Feasibility trial of a digital self-management intervention ‘My Breathing Matters’ to improve asthma-related quality of life for UK primary care patients with asthma

Objective To assess the feasibility of a randomised controlled trial (RCT) and acceptability of an asthma self-management digital intervention to improve asthma-specific quality of life in comparison with usual care. Design and setting A two-arm feasibility RCT conducted across seven general practices in Wessex, UK. Participants Primary care patients with asthma aged 18 years and over, with impaired asthma-specific quality of life and access to the internet. Interventions ‘My Breathing Matters’ (MBM) is a digital asthma self-management intervention designed using theory, evidence and person-based approaches to provide tailored support for both pharmacological and non-pharmacological management of asthma symptoms. Outcomes The primary outcome was the feasibility of the trial design, including recruitment, adherence and retention at follow-up (3 and 12 months). Secondary outcomes were the feasibility and effect sizes of specific trial measures including asthma-specific quality of life and asthma control. Results Primary outcomes: 88 patients were recruited (target 80). At 3-month follow-up, two patients withdrew and six did not complete outcome measures. At 12 months, two withdrew and four did not complete outcome measures. 36/44 patients in the intervention group engaged with MBM (median of 4 logins, range 0–25, IQR 8). Consistent trends were observed to improvements in asthma-related patient-reported outcome measures. Conclusions This study demonstrated the feasibility and acceptability of a definitive RCT that is required to determine the clinical and cost-effectiveness of a digital asthma self-management intervention. Trial registration number ISRCTN15698435.


Introduction
Asthma prevalence in the UK is amongst the highest in the world at nearly 6% of the UK adult population, comprising 5.4 million people, with most managed in primary care. Although hospital admission and mortality rates for asthma improved from 1970 to 2000, these improvements have since stalled [1]. Surveys of asthma symptoms and health status impairment show that sub-optimal control is common and that the majority of people with asthma in the UK frequently experience potentially avoidable symptoms and quality of life impairment [2].
Proactive self-management of asthma has been convincingly shown to improve clinical outcomes and have been advocated in guidelines for 25 year [3]. Guidelines are not always well implemented [4] and consequently some people with asthma do not receive evidence-based interventions which are known to impact positively on outcomes. Recent large-scale systematic reviews demonstrated that supported asthma self-management can reduce healthcare utilisation and increase asthma control, without increasing healthcare costs [5,6]. For example, people with asthma without a management plan are four times more likely to have an asthma attack needing emergency care in hospital, yet only 44% of people with asthma in the UK report having a self-management plan [7]. Self-management recommendations for asthma have also encompassed non-pharmacological strategies to improve control. These include lifestyle interventions, such as smoking cessation, allergen avoidance, weight reduction in those with obesity, and breathing retraining interventions [8].
Digital interventions (DIs) are increasingly recognised as a possible approach to achieve the aims of supporting chronic diseases such as asthma. DIs can be convenient, easily accessed and may provide cost-effective tools by automating routine aspects of patient education, monitoring and support [9]. There is accumulating evidence that DIs are feasible and may be effective in the context of asthma. The SMASHING trial compared usual care with web-based educational resources, self-monitoring and automated feedback on medication titration, plus some group and email nurse support for patients with asthma. After 12 months the intervention group had better quality of life and lung function and more symptom-free days, at no extra cost [10]. The RAISIN pilot trial indicated that self-management interventions that included nonpharmacological (behavioural and psychological) components could be effective at improving quality of life and asthma control, with improvements to 'reach' and response rate, by catering to patients with mild asthma but impaired quality of life [11]. A recent systematic review and meta-analysis indicated that while self-management DIs may be able to improve asthma control and reduce asthma-related quality of life impairment, larger confirmatory trials are required [12].
In the current randomised controlled feasibility trial, we developed and evaluated a digital selfmanagement intervention, that incorporated pharmacological and non-pharmacological selfmanagement support for adults in primary care with impaired asthma-specific quality of life ('My Breathing Matters'; MBM), using evidence, theory and person-based approaches [13] and in line with Medical Research Council (MRC) guidance for developing and evaluating complex interventions [14].

Aim
The aim of the MBM study was to assess the feasibility of a trial to evaluate a digital intervention in primary care to improve clinical outcomes (such as quality of life, health resource use, lung function) of people with asthma, in comparison to usual care (with provision of standard patient information materials produced by the charity Asthma UK).

Research Objectives
1. To assess feasibility of trial procedures including recruitment strategy, eligibility criteria, consent, withdrawal, randomisation and blinding.
2. To assess feasibility of the MBM digital intervention including usage and engagement.
3. To assess feasibility of data analysis, including data collection, data quality and management of trial data across trial endpoint measures to inform sample size calculations for a larger phase 3 randomised controlled trial.  records screened by GP to remove ineligible participants. Invitation letters, study information sheets, consent forms screening questionnaires (Mini Asthma Quality of Life Scale; AQLQ [16]) and freepost return envelopes were posted to participants who returned them if they were interested in taking part. Patients who met screening criteria (AQLQ score of less than 5.5) were contacted by research team staff and attended a baseline appointment at their practice with a trained research nurse. Recruitment began in March 2017 and was completed in August 2017.

Sample size
The target for this trial was to recruit 80 patients overall (40 per arm), in order to assess primary feasibility outcomes and to assess intervention engagement and acceptability.

