Mixed‐methods evaluation of a nurse‐led allergy clinic model in primary care: Feasibility trial

Abstract Introduction It is now widely acknowledged that there are serious shortcomings in allergy care provision for patients seen in primary care. We sought to assess the feasibility of delivering and evaluating a new nurse‐led allergy service in primary care, measured by recruitment, retention and estimates of the potential impact of the intervention on disease‐specific quality of life. Methods Mixed‐methods evaluation of a nurse‐led primary care‐based allergy clinic in Edinburgh, UK undertaken during the period 2017–2021 with a focus on suspected food allergy and atopic eczema in young children, allergic rhinitis in children and young people, and suspected anaphylaxis in adults. Prior to March 2020, patients were seen face‐to‐face (Phase 1). Due to COVID‐19 pandemic restrictions, recruitment was halted between March–August 2020, and a remote clinic was restarted in September 2020 (Phase 2). Disease‐specific quality of life was measured at baseline and 6–12 weeks post intervention using validated instruments. Quantitative data were descriptively analysed. We undertook interviews with 16 carers/patients and nine healthcare professionals, which were thematically analysed. Results During Phase 1, 426/506 (84%) referred patients met the eligibility criteria; 40/46 (87%) of Phase 2 referrals were eligible. Males and females were recruited in approximately equal numbers. The majority (83%) of referrals were for possible food allergy or anaphylaxis. Complete data were available for 338/426 (79%) patients seen in Phase 1 and 30/40 (75%) in Phase 2. Compared with baseline assessments, there were improvements in disease‐specific quality of life for most categories of patients. Patients/carers and healthcare professionals reported high levels of satisfaction, this being reinforced by the qualitative interviews in which convenience and speed of access to expert opinion, the quality of the consultation, and patient/care empowerment were particularly emphasised. Conclusion This large feasibility trial has demonstrated that it is possible to recruit, deliver and retain individuals into a nurse‐led allergy clinic with both face‐to‐face and remote consultations. Our data indicate that the intervention was considered acceptable to patients/carers and healthcare professionals. The before‐after data of disease‐specific quality of life suggest that the intervention may prove effective, but this now needs to be confirmed through a formal randomised controlled trial. Trial Registration ClinicalTrials.gov reference NCT03826953.


Conclusion:
This large feasibility trial has demonstrated that it is possible to recruit, deliver and retain individuals into a nurse-led allergy clinic with both face-to-face and remote consultations. Our data indicate that the intervention was considered acceptable to patients/carers and healthcare professionals. The before-after data of disease-specific quality of life suggest that the intervention may prove effective, but this now needs to be confirmed through a formal randomised controlled trial.
Trial Registration: ClinicalTrials.gov reference NCT03826953. highlighted what we have known from UK-based studies for a decade, namely that primary care providers have limited training, expertise and confidence in allergy care and that demand for specialist allergy service far outweighs supply capabilities. [2][3][4][5] Primary and secondary care allergy pathways are inadequate, leading to poor referral processes, avoidable delays in management of allergic diseases and poor patient outcomes. 6,7 Specialist allergy care provision is patchy, but where it exists, evidence suggests that there are often unnecessary referrals to secondary care for conditions that could be dealt with in primary care settings. 8 Lack of specialist allergy care provision has been highlighted in the UK for over 15 years. [9][10][11] Jutel et al. 12 suggested that there is a need to prioritise provision of allergy care within community settings rather than specialist settings, given the very large numbers of people now affected by allergic conditions. 4 The 2013 Children's and Young People's Allergy Network Scotland (CYANS) report found that, across Scotland, primary care practitioners did not feel they have the skills or knowledge to provide good quality allergy care, especially around diagnostic testing for allergy, and the ability to interpret the results. 13 A pilot study by Levy et al, 14 run by a specialist allergy nurse and a GP with a special interest in respiratory disease and allergy, showed that a primary care intervention for allergy could effectively deal with the majority of cases of allergy seen in primary care, resulting in a reduction in inappropriate referrals into secondary care, an increase in selfsupported care for patients, and a saving in costs. These data were however from an uncontrolled study and therefore need to be interpreted with caution. A more recent study found that a large percentage of referrals, that would otherwise have been seen in secondary care, could adequately be dealt with in primary care by a practitioner with a specialist interest in allergy. 15 A systematic review supports the need for alternative models of allergy care provision, 1 however none of these models were tested with a control arm. Systematic reviews of diseasespecific nurse-led clinics have shown high levels of patient satisfaction, however there remains a need for more robust studies. 16,17 Primary care based nurse-led allergy clinics offer a new model of care for people with allergic conditions. This paper describes the feasibility of delivering and evaluating a nurse-led allergy service to patients in Edinburgh, Scotland.

