the resulting cultures.

Elicitation - Acceptability: Participants across the focus groups stated that bronchoscopy was invasive, and acceptability amongst pwCF was poor. Participants in this group said their preference was to avoid undergoing bronchoscopy if possible, and the clinical teams would consider other tests first before deciding to perform a bronchoscopy.

Elicitation - Other comments: Bronchoscopy comes with inherent risk associated with the procedure. While those risks are generally low, they do have to be considered when planning clinical management.

Spirometry

Diagnostic type: Surrogate biomarker

Population: Adults and children with CF

Indication: first-line diagnostic, routine, monitoring, symptomatic and asymptomatic patients.

Setting: at-home, GP/primary care, community, secondary care.

Time to results: Immediate

Elicitation - Use: Spirometry, in particular Forced expiratory volume (FEV1) is used routinely for monitoring lung function and general health. Many pwCF have a spirometry device at home, which can be used daily and has utility as an early indicator of infection or exacerbation. Lung function testing can also be performed during routine clinic visits for continued monitoring and may be utilised during periods of infection or exacerbation to determine treatment efficacy.

For acute exacerbations, the goal of treatment is to ensure the patient returns to their baseline FEV1. Continued decline in lung function over time impacts morbidity and mortality. Regular monitoring of lung function is therefore encouraged by clinical care teams, both in periods of health and during periods of acute exacerbations

Elicitation - Accuracy: These techniques are not typically used in isolation for diagnosis of infection, but do indicate a potential issue that would be followed-up by other diagnostic testing and clinical investigations. Individual lung function measurements are not considered reliable, confidence in these tests comes from repeated measures, where longer term changes compared to a healthy baseline reading can flag decline in lung health or indicate improvements in health.

Some pwCF noted issues with the reliability of spirometry readings performed in the clinic, highlighting that different machines would give different readings, even when used sequentially. They described some machines as ‘temperamental’, requiring multiple attempts to generate a valid reading. This would occasionally prolong clinical visits and cause issues with interpretation of results.

Elicitation - Acceptability: Spirometry is an integral part of pwCFs clinic visits. Adult pwCF appeared to have accepted the technology as a standard part of managing their life with CF. Clinical staff did not note any particular acceptability issues with spirometry from a patient or clinical perspective.

Regarding home spirometry for general monitoring purposes, participants in the pwCF group offered opposing views on the importance and helpfulness of these tests for monitoring of their lung health. While all of the participants stated they were generally compliant with adhering to treatment regimens and attending clinic visits, some found home spirometry to be too intrusive for them to use regularly. One participant stated that they no longer used their device due to the impact it had on their mental health, largely due to the perceived expectation for them to do the test regularly. Other participants claimed that home spirometry had allowed them to keep on top of their health and gave them a sense of control over their own health and monitoring.

Elicitation - Other comments: The role of spirometry in CF is primarily in long term health monitoring, although it is also used to monitor treatment response and indicate potential infection. PwCF are encouraged to monitor themselves at home on a regular basis, although compliance varies significantly between people.

X-ray and CT scans

Diagnostic type: Imaging

Population: Symptomatic adults and children with CF, where infection persists despite treatment, or where initial diagnostic tests have proved inconclusive.

Indication: Second-line diagnostic, non-routine, emergency, ad-hoc

Setting: Secondary care, specialist centres, emergency department

Time to results: Typically 1-3 days for reporting.

Elicitation - Use: Clinical team member noted that X-ray and computed tomography (CT) scans were used to obtain baseline images in new, and transitioning pwCF but were not performed routinely. X-rays and CTs are used in part to diagnose infection, but also to rule-out other conditions such as pneumothorax.

These imaging techniques are used to inform clinical management and in the case of serial imaging, to evaluate the effectiveness of treatment or interventions.

Elicitation - Accuracy: There were no notable concerns elicited from the focus group participants regarding the accuracy of imaging techniques such as X-ray and CT. Generally an X-ray was considered under circumstances where it was indicated to help guide clinical decision making. It was not typically a first-line infection diagnostic tool, and would be used in conjunction with other measures, eg. sputum culture, spirometry, and self-reported signs and symptoms.

Elicitation - Acceptability: Participants in both the clinical groups and pwCF group felt that X-rays and CT scans had a purpose in symptomatic, unwell patients, however there was hesitation around frequent use due to concerns around the long terms effects of recurrent radiation exposure.