Randomisation and Blinding
After completing outcome measures at their baseline appointment participants were randomised (block randomisation stratified by an average primary care AQLQ score [4.3] taken from a previous trial using the same inclusion criteria [8]). Information packs were given to participants after randomisation with instructions on signing up to MBM (if randomised to intervention group) or just usual care materials (if randomised to control). Research nurses conducting baseline appointments were blinded throughout the study until the final questionnaire which was only delivered to the intervention group.

Interventions described according to Tidier checklist[17]
Intervention Group: Usual Care with MBM and Asthma UK booklet.
Patients in the intervention group continued to receive usual care but were also given a code that allowed free unlimited access to MBM. MBM is a digital asthma self-management intervention that supports asthma self-management using both pharmacological and nonpharmacological approaches, developed using the LifeGuide Software [18]. A demonstration version of the intervention is available here: http://www.mybreathingmatters.co.uk. After signing up and completing quality-of-life related self-monitoring questions, patients were Well with Asthma', an asthma self-management intervention that previously demonstrated feasibility for self-management [11,19] and was developed in collaboration with people with asthma and with input from Asthma UK (a national asthma charity). This section was designed to answer common concerns about medication, incorporating a strategy described as 'the 4- week medication challenge' that encouraged participants to realise the benefits of adherence to regular medication, by self-monitoring their symptoms during 4 weeks of continuous inhaler use. The non-pharmacological support included sections on a number of strategies to improve asthma control, such as Breathing Retraining, stress reduction, and additional healthy lifestyle resources (physical activity, weight reduction, hand hygiene and smoking cessation). Optional nurse support was available by Asthma UK who provide a dedicated nurse helpline that was advertised through the intervention.
The intervention was developed using the person-based approach [20] which places patients at the heart of the development process. Evidence from primary mixed methods research (such as [19]) and qualitative and quantitative reviews was used to develop guiding principles. A prototype intervention was piloted using 'think aloud' interview studies in which patients with asthma used prototype versions of the MBM website and provided feedback on intervention acceptability and feasibility as they used it. In 46 interviews with 30 patients, the intervention was iteratively modified and updated to address patient feedback until participants indicated no further modification was required, confirming the intervention was as acceptable and engaging as possible (for more details on this process see [21]).
Due to the digital nature of the intervention, participants could engage with components of the intervention as much or as little as they wished. Tailored advice was offered according to participants' preference to find out more about pharmacological or non-pharmacological selfmanagement techniques (patients selected a check box option of "I'd like to find out more about how my asthma could be helped by i) making the most of my asthma medicine, or ii) 'nonmedicine' ways to help my breathing."), with automated reminders whenever patients had not accessed the intervention for several weeks, or when content was made available that they had not previously seen. The intervention was not modified during the study.
Intervention usage was monitored through digital usage metrics (reported below).Nonengagement with the website was not addressed, in line with the pragmatic nature of the feasibility study. Participants were sent one email and received one phone call in which they were offered technical support if they had not logged onto the intervention at all for one month following their baseline appointment.

Control Group: Usual Care with Asthma UK Booklet
To provide 'good quality' usual care to participants allocated to this arm, as well as usual care from their practice, participants were given an Asthma UK booklet 'Live Well with Asthma' at their baseline appointment. The booklet was created by a multidisciplinary team and expert patients, and aimed to provide essential information and advice to enable effective selfmanagement to occur. It is available to anyone via the Asthma UK website [22]. The booklet was provided in hard copy and provided information about asthma symptoms and triggers, medication adherence and usage techniques, PAAPs and support from families. The booklet also advertised the Asthma UK support line. Booklet usage was not monitored.

Outcome Measures
In line with objectives, trial outcomes are reported below as 1) feasibility outcome measures, 2) intervention usage outcome measures, and 3) trial endpoint measures to inform a larger trial. -Health-related quality of life. Measured using the EQ-5D-5L [24] in which participants select their functioning level across five dimensions (immobility, self-care, usual activities, pain and discomfort, and anxiety and depression) on a five-point scale, with higher scores indicating greater problems with functioning.
-Health-related capability. Measured using the Icepop Capability measure for adults (ICECAP-A [25]), in which participants select their capability across five dimensions (stability, enjoyment, achievement, attachment and autonomy) on a four-point scale, with higher scores indicating better capability.

11
-Anxiety and depression. Measured using the Hospital Anxiety and Depression Scale (HADS [26]), a 12-item questionnaire in which higher scores on depression and anxiety subscales indicate higher anxiety across two subscales (anxiety and depression).
-Enablement, measured using a modified version of the patient enablement instrument (PEI [27]) that has been validated in previous RCTs [28]. The modified PEI is a 7-point scale consisting of 6 items, in which higher scores indicate more enablement.
-Patient satisfaction was measured by asking patients whether they saw any benefits or disadvantages to using MBM, and whether they would recommend it to friends and family based on the NHS Friends and Family Test (FFT [29]).
-Patient burden was measured using a specifically developed questionnaire exploring time and costs via self-report based on Burden of Treatment Theory [30]. The questionnaire consisted of 4 questions with descriptive responses that explored whether new programmes were signed up to (such as gym membership, yoga/meditation), and the financial burden of doing so (see Appendix 1).
Physiological measures of lung function were taken at baseline and 12 month appointments: forced expiratory volume (FEV 1 ), ratio of forced expiratory volume to forced vital capacity (FEV 1 /FVC) and peak expiratory flow rate (PEFR).
At 12-month follow-up, we also monitored health resource use GP consultations, A&E visits, hospital admissions, asthma medication use and use of antibiotics for chest infections using GP practice patient notes.
Healthcare utilisation data were collected via retrospective notes review conducted by practice staff. Staff were provided with a template for reviewing data, and an instruction manual to ensure correct data were provided. Initial notes reviews were completed within two months from completion of primary data collection. Nine patients (10% of the total patients in each practice) were also reviewed by research nurses to assess data quality. Asthma UK was involved in the initial project proposal and supported the project throughout.