| Overview of methods
Our feasibility trial protocol is published elsewhere. 18 We undertook a mixed-methods study, through general practices in Edinburgh, Scotland between July 2017 and February 2021. General practices in South East and South West Edinburgh were invited to refer patients with specific suspected or clinician diagnosed allergic problems to a specialist nurse-led allergy clinic (intervention) located in three hub practices (Phase 1). Referrals were paused in March 2020 due to COVID-19 restrictions, and following an ethics amendment, were restarted in September 2020 using a secure remote model of care (Near Me -an NHS Scotland approved video consulting service) to the end of February 2021 (Phase 2).

| The primary outcomes were
� Recruitment of practices to facilitate establishing the new service and make referrals.
� Referral and consultation rates for nurse-led allergy clinic and retention rates.
� The change in disease-specific quality of life questionnaires between baseline and 6-12 weeks post intervention.

| Ethical considerations and permissions
We obtained written informed consent from patients and for young children from their parents. Children with capacity provided informed written consent and assent was obtained from their parents/carers. All participants were given a study ID and all data were anonymous to the research team.

| Eligibility criteria
Patients were referred to the nurse-led allergy clinic via secure NHS email. The eligibility criteria are summarised below.

| Inclusion criteria
� Children aged <36 months with suspected food allergy.
� Children aged <36 months with moderate-to-severe atopic eczema not responding to standard treatment.
� Children and young people up to 16 years of age with suspected allergic rhinitis symptoms not responsive to a combination of oral antihistamines and nasal steroids.
� Young people and adults (from 16 years of age) with a history of anaphylaxis or suspected anaphylaxis.
� Able to give informed consent/assent for children under 16 years.

| Exclusion criteria
� Children aged <36 months with suspected or confirmed non-IgEmediated food allergy presenting primarily with gastrointestinal symptoms.
� Mild-to-moderate atopic eczema without any obvious allergic trigger.
� Unable to give informed consent/assent.

| Recruitment
Primary care-based health care professionals (HCPs) were invited to complete a referral proforma for eligible patients via secure NHS email sent to the allergy nurse. If the referred patient met the inclusion criteria the patient/parent/carer were sent a patient information sheet and consent/assent form prior to being seen in the clinic.
If they did not wish to take part in the study they were referred back to their HCP. Suitable patients were contacted by telephone or email and offered an appointment, where consent was taken, and baseline data were collected.

| Intervention
The intervention is described in detail in the feasibility trial protocol. 18 In summary, this was a nurse-led allergy service in primary care.
Nurses with a postgraduate allergy qualification and extensive secondary care experience were supported by a team of local specialist services, including paediatric and adult allergy, dermatology, ear, nose and throat (ENT) and respiratory medicine. Trial nurses took an allergy-focussed history and clinical examination, with investigations as considered clinically appropriate (skin prick tests and blood samples) and provided a diagnosis, management advice and relevant education. A letter summarising the consultation was sent to the referring HCP including details of any medication(s) to prescribe.
Patients were then discharged back to the care of their referring HCP.
In some instances, further onward referral to secondary care services was required.

| Recruitment to semi-structured interviews
Patients, or parents/carers of young children, were invited to be interviewed and indicated on the consent form if they were happy to be approached by the qualitative researcher. If patients consented, their details were passed to the researcher who contacted them by telephone to arrange an interview at a time and place to suit them.
Referring HCPs were asked by the allergy nurses if they would take part in an interview and, if so, their contact details were passed on to the qualitative researcher, who contacted them by telephone to arrange an interview.