Magnetic resonance imaging (MRI) scans

Diagnostic type: Imaging

Population: Symptomatic adults and children with CF, where infection persists despite treatment, or where initial diagnostic tests have proved inconclusive.

Indication: Second-line diagnostic, non-routine, emergency, ad-hoc

Setting: Secondary care, specialist centres, emergency department

Time to results: Typically 1-3 days for reporting.

Elicitation - Use: MRI scanning is used similarly to other imaging techniques such as CT and X-rays, but was less commonly used due to increased cost and lower availability. Some participants in the pwCF group stated that their clinic offered them regular MRI scans every six months, however, in the clinical team focus group, some participants commented that they ‘wished’ they could offer MRI scans, but such testing was not feasible for them.

Where MRI scanning was used, it was considered a valuable diagnostic tool for detecting exacerbations and infections by pwCF and clinicians.

Elicitation - Accuracy: MRI was considered more accurate than X-rays. There were no concerns raised about MRI regarding test accuracy. It was considered superior to the more frequently used X-ray.

Elicitation - Acceptability: One participant in the pwCF group who’d admitted to having regular MRI scans felt that MRI was much more acceptable to them than X-ray techniques as they were concerned about the effects of radiation exposure following frequent X-ray imaging.

Elicitation - Other comments: It is likely that funding allocations between individual centres influence the use of MRI scans across CF centres. This may indicate a degree of regional inequality.

Blood tests

Diagnostic type: Surrogate biomarker

Population: Adults and children with CF

Indication: first-line diagnostic, routine, monitoring, symptomatic patients.

Setting: GP/primary care, community, Secondary care, specialist centres, emergency department.

Time to results: Typically 1-5 days for reporting.

Elicitation - Use: pwCF described the use of blood counts during periods of infection, with particular reference to whole blood cell counts being used as a measure of infection and response to treatment. These measurements would be used alongside spirometry, sputum testing and potentially imaging techniques to support clinical decision making, but would not be used alone as a diagnostic test for infection or exacerbation.

Elicitation - Accuracy: No concerns were raised from the pwCF group or clinical care groups regarding the accuracy of blood counts, however, as noted in the use section, this testing forms only part of a series of diagnostic tests.

Some participants in the pwCF group did note that on occasion, the sample of blood provided would be insufficient for the intended tests, which would result in incomplete information, or a possible need for resampling. This was seen as an inconvenience, but was not a major concern as treatment and management decisions would continue without the need for this test.

Elicitation - Acceptability: The clinical care groups found blood testing a useful and acceptable addition to the diagnostic and management process. Participants in the pwCF group were divided on their acceptability of blood tests based on their reluctance or dislike of the blood drawing process. Even those who strongly disliked giving blood commented that it was an acceptable option if the resulting test influenced clinical decision making. Upon further questioning, participants in this group generally considered finger-prick testing, as used for glucose monitoring, to be more acceptable than a venous blood draw based on their opinion of the relative discomfort and ease of performing each sampling type.

Elicitation - Other comments: Participants in both the clinical groups and pwCF group stated that they would be happy to see more research in diagnostics focused on using blood to detect infection/exacerbation. One clinical participant commented on the limited utility of c-reactive protein (CRP), adding that an alternative measure would be welcome. As blood is considered a low-invasive, easily collected sample type, with established pathways and protocols for sample collection, transport, storage, and analysis within both primary and secondary care.

Passive monitoring techniques

Diagnostic type: Surrogate biomarker

Population: Adults and children with CF

Indication: routine, ad-hoc, continuous monitoring, Symptomatic and asymptomatic patients.

Setting: at-home

Time to results: immediate, continuous

Elicitation - Use: While not in clinical use, ongoing trials such as Project Breathe involve a range of wearables, digital health devices, and at-home monitoring kits that are expected to inform early detection of exacerbations. Some participants in the pwCF group noted an elevated resting heart rate via their smart watches in the days leading up to symptomatic infection. While this is anecdotal evidence, early feedback from these studies show some promise for the use of continuous monitoring devices in the CF care pathway.

Elicitation - Accuracy: The accuracy of such digital health and continuous monitoring techniques has not yet been formally assessed. These technologies are not used in the CF care pathway, outside of individual clinical trial use and research experiments.