Patient and Public Involvement
Patient and public representatives (recruited with help from Asthma UK) participated in intervention development (providing feedback on prototype versions of the intervention, attending study management meetings, helping to develop trial materials and procedures, and discussing responses to participant feedback). Asthma UK are involved in dissemination of this research and ongoing projects related to the research.

Data Analysis
Primary analysis of the study was a description of key feasibility outcomes including patient eligibility, recruitment rates, withdrawals, 3 and 12-month follow-up response rates and digital intervention usage.
Descriptive statistics were used to identify any floor or ceiling effects. For continuous measures, means, standard deviations (SD) and 95% CIs were reported at baseline, 3-month and 12month for each group, as well as for the sample as whole.
Exploratory analysis explored group differences in continuous outcome measures (AQLQ, ACQ, HADS, PEI, ICECAP-A, EQ-5D-5L) using linear regression models. Participants were analysed in the group to which they had been randomised and comprised complete cases only.
Proportions of patients achieving a minimal clinically important difference (MCID) was described for asthma quality of life (AQLQ).
Sensitivity analysis explored missing data at 3 and 12 months.
Health economic analysis was descriptive, reporting estimates of cost and outcomes measures and baseline and follow-up. The completeness and suitability of EQ-5D-5L and ICECAP were compared as was the appropriateness of the resource use, and time and cost tools developed for the study.
Intervention engagement was descriptive.

Recruitment and retention
Six practices were initially recruited and after monitoring recruitment rates a seventh practice added. In this additional practice only half the list (randomly selected) were offered participation in the study to avoid over recruiting. Across the 7 practices, 68478 patients were assessed for eligibility with 3199 meeting initial eligibility criteria (asthma diagnosis, >1 asthma medicine prescription in last 12 months, screened by practice). 266 patients completed postal screening measures before the recruitment period finished, of whom 125 were eligible to take part (impaired asthma-related quality of life, AQLQ score less than 5.5). Ninety patients responded to further contact. Two patients did not attend their baseline appointment leaving a final sample of 88 patients (intervention N = 44, usual care N = 44) who were recruited into the study (13.5 per practice) and were randomised over a 5 month period. Figure 1 presents the study CONSORT diagram.
During the study, 2 patients withdrew before 3-month follow-up and 2 before 12-month followup. All were in the intervention group. Patients withdrew for several reasons including lack of time (N = 1), illness (N = 1), death of family member (N = 1), and lack of perceived benefit (N = 1). 3 of 4 participants who withdrew had used the intervention; one had not. One participant was withdrawn from the study prior to 12-month follow-up they were no longer eligible (i.e. they were referred to secondary care).
76% attended a baseline appointment and provided secondary clinical data (67/88). Four (5%) patients did not complete 12 month follow-up measures but did not withdraw (

Patient characteristics
Demographics and baseline characteristics of participants are presented in Table 1 and were reasonably well-balanced between arms across all measures.   Table 1 compares those lost to follow-up to those who remained in the study at 3 months.
Those lost to follow up were slightly more likely to be female, have a higher BMI, a longer time since diagnosis, a lower AQLQ score, a higher HADS-A and HADS-D score and to be from a more deprived postcode.

Trial Endpoint Measures
The full data of the trial endpoints is set out in Table 2.
Both the intervention group and control group improved from baseline to 3-month and 12month follow-up, with numerically (but not statistically significantly) larger improvements in the asthma-related patient reported outcomes measuring quality of life and symptom control (AQLQ and AQC) at both time points; one or both these measure are anticipated to to be the primary outcome of a subsequent fully powered study.

Adverse Events
Adverse events were reported by GPs and nurses who contacted the study team to report both adverse and serious adverse events. Nine adverse events were reported (intervention N = 6, usual care N = 3). These were assessed by research team clinicians and all were considered unlikely to be related to the study. Three were related to participant asthma (asthma exacerbation not leading to hospital admission, upper respiratory tract infection, sinusitis).
Three serious adverse events were reported (intervention N = 2, usual care N = 1). These were considered unlikely to be related to the study and the condition (atrial fibrillation, open distal radius fracture, cardioversion).

Healthcare Utilisation Outcomes
Data were collected from retrospective notes reviews (conducted by practice nurses) from 83 participants, reported in Table 3. Data was collected from 84 practices for 7 participants, with 4 participants from 1 practice incomplete. The data quality check, and subsequent examination by research team clinicians (MT) found that reviews completed by the practice nurses varied substantially in quality with varied levels of detail, and the quality of data achieved in this way was insufficient for a health economic analysis.  Comparisons between group count data were reported using a negative binomial model but given the issues with the reliability of the data, should be interpreted cautiously. The prescription rate was approximately 8% higher in for both SABA (IRR 1.08, 95% CI 0.82, 1.43) and ICS (IRR 1.08, 95% CI 0.86, 1.35) in the control group compared to the intervention group.
Both groups had a low number of prescriptions for oral steroids, oral steroids and antibiotics, with only 15 prescriptions in total for either of these medications, making between group comparisons unreliable.
Due to unreliability of data, frequency of GP consultations, A&E admissions and hospitalisations have not been reported.