| Interviews
Patient/parent/carer participants were asked to give a description of their experience of their allergy in a semi-structured interview conducted either face-to-face, by telephone or by email dependent on participant's choice. They were also asked about their clinic visit, and their views on current provision of allergy care.
Interviews lasted between 15 and 30 min, and were audiorecorded.
HCPs were asked about their experience of referring into the new clinic, how it impacted their workload, and their views on the provision of allergy care.

| Data management and analysis
Disease-specific quality of life (QoL) was measured at baseline and 6-  interview recordings were transcribed, checked, and anonymised.
The interview transcripts were then analysed thematically, framed around both the research questions and themes arising from the data.
To examine socio-economic inequality in the patient population Scottish Index of Multiple Deprivation (SIMD) 2020 was used, which was based on the 2020 census population. 25 In Scotland in 2020 there were 6976 similarly sized areas of approximately 700 people, called data zones, where each had an allocated deprivation rank, the SIMD. SIMD rank for a data zone represents a comparative magnitude of deprivation across seven domains: income, employment, education, health, access to services, crime and housing. We used the SIMD ranks organised as quintiles (20% of the data zones), where quintile 1 represented the 20% of the data zones which were most deprived and quintile 5 the 20% data zones which were least deprived. SIMD quintile was obtained for each patient who attended the clinic by looking up their postcode of residence in the SIMD postcode lookup file. 26 Edinburgh Southern had 96 data zones in 2020, of which one data zone was in the most deprived quintile. That most deprived data zone represented 1.04% of Edinburgh Southern and 0.07% of Scotland. 26

| Referrals
Of the 37 practices in the two referring localities, South East and South West Edinburgh, 35 referred patients to the allergy clinic.
After assessing a total of 506 referred patients against the inclusion/ In Phase 1, GPs referred most patients to the clinic (n = 320, 69%), followed by health visitors (n = 129, 28%) and practice or community nurses (n = 17, 3%). In Phase 2, both GPs and health visitors referred 20 patients.

| Demographics of referred patients
In both phases of the study, similar numbers of males and females were referred, with similar percentages of adults and children in The majority (49% in Phase 1% and 53% in Phase 2) of patients referred to the clinic were in SIMD quintile 5, representing the 20% least deprived areas in Scotland (Figure 2).

Patient satisfaction survey
Immediately after their 6-12 weeks follow-up appointment, 371 patients/parents/carers completed a satisfaction (Table 5). Between 64% and 91% of participants were very happy 5 with the individual aspects, with the highest scoring domain being the general information given by the nurse.
When asked about self-assessed improvement of their allergy, 42% (n = 155) of patients said their allergy was much better since attending the allergy nurse clinic, 40% (n = 148) said they were slightly better, and 16% (n = 59) said there was no change. Less than 2% said their allergy was worse or slightly worse. The majority of patients (92%) said there were able to keep to their personal management/treatment plan advised by the allergy nurse.

HCP satisfaction survey
Twenty one HCPs who had made referrals into the service completed a satisfaction survey (12GPs, 6 Health visitors, 1 practice nurse, 1 community nursery nurse, 1 other). The majority of HCPs had referred between 1 and 10 patients into the clinic, two had referred 11-20 and 2 had referred 21-30 patients. Most respondents found the referral process very easy, but requested as the main improvement to set up the clinic on the Scottish Care Information (SCI) Gateway, a national system that integrates primary and secondary care systems using highly secure Internet technology.
In a satisfaction survey, HCPs scored all aspects of their patients care in the nurse-led allergy clinics very highly with mean scores between 4.3 and 4.7 (scale 1-5, not happy -very happy) ( Table 6).
When asked about the best aspect of the clinic, HCP free text responses focussed on ease of access to the clinic, quick appointment times and reduced waiting times, as well as seeing an experienced HCP and having time for discussion (See Box 1 for full list).
When asked about the worst aspects of the clinic, the response were limited to requests for the use of SCI-Gateway for referrals, discharge summaries including prescribing information, and for shorter waiting times for clinic appointments.