Elicitation - Acceptability: Both the pwCF group and clinical teams were supportive of continuous monitoring technologies. They were considered non-invasive and required the least amount of time and effort from a patient perspective.

A participant in the pwCF group expressed concerns around the potential impact and toll of regular, ongoing data collection on their mental health if a monitoring technology involved frequent data-sharing with clinical care teams and other NHS infrastructures.

The majority of participants in this group were very supportive of such technologies and stated that they would feel a greater degree of control over their condition if they could monitor themselves passively and visualise the data generated in their own time.

Elicitation - Other comments: While not in clinical use, passive, continuous monitoring technologies appear to offer low impact methods of assessing general health and are used by some pwCF as an early indicator of infection. This does affect the care pathway, as these people are more inclined to seek help earlier in response to changes in parameters such as resting heart rate, even in the absence of other symptoms of infection.

Interactions between pwCF and clinical care teams

PwCF and clinical team members were asked to describe their interactions with each other to determine the influence of decision making and clinical interactions.

Stress and anxiety regarding clinic visits

Some pwCF claimed that their routine clinic visits could be a cause of stress and anxiety on occasion. This was centred on ‘the need to perform’ clinical assessments within the time allocated for their visit. Clinics are an opportunity for staff and pwCF to review their health and management plan, and also to collect routine samples for surveillance purposes, antimicrobial resistance profiling, and obtain up to date spirometry readings. Given the difficulty of expectorating sputum for those on HEMT, and the unreliability of some spirometers, some pwCF reported feeling anxious prior to their clinical visits.

Clinical participants also reflected on the anxiety and stress that clinic visits may incite for pwCF. One care team member commented that the way in which clinic rooms are set up may contribute to this, typically with a clinical team member standing, or elevated on an office chair, with the pwCF in a smaller chair in a corner of the room. This is further compounded by instances where the clinical team member stands over the person while attempting to obtain a respiratory sample to send off for culture. Combined with a busy clinic environment, some pWCF may feel a sense of urgency during their clinic visit.

Reporting of results between clinical care team and pwCF

There was discussion amongst the pwCF focus group about the reporting of test results. Participants stated they would receive phone calls or messages about their results if the information wasn’t immediately available (e.g cultures taking a minimum of three days), they may also be called into the clinic for some results or if urgent interventions are required.

There were some concerns regarding reporting of results from some pwCF. One noted that they had access to ‘MyChart’, an app where they can view results of any recent tests. They mentioned receiving notification of results far more quickly through this app than from their clinical care team, and occasionally there would be discrepancies between the information accessible via the app and the information obtained via the care team. This appeared to be a discrepancy of detail. One example was an MRI result indicating suspicion of a mild new infection according to their patient record, and their care team relaying there were no concerns from the MRI. It is possible that the care team, having reviewed the report, concluded no further intervention was required, however the pwCF described feeling a loss of trust in their clinical team as a result of this discrepancy.

Shared decision making

Participants were asked to describe how decision making was shared between patients and their care team. Both groups described similar experiences. A participant in the clinical team explained that parents of young children with CF, and adults with a late diagnosis of CF, preferred to let the clinical care team lead on decision making as they were still learning about the condition and its management. One participant in the pwCF group with a young child with CF agreed that in paediatrics, decision-making was largely consultant-led, but caregivers were given time to ask questions, discuss the impact of decisions, and process what was happening.

Adults with CF reported greater autonomy and greater say on the decision-making process. The majority of adult CF representatives noted an equal, collaborative approach to decision making between them and their care team. As they learned more about their condition and the signs and symptoms of infection and exacerbation, they were more confident to lead on treatment and management decisions.

One pwCF stated that decision making for them was based on whether or not they had experienced the event before. A repeat infection was something they felt they could handle, but they were happy for the clinical team to lead on new infections or treatments that they were unfamiliar with.

For the clinical care team, shared decision making was an important way to ensure pwCF are engaged with their care and understand the impact of their own decisions. The approach for some consultants involves laying out multiple options and explaining the impact or consequences of each, and then leaving the final decision to the individual with CF. One consultant argued that dictating management decisions to teens and adults would likely reduce compliance and adherence to treatment and testing, and should be avoided where possible.