Health Economic Outcomes
Both EQ-5D-5L and ICECAP-A had the same completion rates as other secondary measures completed at follow up (see Table 2).
Patients reported several programmes across both groups including gym, walking, yoga, sewing, language courses, physio and signing (see Appendix 2). There were no substantial differences in terms of numbers or costs although the sample size was small.

Discussion
Findings from our randomised controlled feasibility trial demonstrate that a full-size confirmatory trial to confirm effectiveness of MBM, a digital self-management intervention for adults in primary care with asthma is feasible and acceptable. There were also trends to improved asthma control and quality of life in our underpowered sample, so supporting the need for a definitive fully-powered study. Our recruitment procedures recruited a specific patient sample (those impaired asthma-specific quality of life) to target from a range of urban and rural practices. Our sample varied in age with a relatively high mean (56 years) indicating that our digital intervention can provide benefit to older adults. Both male and female adults were wellrepresented in our trial. A notable proportion of our sample was obese (41%), in line with previous findings [11]. Given that obesity is a risk factor for asthma, a larger trial could further improve effectiveness by providing more specific behavioural content for obese adults with asthma (such as tailored content to increase motivation to use weight-loss related lifestyle components in obese patients). Our sample was also predominantly white. Underrepresentation of minority ethnic groups in medical research in the UK is an ongoing issue [12] and should be addressed in recruitment procedures in the full trial.
Trial procedures were completed effectively, supporting the feasibility of a full trial. All patients who completed baseline measures were randomised. Completion of measures was good at both follow-up points (3-month via post and 12-month at participants' practice). Where participants were not able to attend a follow-up appointment at practices, they were satisfactorily followed up via post or telephone for main trial measures. 8 patients were lost to follow-up (4 withdrew and 4 no longer responded to attempts to contact them). Notably, these patients were all in the intervention group; it may be that the control group were more likely to maintain contact as they were only able to access the intervention upon completion of 12-month follow up measures. Although loss to follow-up is low, it is important to consider whether that loss is differential. Those lost to follow-up were more likely to be more socio-economically deprived, female, have a higher BMI, a longer time since diagnosis and a higher HADS-A and HADS-D score. It is possible that these patients would benefit from using MBM more than most, and therefore we have proposed several ways to further increase trial efficacy. Automatic email intervention registration at baseline (patients cannot attend baseline appointment without enrolling on the intervention) would increase initial engagement and engagement with trial procedures throughout duration of study. Online questionnaire completion during screening process would i) screen patients who are unable to interact with online trial/intervention and  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   BMJ Open   22 therefore unable to benefit from the intervention (feasibility trial estimate = 2%), and ii) streamline baseline/follow-up procedures.
Both health economic outcomes had high completion rates but did not suggest substantive change, similar to EQ-5D measures in previous non-pharmacological self-management trials (such as [8]). It is possible that an alternative measure such as the Short Form 12-item Survey (SF-12 [31]) in which participants consider the previous two weeks (whereas in the EQ-5D they consider the immediate present) may be better suited to measure small yet valuable changes in well-being over a full trial. Our detailed mixed-methods process analysis explored issues of trial acceptability in more detail, and will be reported in a subsequent paper. A full trial of this nonpharmacological intervention should accurately capture 'non-medical' costs (such as gym membership) that are likely to impact disease-specific quality of life, as well as medical costs that would be affected by changes in healthcare utilisation.
Healthcare utilisation data were collected by practice nurses whose main role was to provide usual clinical care at the practices, using a manual to guide data collection, rather than by trained research nurses, and our quality check demonstrated that the data collection process used was unreliable some centres. We conclude that in a full subsequent study, these data should be collected from the medical record by a trained member of the study team (such as a trained research nurse), as has been successfully used in previous studies [8].
Engagement with the intervention was increased compared to a previous similar digital asthma self-management interventions [18] at initial sign up (82% vs 76%) as well as maintaining a higher number of ongoing engagement throughout the follow up period (median 3 additional log ins vs. 1), although our study used a broad primary care population while the RAISIN protocol primarily recruited from areas of high deprivation. This finding demonstrates that the use of the person-based approach to develop the intervention resulted in an intervention that was acceptable and engaging to patients, even using a pragmatic methodology in which patients  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   BMJ Open   23 self-registered at home instead of being registered by a GP during their baseline appointment.
Participants accessed both pharmacological and non-pharmacological self-management content.
Estimates of effect size demonstrated that participants who received the intervention and completed follow-up measures showed improved and clinically relevant quality of life and asthma control. The order of magnitude of the mean between-group improvements in the patient reported measures of control (ACQ) and asthma-related QOL (AQLQ), although not statistically significant with the sample size of this feasibility study, was comparable to that reported in controlled studies of pharmacological [32] and non-pharmacological [8] interventions in asthma, and so justify a fully powered study.
There was no suggestion of an effect on physiological measures of lung function. These results are in line with previous studies of behavioural self-management interventions in primary care adults with asthma (such as BREATHE, RAISIN), and demonstrate the importance of interventions targeting outcomes that incorporate elements of functional wellbeing (disease specific quality of life, subjective symptoms), rather than solely focusing on objective, physiological measures that are not correlated with quality of life.
The effectiveness of our intervention could be further increased according to findings from our process evaluation. This analysis, which will be reported separately, broadly agrees with previous research [15] in finding that many patients consider their asthma to be 'wellcontrolled' despite having important levels of symptoms and quality of life impairment on validated questionnaire. This implies that many people had become accustomed to their ongoing symptoms and had altered their life to try to reduce their impact, using denial as a coping mechanism. As a consequence, the means of appropriately targeting and framing selfmanagement interventions should be carefully considered in future work, focussing on maintaining good health rather than improving poor health. Some of our findings (such as the association between quality of life improvement and ongoing intervention engagement) demonstrate that framing content as positive and not focusing on illness -for example 'How to  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   BMJ Open   24 keep your breathing healthy' rather than 'How to reduce asthma symptoms' may lead to an acceptable, engaging intervention that benefits this patient group.
There were some limitations to this small feasibility study. Although our researchers and statisticians were blind to group allocation, patients would have known that they were allocated to the intervention rather than the usual care control. This is common in complex behavioural interventions. Furthermore, although we endeavoured to recruit participants across a broad demographic range, the reach of our intervention could be improved. While the reach of digital interventions improves as digital literacy increases nationally, care must be taken to ensure that 'digital transformation' of NHS services does not entrench healthcare inequality, by facilitating a 'digital divide' that fails to provide adequate health and social care to those who do not have the digital skills to benefit.