Box 1: HCP free text responses to satisfaction survey
♦ Ease of access, quick appointment turnaround and reduced waiting time.
♦ being seen quickly by knowledgeable professionals and having more time.
♦ parents appreciate the ability to spend time discussing food exclusion/reintroduction. ♦ much needed allergy service that is otherwise lacking. ♦ prompt access for a difficult problem.
♦ useful tool for patients with anaphylaxis. This is a big gap in referral pathways otherwise.  that being able to spend time with an experienced specialist nurse led to improvements in their knowledge about their allergies, a greater feeling of control over them, and a consequent improvement in their QoL. Participants liked the fact that the consultation focused not just on their physical symptoms, but also on practical ways to minimise their impact on their daily lives, and on ways to cope with the anxieties raised by these symptoms. All said that they felt more supported, and appreciated the convenience of a local clinic. 1 teenage boy (interviewed with his mother) with anaphylaxis.

Box 2: Description of patients interviewed by allergic condition and sex
1 teenage girl with food allergies.

| Length of the consultation
The participants commented that the consultation lasted longer than they had expected, and that they therefore had time to discuss all the aspects they were concerned about. Many contrasted this with their previous, and often repeated, visits to see a GP where they felt rushed, and where their concerns had not really been attended to as they had hoped. Several said that they felt as if they were bothering the GP with concerns that were perhaps minor to him/her, but important to them, and came away from GP consultations still feeling worried and anxious.
The nurse was like really, really helpful and she was really sweet as well. She made me feel really welcome, and she wasn't in a hurry, which I think is a really

| Patient empowerment
When participants talked in general about the impact of their visit to the clinic, the aspects they found most helpful were two-fold: practical and emotional. Practically, participants valued the increased knowledge and understanding of their allergies, and the information on how to cope with them, that they gained from the consultation.

| HCP interviews
Nine HCPs were interviewed, five referring HCPs (2 HV, 3 GPs), one practice manager and the two specialist allergy nurses, one of whom took part in two interviews, one in the middle of the study and one at the end.
All the HCPs interviewed were very positive about the service. In terms of improvements to the service, HPCs felt that although the referral system was straightforward and easy to use, it would be improved by implementing the more recognised referral pathway via SCI Gateway.
If it could be continued, then it could be put on SCI-Gateway in this clinic and that will be even better. (GP2) In terms of improvements I would say having a for- There are some limitations to our work that need to be considered. Chief amongst these is that we did not get complete follow-up data on all participants. This is important, because participants who did not complete the post-intervention follow-up assessments may have been less likely to have valued the intervention and/or improved, potentially introducing bias. It is also important to note that although we had four clinical conditions that we sought to improve, over 80% of referrals were for suspected food allergy and anaphylaxis. Numbers of participants recruited with severe childhood eczema or seasonal allergic rhinitis unresponsive to pharmacotherapy were therefore small.
Moving forward, these findings provide a firm basis to move to a pilot randomised controlled trial followed by a possible definitive randomised controlled trial. Given the referral and recruitment patterns seen, it would seem prudent to focus follow-on work on food allergy and anaphylaxis; this would also mean the need to work with fewer disease-specific QoL measures. Although retention was good overall, there is a need to improve this, which is probably best achieved by emphasising the importance of these final study outcome measures at the time of recruitment and also by making it possible to complete questionnaires through a range of channels, including in person, over the phone or over the Internet through a smart phone, tablet or personal computer. Future work also needs to involve a formal health economic evaluation from the perspectives of the NHS.

| CONCLUSION
We have demonstrated the acceptability, feasibility and potential effectiveness of a nurse-led primary care-based allergy clinic for patients with suspected/confirmed food allergy, eczema, allergic rhinitis and anaphylaxis. There is now a need to build on this work to formally evaluate the effectiveness of the intervention through a pilot and then definitive randomised controlled trial.

AUTHOR CONTRIBUTIONS
Aziz Sheikh and Jurgen Schwarze conceived and led this study and together with Lynn Morrice secured funding for this project. Vicky

Hammersley was the project manager. Margaret Kelman and Susan
Harley were the trial nurses. Mome Mukerjhee supported quantitative data analysis. Marilyn Kendall undertook the interviews and led the qualitative evaluation. All authors contributed to the writing of the manuscript.