The importance of rapport building

Confidence and trust in the clinical care team appeared to correlate strongly with adherence and compliance with treatment and management regimes in adult pwCF. Due to the chronic nature of the disease, and the ongoing level of interaction involved as part of their care pwCF build a close bond with their care team. As such, most participants felt they were ‘on par’ with many of the clinical care team regarding knowledge of CF.

Building familiarity and rapport with the CF team was particularly important for pwCF with unusual symptoms or specific needs. Participants in the pwCF focus group commented that they all experience CF differently, and it takes considerable time to understand their particular requirements. Thus some stated they would prefer to delay or re-arrange their clinic visits if they weren’t able to be seen by their usual, preferred clinical team.

Impact of managing CF on lifestyle

Multiple pwCF noted the impact that managing their CF had on their life, with one describing it as ‘another full time job’. There was discussion amongst the group about the burden of diagnostics performed at the clinic and the shift towards repeated measure tests at home (such as spirometry). Combined with on-going complex treatment regimens which for some involve using nebulised therapies multiple times a day, participants believed that managing their CF interfered with their daily lives.

The parent of a young person with CF added that the clinic visits were particularly disruptive to their lives and their child’s learning. They understood the need to visit the clinic, but claimed that the care team were usually very busy, and the clinic would be running late. Additionally, the CF team would use the opportunity to obtain other clinical measurements which would normally be taken at a routine visit like, height, weight, and physical checks. These would increase the clinic time, adding to the parent’s anxiety for their child’s health.

One adult with CF had removed themselves from clinical trials due to the added burden placed on them to repeatedly perform specific tests or attend extra clinics for additional measurements. Another supported the statement, adding that frequent testing and referrals would impact their adherence to treatment and monitoring.

Unmet needs in CF infection and exacerbation

Monitoring and re-testing in response to treatment

In paediatric clinics, participants from the clinical team and pwCF focus group noted that cough-swabs and cough-plates were used very frequently in instances where an infection was suspected, and following completion of treatment to determine efficacy.

In the adult setting, there was less focus on re-testing or monitoring during periods of infection. Some pwCF stated that once they had started a course of antibiotics, no further tests were performed during or after the course, with decision making for escalation or de-escalation of treatment guided by clinical signs and symptoms.

Members of adult clinical care teams expressed the need for determining treatment response and treatment efficacy, citing concerns of overtreatment. Clinical focus group members felt that benefit could be derived from shorter antibiotic courses in some instances, and determining effective treatment duration would support antimicrobial stewardship.

Empiric antibiotic use is standard practice, as the reference standard culture techniques take at least three days to provide actionable results, or longer for pathogens like NTM. Clinicians stated that for an exacerbation, they will treat empirically, and with the current standard of diagnostics available, this practice is unlikely to change.

Equity of care

Both the clinical care team and pwCF noted issues with equity of care across the devolved nations and within regions of the UK. PwCF described differences in their care pathways, access to diagnostics and interventions, responsiveness to need, and frequency of clinic visits.

Clinical representatives noted differences in local Trust guidelines, funding streams and allocation of funding, availability of services and staff, and access to treatment options all contributed to inequity of care for pwCF across the UK NHS, with one clinician stating an individual’s care was largely dependent on their postcode.

A participant from the microbiology lab noted significant differences in the preparation and analysis of samples, particularly cough swabs and sputum. This difference was attributed to the size and capacity of the laboratory, as well as individual Trust guidelines. They suggested that additional guidelines for laboratory methods would improve consistency across the UK.

Equity of care was also linked to funding structures between the four UK nations. This also extended to availability of drugs and other interventions across the UK nations, which is dependent on inclusion of these interventions in national formularies in different parts of the country.

More acceptable methods of sampling

The clinical care teams, particularly physiotherapists involved in the collection of sputum samples, stated that sputum sampling was not well received by some pwCF, particularly with the increased use of HEMT, where those on HEMT are increasingly non-productive, and must undergo a less tolerable induced method of sputum sampling.

Collection of induced sputum has impacted clinic times, as the process takes longer and increases the burden of time for the physiotherapists. The reduced acceptability of induced sputum sampling was also highlighted across all focus groups - the potential risk is that fewer infections are detected due to fewer high quality samples being collected.

Cough swabs, while generally more acceptable to pwCF, were less acceptable to clinicians due to their poor accuracy, in particular, low negative predictive value (NPV). Negative predictive value relates to the probability that following a ‘negative’ test result, that individual will not have that condition. A poor NPV means that a negative test is potentially incorrect, and the patient may in fact have the condition being tested for, despite a negative result.