Conclusion
Our findings demonstrate the feasibility of a new digital self-management intervention for asthma (MBM). Using the person-based approach to intervention development means that MBM is both acceptable and engaging for adults with asthma in primary care. MBM reflects the varied experiences of people with asthma, by including both non-pharmacological and pharmacological components. Our data support the feasibility of moving towards a fullypowered RCT, with only minor modifications to some trial procedures required.

Protocol Information
This protocol describes the My Breathing Matters trial and provides information about procedures for entering participants. The protocol should not be used as a guide for the treatment of other participants; every care was taken in its drafting, but corrections or amendments may be necessary. These will be circulated to investigators in the trial, but sites entering participants for the first time are advised to contact the Centre for Applications of Health Psychology to confirm they have the most recent version.

Compliance
This trial will adhere to the principles outlined in the International Conference on Harmonisation Good Clinical Practice (ICH GCP) guidelines. It will be conducted in compliance with the protocol, the Data Protection Act and all other regulatory requirements, as appropriate.

Impaired asthma-related health status (Asthma Quality of Life
Questionnaire score of <5.5) 5. Informed consent 6. Able to access the internet and understand written english.
Exclusion Criteria: 1. Asthma judged at the baseline assessment to be dangerously unstable and in need of urgent medical review (if unstable asthma is found, the patient will be referred back to usual primary care clinician for review) 2. Terminal disease or other condition which in the opinion of the family doctor makes them inappropriate to take part 3. Diagnosed with 'difficult asthma' as defined by BTS.

Patient objectively recorded measures
Website usage X* Usage of asthma action plan (PAAP) X* Usage of asthma review pages X* Booked asthma review X* Usage of '4-week medication challenge' X* Usage of medication information X* Usage of breathing retraining challenge X* Usage of friends & family section X* Usage of stress reduction X* Choice of lifestyle changes X* Reported progress on lifestyle change (e.g. weight change) X*

Economic measures
Patient quality of life X X X Patient costs X X X Patient time and resource use X (NR)

Qualitative analysis
Patient experience and views of the DI X Key: NR = Notes review *My Breathing Matters arm only -measured via Lifeguide website 1 This will be measured directly before and after the My Breathing Matters training completion

BACKGROUND
The UK has one of the highest prevalence's of asthma in the world; nearly 6% of the UK population have asthma, comprising 5.4 million people, most of whom are managed in primary care [1] . Hospital admission and mortality rates for asthma showed improvements in the last decades of the last century, but these improvements have stalled since the millennium. Premature mortality from asthma was 1.5 times as high in the UK as in the rest of the EU in 2008, with around 1000 to 1200 deaths a year recorded since 2000. It is estimated that 90% of deaths are associated with preventable factors. Asthma is associated with high numbers of admissions and Emergency Department attendances, and it is estimated that 70% of these could have been prevented by appropriate early intervention and self-management [1] . Surveys of asthma symptoms and health status impairment continue to show that sub-optimal control is common and that at any given time the majority of asthmatics in the UK suffer potentially avoidable symptoms and quality of life impairment [2] .
Although the UK leads the world in providing guidelines for asthma management, these have been poorly implemented and people with asthma do not receive evidence-based interventions, particularly individual action plans, which are known to impact positively on outcomes [3] . Patient education and proactive self-management have been convincingly shown to improve clinical outcomes in asthma [3] and have been advocated in guidelines for 20 years [4] . People with asthma without a management plan are four times more likely to have an asthma attack needing emergency care in hospital [5] . However, a representative (Ipsos-MORI) survey carried out by Asthma UK in 2010 [6] suggested that only a quarter of people with asthma in the UK have a self-management plan. Self-management in asthma can also encompass nonpharmacological interventions to improve control and empower the patient, such as breathing exercises or lifestyle changes such as smoking cessation and weight reduction (since smoking and obesity are associated with worse prognosis in asthma [7] ).

RATIONALE AND RISK BENEFITS FOR CURRENT TRIAL
Increasingly widespread access to the internet and mobile phones [8,9] means that healthcare Digital Interventions (DIs) are accessible to the majority of patients, and can be used to provide information and support at any time the patient needs it [10] . DIs can empower patients by providing better access to personalised information, and support for active involvement in treatment and self-management [10] . A large meta-analysis found a small but significant positive effect of DIs on health-related behaviours [11] , whilst a Cochrane review found evidence that computer-based health interventions for those with chronic health conditions significantly improved knowledge, health behaviours and clinical outcomes [12] . DIs have the potential to make significant savings by automating routine aspects of patient education, monitoring and support, freeing up health professional resources for when patients most need them [13] . These savings can play an essential part in meeting the NHS QIPP agenda to achieve increased efficiency gains despite the growing demand created by an expanding and ageing population. There is accumulating evidence that DIs can deliver better and more efficient healthcare in the context of asthma. A recent systematic review by our group found that self-management DIs could improve asthma control and reduce asthma-related quality of life impairment but called for larger, more robust trials [14] . We identified several existing self-management DIs for asthma, but half of these were for children, and many were not in English. While there are numerous commercial DIs for asthma, only one has been evaluated [15] . The RAISIN pilot trial demonstrated that self-management interventions could be effective at improving quality of life and asthma control, with improvements to 'reach' and response rate by catering to patients with mild asthma but impaired quality of life through non-pharmacological means [16,17] . A Danish trial compared GP and specialist care with web-based self-monitoring with automated feedback and a stepped care medication plan (with GP advice when required) [15] . After six months those allocated to the web-based self-monitoring had greater improvement in symptoms, quality of life and lung function. The SMASHING trial in the Netherlands [18] compared usual care with web-based educational resources, self-monitoring and automated feedback on medication titration, plus some group and email nurse support. At one year the intervention group had better quality of life and lung function and more symptom free days, at no extra cost [19] .

TRIAL OBJECTIVES
The primary aim of the My Breathing Matters trial is to assess the feasibility, acceptability, effectiveness and cost-effectiveness of a DI in primary care for the self-management of asthma, in comparison to usual care (with provision of standard patient information materials produced by the charity Asthma UK).
To assess the feasibility and acceptability of a trial of My Breathing Matters, an intervention designed to assess improvements in clinical outcomes (e.g. quality of life, health resource use, lung function) of people with asthma.
Secondary research questions 1.
To assess feasibility of trial procedures including: clinical research governance, recruitment strategy, trial documents (e.g. PIS), eligibility criteria, consent/withdrawal, randomisation and blinding.

2.
To assess feasibility and acceptability of MBM intervention including: usage and engagement, adherence and completion, fidelity of providers.

3.
To assess feasibility of data analysis, including data collection, data quality, management of trial data and estimates of effect size across trial outcome measures (see primary/secondary endpoints) to inform sample size calculations for a larger phase 3 RCT.
Tertiary research questions 1. To examine intervention usage, progress and engagement to inform well-powered usage analysis in larger trial My Breathing Matters is a digital intervention for the self-management of asthma, consisting of pharmacological support, (advice about asthma reviews and personal asthma action plans, information about medication and side effects) and non-pharmacological components (stress reduction, online versions of breathing retraining courses shown to be acceptable and feasible [20] ) and optional user-selected lifestyle modifications. 80 participants will be randomised either to the My Breathing Matters programme or to a control group receiving usual care with provision of an Asthma UK information leaflet. This trial will be coordinated from UK facilitated by the PCRN's in these areas with researchers and research nurses employed at each centre.

TRIAL DESIGN
The MBM study will comprise: 1) Feasibility trial: 80 participants (40 per arm) will be recruited from practices to confirm the acceptability and feasibility of the intervention, full trial protocol and study procedures. 2) Qualitative analysis will also be embedded in to the My Breathing Matters study.

TRIAL OUTCOME MEASURES
The primary outcome of the trial will be the feasibility and acceptability of a trial of My Breathing Matters, an intervention designed to improve clinical outcomes (e.g. symptom control, quality of life) of people with asthma.
Secondary outcomes will be to assess the feasibility of measuring (and estimates of effect size to generate hypotheses and perform sample size calculations with which to test them) in the following trial measures: 3-month questionnaire packs will be mailed out with a freepost return envelope. Baseline and 12-month follow up questionnaires will be taken using paper questionnaire packs and clinical measures will be taken by a research nurse.
For 3 month and 12-month measures, a second pack will be sent out after 2 weeks if no response; followed by questionnaire measures (AQLQ) completed over the phone if still outstanding after a further 2 weeks.
Further detail regarding clinical measures:  Lung function will be measured using spirometry by an appropriately trained research nurse.
Measures of lung function will be: FEV 1 (forced expiratory volume in 1 second), FEV 1 /FVC (ratio of FEV 1 to forced vital capacity), PEFR (peak expiratory flow rate).  Medication prescriptions, any changes and support provision (consultations) will be identified from patient notes reviews after completion of the 12-month follow-up. Medication use will be converted into defined daily doses of medication.
Further detail regarding patient reported measures:  Asthma-specific quality of life (AQLQ) at 12-month follow-up  Asthma symptoms will be measured using the Asthma Control Questionnaire (ACQ)  Anxiety and depression will be measured using the Hospital Anxiety and Depression inventory (HADS)  Patient enablement (PEI): will be measured using the Patient Enablement Instrument [31].  Patient medication adherence: all participants will complete the MARS as a measure of medication adherence [2] .
In the event that the results from any of the questionnaires are of concern (e.g. demonstrate significant anxiety or depression), the research team will inform the participant of their scores and the normal range of scores, and suggest that they may wish to discuss this result with their GP.
Economic measures:  Health resource use (professional contacts, referrals, prescriptions)  Patient time/burden needed to take part in intervention  Health-related quality of life: the EuroQol (EQ-5D 5L) provides a measure of quality of life for economic analysis [5] and the ICEACAP-A measures non-health-specific quality of life in adults [21] .  Patient costs (e.g. additional costs incurred if patient engages in extra activities beyond intervention, such as gym membership)