Faster microbiology and bacterial culture testing

For the clinical care team, the time to results for culture-based diagnostics is usually too long to consider delaying the use of empirical antibiotics. Treatment decisions are based primarily on clinical symptoms and choice of antibiotics guided by results from previous cultures. Mycobacteria in particular takes weeks to grow sufficiently in culture for analysis.

In addition to the culture time, antibiotic susceptibility testing also takes time. In acute situations, this delay is unacceptable for clinical decision making, further emphasizing empirical antibiotic use.

Changes to the care pathway, and a move towards postal-based sputum samples was preferred by some participants in the pwCF group to samples collected in clinic, however clinical participants flagged some issues around transporting the samples. One clinical representative noted that they had observed more issues with contamination and false-positive cultures following a period of delays within the postal system. Transit time of posted sputum samples affected the quality and resulting accuracy of culture-based diagnostics.

Clinical representatives agreed that a rapid alternative to culture-based diagnostics would be valuable to delivering timely care. Rapid detection of non-Tuberculosis mycobacteria and Pseudomonas were explicitly referenced as cases where earlier, faster diagnosis would be of value in the care pathway, particularly if there was a quantitative element to the output.

Communication and integration of services

Laboratory staff involved in analysing samples from pwCF described a desperate need for greater communication between laboratory staff and clinical staff. One microbiologist noted that they had been given a seat at their local multi-disciplinary team meetings, which had improved communication and allowed the clinical team and laboratory staff to discuss clinical cases in greater detail. This was not standard practice for many CF centres though.

Another microbiologist stated that samples frequently arrived with inaccurate information, or inappropriate fields completed on accompanying forms to inform the laboratory of what the sample is, and what is required from it. They stated that this leads to inappropriate handling of the sample, failed tests, and potentially incorrect reporting of results.

Improvements to education and training of laboratory staff for the handling and identification of unusual bacterial and fungal infections, particularly those more commonly found in CF, was also considered an unmet need for laboratory staff, along with greater, more consistent standards for the testing of samples. Laboratory staff had noted that non-specialist laboratory staff were more likely to misidentify strains isolated from a pwCF and may inaccurately report findings that clinical staff use to inform their clinical decision making.

Stakeholder preferences in TPP characteristics

All of the focus groups involved a TPP characteristic preference and ranking exercise. After an introduction to the topic, each participant in turn was given time to think about what characteristics they felt were most important when considering current and future diagnostics in CF infection and exacerbation.

Preferences for pwCF

Accuracy and confidence in testing – For participants in the pwCF group, accuracy of a test and the confidence they had in a test were discussed simultaneously. PwCF expressed concerns about at-home testing and ensuring that they were performing the test correctly. Some of them alluded to COVID-19, and the proliferation of at-home nasal-swab lateral flow devices. Reliability is of particular importance to pwCF, as multiple participants stated that if a test failed, it would usually mean that they’d have to re-attend clinic in order for the sample to be collected again, or it would delay the time to results which would inform their treatment. With many pwCF becoming less productive secondary to HEMT, the importance of reliability of testing on increasingly hard to get samples is significant.

Regarding test accuracy, participants noted higher accuracy would be preferable as diagnostics for infection and exacerbation in CF are used to inform treatment and management plans.

Time to results – Participants in the pwCF group stated one of their main unmet needs involved faster time to results, particularly for culture-based methods.

Clinical care team representatives outlined time to results for culture-based methods are typically 2–3 days for a ‘negative’ culture, 3–5 days for a ‘positive’ culture of a fast-growing pathogen like Pseudomonas, but could take 6–8 weeks for some pathogens like NTM and some fungal species can be up to 28 days (e.g. Exophiala) (clinical focus group 1). New infections with either of these bacteria can be a health emergency for a pwCF, usually requiring a treatment regimen intended to eradicate the pathogen. Participants in the pwCF group stated that the delay in time to results generally means they are prescribed antibiotics empirically, or treatment decisions are delayed, thereby increasing the risk of the infection worsening before an intervention takes place.

A faster time to results was considered by pwCF to lead to earlier intervention, and reduced risk of complications.