Tertiary outcomes (intervention group only) will include:
 Lifestyle change choice, usage and progress  Adherence to recommendations  Engagement with program and reasons for such engagement All measured over the 12 months of the study. Assessed at baseline, after 3 months and after 12 months of online training.
Socio-demographic data  Participants' age, gender, internet experience, disease status (time diagnosed), smoking habits, education level, and social deprivation indices based on postcode will be recorded at baseline Eligible participants will be identified from around 6 general practices located by primary care research networks (PCRN); Practices will be selected in order to confirm feasibility for a wider trial. Principal Investigators will be identified at the regional level, with lead GPs nominated at each site.

CENTRE SELECTION
The following documents must be in place and copies sent to the MBM Clinical Trial Coordinator (CTC) (see contact details on page 2):  The approval letter from the relevant R&D Department, following submission of the Site Specific Information (SSI) form (where required)  A signed Study Agreement (PI and sponsor signature)  Completed Signature List and Roles and Responsibilities document  Completed contacts list of all site personnel working on the Study Upon receipt of all the above documents, the MBM CTC will send a confirmation letter to the lead GP. This letter must be filed in each centre's Site File. Along with this confirmation letter, the practice should receive their trial supplies and a study pack holding all the documents required to recruit a patient into the MBM Trial.

SCREENING AND PRE-REGISTRATION
Screening  Electronic database searches will be conducted, and records screened by general practitioners to remove participants who meet exclusion criteria. The latest version of the participant information sheets along with the consent and screening questionnaire (including a freepost response envelope) will be posted with a letter inviting the patient to take part in the study;  Eligible patients may also be identified and referred opportunistically during routine consultations with the practice nurse or GP.  Study advertisements will be displayed in participating practices;  Patients responding to the study advertisements will phone the research team who will conduct a minimal screening of participants by telephone using a standard set of questions (included as a supporting document) to establish whether they meet the basic inclusion criteria, e.g. diagnosis of asthma, access to internet. The research team will then send copies of patient study materials (consent form and questionnaire) to confirm eligibility.
Patients who do not respond within one month of the practice invite will be followed up by telephone. This procedure will be explicitly outlined in the invite letter sent from the practice. In addition, all invited participants will be provided with a response form to return should they wish to decline to take part in the trial. This will ask for basic demographic information and their reasons for declining, including an option not to give a reason if they prefer not to (see patient opt-out).

4.3
INCLUSION CRITERIA 1. Physician diagnosed asthma in medical record (confirmed via practice) 2. I , anti-asthma medication prescription in the previous year (determined from the physician prescribing records)

EXCLUSION CRITERIA
6. Asthma judged at the baseline assessment to be dangerously unstable and in need of urgent medical review (if unstable asthma is found, the patient will be referred back to usual primary care clinician for review) 7. Terminal disease or other condition which in the opinion of the family doctor makes them inappropriate to take part 8. Diagnosed with 'difficult asthma' as defined by BTS. 9. Documented diagnosis of Chronic Obstructive Pulmonary Disease (COPD) 10.Household member already enrolled on the study

Informed consent
All patients will receive a copy of the consent form in their invitation letter pack, and will be asked to complete it when contacting the research team. All patients will receive a face-to-face baseline consultation with the practice nurse where additional informed consent (see consent form) will be recorded before clinical measures, such as confirmation of asthma diagnosis, are undertaken (see primary outcomes data document).

Randomisation
Patients will be randomised at the practice where if in the intervention group they will be given access to the My Breathing Matters programme, or if in the control group will be given Asthma UK information leaflet. (see section 9.2 for further details of randomisation). Participants in the intervention group will be instructed to sign up to the My Breathing Matters programme upon which they will be notified by email of their involvement in the study. Only one patient per household will be randomised.

CONSENT
Consent to enter the trial must be sought from each participant only after a full explanation has been given, an information leaflet offered and time allowed for consideration. Signed participant consent will be obtained. The right of the participant to refuse to participate without giving reasons will be respected. After the participant has entered the trial the clinician remains free to give alternative treatment to that specified in the protocol at any stage if he/she feels it is in the participant's best interest, but the reasons for doing so should be recorded. In these cases the participants remain within the trial for the purposes of follow-up and data analysis. All participants are free to withdraw at any time from the protocol intervention without giving reasons and without prejudicing further treatment. Participants will give informed consent (including explicit consent to access relevant data from the DI and their medical records) before they first log onto MBM. They will be free to withdraw at any time simply by contacting the research team or practice staff All patients who withdraw from the intervention will be asked if they are prepared to attend the final follow-up appointment with the research nurse and/or complete self-report follow-up measures. If they agree, they will be invited according to the follow-up procedure and will have the option to answer follow up questionnaires without using the website. .