Many participants in the pwCF group claimed their symptoms would start in their sinuses prior to signs of infection on the lungs. They added that using samples from the sinuses or upper airways could be a way of detecting problematic pathogens before they cause a lung infection.

Acceptability and convenience of diagnostics – While there were mixed views on which specific diagnostic tests were deemed acceptable, there was a preference for convenient and minimally invasive tests.

As discussed in other sections, pwCF preferences for rapid testing was in part linked to the significant impact that CF has on their lives. They wanted to ensure that any tests they did were useful, quick, and had some clinical impact. Many wanted to ‘get on with their lives’ and not have to think about their condition more than necessary.

At-home testing was particularly preferred by the majority of pwCF. The opportunity for postal sputum, was seen as convenient and time saving, and also minimised the risk of cross infection associated with the hospital setting.

Clinical care team preferences

Accuracy, sensitivity and specificity - Multiple clinical representatives stated that accuracy was more important than time to results, and that they would prefer to wait a longer for a more accurate test if the alternative was a quicker test that required additional follow up testing due to an inferior accuracy.

Clinical representatives noted that high accuracy, both in terms of specificity and sensitivity, would be important if a diagnostic is to inform clinical decision making.

Some clinical representatives suggested that sensitivity of the test would be more important than specificity, claiming that over-treatment (in part linked to false positives) may be less severe for a patient than missing an infection (due to false negatives).

Other team members argued that false positives should be avoided, due to the anxiety this can cause for patients, particularly with pathogens like Pseudomonas.

One care team representative offered preferred accuracy values, stating that 70% accuracy would be unhelpful, and 95% accuracy would be reasonable for them to take actions based on the result of that test.

Justification for high test accuracy was provided by one clinical representative. They noted that treatment decisions, including lung transplantation was contingent on the person’s microbiology results. False results would severely impact the patient and their management.

Patient acceptability – Clinical team members considered patient acceptability of significant importance to the overall effectiveness of a novel diagnostic. Compliance is required for optimal management and improved outcomes, and patient acceptability of interventions is a key driver of compliance and adherence.

Along with driving compliance, physiotherapists described the impact that acceptability has on the time spent collecting a sample, and the impact that this has on the relationship and rapport built with the patient.

Time to results –An acceptable ‘time to results’ is vital to informing timely clinical decision making, particularly in the setting, or in cases where pathogens like Pseudomonas are detected for the first time and eradication therapy is required.

A faster time to result was considered to drive informed and appropriate clinical decision making.

Increased reliance on induced sputum collection has impacted the time and efficiency of CF physiotherapists, a faster time to result should therefore also consider the time it takes to collect the sample material required and the time to transport the sample to the laboratory.

Clinicians also considered the time involved in analysing a sample and performing a test, providing examples of cases where non-24 hour services have resulted in delays to clinical management. This delay resulted in a patient continuing a course of antibiotics in the absence of information validating the appropriateness of the therapy.

Clinical relevancy (impact on treatment decisions) – Clinical care teams claimed that a test would only add value to the care pathway if it was utilised in such a way that it could impact clinical decision making.

When questioned further on the topic, one clinician noted that diagnostic tests should inform antibiotic strategies in order to be of clinical benefit as empirical use of antibiotics is commonplace in CF management, an issue noted by both clinical care staff and pwCF.

Integration into the clinical care pathway (staff acceptability) – Disruption to the current care pathway when integrating a novel diagnostic can significantly impact staffing arrangements, cost and clinical decision-making processes, potentially delaying clinical care, or resulting in poor functionality of the care pathway. Clinical representatives stated that it would be important for a novel diagnostic to fit within the current care pathway with as little disruptions as possible.

One clinical representative stated that the logistics of integrating and delivering a novel intervention is often overlooked by scientists and researchers, and would need additional thought to be successfully integrated into clinical practice.

The future of CF care and diagnostics

Participants in the pwCF and clinical focus groups provided opinions on what they would like to see from novel diagnostics and care pathway changes in CF.

Improved sample usage – The clinical care team were keen to see greater use of existing and easy to obtain sample material. With sputum becoming increasingly difficult to obtain, being able to perform multiple tests on a single sample, was considered an important step in the advancement of CF diagnostics. An example of this in practice is with viral testing, where some CF centres require multiple samples for an array of tests, while other centres use one swab for multiple assays. Using a single sample across multiple assays was considered a convenient approach and one viewed as less burdensome for pwCF.

Better staff training for assessment of CF cultures – One microbiology representative highlighted that laboratory staff unfamiliar with handling CF samples would frequently miss atypical or unusual pathogens. With CF centres operating differently in terms of laboratory staffing and training, sample handling may vary between the centres. The representative expressed a desire for greater training and awareness amongst laboratory staff in the detection of CF-relevant pathogens, as well as further guidelines on culturing methods and restrictions on which staff members can report on microbiology results.

New biomarkers for exacerbation and infection – Clinical care representatives stated that they would like to see novel biomarkers for CF infection and exacerbation, and greater research into optimal duration of treatment.

One clinician argued for a biomarker to identify ‘risk of deterioration’ aiming for ‘something better than CRP’ that could be performed at home to distinguish bacterial or viral infection rapidly. They added that understanding if a someone is experiencing a viral or bacterial exacerbation earlier would support antibiotic stewardship and expose fewer pwCF to ineffective treatments (eg. a case of a viral infection and empirical antibiotic use).

Other clinical team members also challenged the standard 14-day antibiotic course, and considered a biomarker for detection and monitoring would allow for tailored antibiotic use for the optimal duration. One clinical care participant added that they had shortened the duration of antibiotic treatment to 7–10 days for many of their patients.

Further research into digital health – Both pwCFs and clinicians referenced ongoing clinical trials including ‘Project Breathe’, which utilises a number of digital health technologies including Bluetooth enabled devices and smart watches to collect information on basic health measures. Spirometry, resting heart rate, and temperature are measured, and will be later analysed with machine learning techniques to determine if ‘passive’ monitoring can flag potential exacerbations and infections before clinical signs and symptoms are evident. Such techniques were considered generally acceptable and low impact for participants in the pwCF group, and may support earlier decision making for CF care teams.

Determining colonisation from infection – An outstanding clinical and research question concerning the separation of colonisation from acute exacerbation was discussed in the clinical focus groups. Clinical experts suggested further research into the microbiome in pwCF with and without a chronic infection would lead to greater understanding of the difference between normal flora, colonisation and acute exacerbation and help better inform clinical decision making.

One clinical team member argued that detection of pathogens alone is not sufficient to determine treatment, and that newer, genetic techniques may be overly sensitive, capable of detecting pathogens that would not actually result in clinical deterioration. Combining pathogen detection with a biomarker of disease activity and or severity may address concerns regarding oversensitivity and provide additional clinical context for better decision making.

Therapeutic drug monitoring- Timely turnaround of drug levels for therapies which require therapeutic drug monitoring to enable safe and effective drug dosing.

Route of drug delivery - While not related to diagnostic tests, participants in the pwCF group shared some insights on the use of nebulised antibiotics delivered via mask. Participants suggested that more research would be welcome to determine the efficacy of nebulised antimicrobials delivered via mask rather than mouthpiece, particularly in the context of sinus disease. Participants reported experiencing sinus symptoms in the lead up to a pulmonary infection, and felt in such instances nebulised antibiotics delivered via mask should be considered.

Study limitations

While the depth and quality of the information gathered in this report is valuable, the small sample numbers involved means that the conclusions drawn in this report may not be representative of the UK NHS or CF population as a whole. Indeed, the pwCF who interact and volunteer in such activities as those utilised in this body of work may not be representative of the general CF population. It is likely that these pwCF are more informed, more interested in medical research, and may be more responsive and compliant in their own management of their condition. Equally, clinical representatives may have research interests or other motivations which drive them to take part in this work and thus, their opinions and conclusions may not be representative of the wider clinical body.

The nature of focus group-based elicitation means that some representatives may overshadow other members of the group, and breadth of opinion may be lost under more generally accepted opinions. This is addressed to some extent by the methodology team directing questions to specific representatives and ensuring opportunities to comment on the conversation topics were available, however it is possible that some representatives may have refrained from sharing their opinion on some topic areas in front of the group.

Not all of the care staff involved in the care of pwCF were included in the focus groups, and only one or two representatives were involved from each discipline, further biasing the opinions gathered.

Despite these limitations, the conclusions drawn from this work formed a valuable step in the generation of a patient-focused TPP for CF infection and exacerbation diagnosis. The methodology team aimed to address the limitations discussed here in their approach to the formal TPP elicitation exercise.