INTERVENTION ARM
The online My Breathing Matters programme will comprise three main components:

Intervention components
The MBM programme will use behavioural techniques to improve functional quality of life of primary care patients with asthma, by supporting illness self-management by pharmacological and non-pharmacological means (thereby reducing risk of asthma exacerbation). The program has three main design objectives: i) to engage people who do not view themselves as having active asthma, ii) to persuade and educate users to implement appropriate pharmacological management and iii) to encourage users to employ nonpharmacological methods of improving QoL. Key features to address these objectives include: i) maintaining positive illness context throughout (i.e. promote health rather than manage illness) and offering a simple, unobtrusive interface to provide optional (and flexible) support only when needed, ii) focusing on persuading/educating users regarding the necessity, efficacy and safety of preventative asthma medication, and facilitating easy completion of an action plan with primary care support, and iii) educating users on benefits and offer psychological methods to improve quality of life (e.g. cognitive behavioural techniques for symptom management), tailored access and address patient concerns about relevant positive lifestyle changes, such as weight-loss if overweight, smoking cessation if current smoker, physical activity if inactive.
Patients in the intervention group will be offered optional support from Asthma UK helpline -as a source of asthma advice and support for those who would like personal contact outside of that offered by MBM or their healthcare team, to support selfmonitoring and lifestyle modification. This support would be the same as that offered to anyone who calls (outside of MBM) so the Asthma UK nurse team would not require additional training. Motivating and reassuring support messages will be sent by email to all patients in the intervention group every 2 to 4 weeks; these will encourage and reinforce patient adherence to lifestyle changes (where applicable).

USUAL CARE
All patients will be able to access the Asthma UK information leaflet following randomisation. Patients within the usual care condition will be informed that they should simply follow their usual instructions for asthma management. After the patients have completed the 12-month follow-up they will be given access to the intervention, in line with recommendations from PPI representatives.

ADVERSE EVENTS AND REPORTING
7.1 DEFINITIONS Adverse Event (AE): any untoward medical occurrence in a participant or clinical study participant which does not necessarily have a causal relationship with study treatment or participation.
Serious Adverse Event (SAE) or Serious Adverse Reaction: any untoward medical occurrence or effect that at any dose:  Results in death  Is life-threatening -refers to an event in which the participant was at risk of death at the time of the event; it does not refer to an event which hypothetically might have caused death if it were more severe  Requires hospitalisation, or prolongation of existing hospitalisation  Results in persistent or significant disability or incapacity  Is a congenital anomaly or birth defect  Other important medical events***.
*'life-threatening' in the definition of 'serious' refers to an event in which the patient was at risk of death at the time of the event; it does not refer to an event which hypothetically might have caused death if it were more severe. **Hospitalisation is defined as an inpatient admission, regardless of length of stay, even if the hospitalisation is a precautionary measure for continued observation. Hospitalisations for a preexisting condition, including elective procedures that have not worsened, do not constitute an SAE. ***Other important medical events that may not result in death, be life-threatening, or require hospitalisation may be considered a serious adverse event/experience when, based upon appropriate medical judgment, they may jeopardise the participant and may require medical or surgical intervention to prevent one of the outcomes listed in this definition.
Note: It is the responsibility of the PI or designated to grade an event as 'not serious' (AE) or 'serious' (SAE).

CAUSALITY
The assignment of the causality to trial procedures of any serious event should be made by the investigator responsible for the care of the participant using the definitions in the table below.
If any doubt about the causality exists the local investigator should inform the trial coordinator who will notify the Chief Investigator. Other clinicians may be asked for advice in these cases.
In the case of discrepant views on causality between the investigator and others, all parties will discuss the case. In the event that no agreement is made, the Ethics Committee will be informed of both points of view.

Relationship Description Unrelated
There is no evidence of any causal relationship Unlikely There is little evidence to suggest there is a causal relationship (e.g. the event did not occur within a reasonable time after administration of the trial medication). There is another reasonable explanation for the event (e.g. the participant's clinical condition, other concomitant treatment).

Possible
There is some evidence to suggest a causal relationship (e.g. because the event occurs within a reasonable time after administration of the trial medication). However, the influence of other factors may have contributed to the event (e.g. the participant's clinical condition, other concomitant treatments).

Probable
There is evidence to suggest a causal relationship and the influence of other factors is unlikely.

Definitely
There is clear evidence to suggest a causal relationship and other possible contributing factors can be ruled out.

REPORTING PROCEDURES
All adverse events should be reported. Depending on the nature of the event the reporting procedures below should be followed. Any questions concerning adverse event reporting should be directed to the SCTU in the first instance.

Pre-existing Conditions
A pre-existing condition should not be reported as an AE unless the condition worsens by at least one CTCAE grade during the trial. The condition, however, must be reported in the pre-treatment section of the CRF, if symptomatic at the time of entry, or under concurrent medical conditions if asymptomatic.

Non serious AEs
All adverse events that may be related to the study will be recorded in the relevant case report form and Adverse Event form and sent to the SCTU within one month of the form being due. As adults on average see their GP approximately 5 times per year for a variety of routine and unscheduled appointments (e.g. for medication review, self-limiting minor illnesses and long- term conditions unrelated to asthma), many medical encounters are of no relevance to the study. Events that will be recorded include any judged by the study nurse to be possibly related to the study. In particular, all medical encounters related to the following medical areas or symptoms will be recorded in the Adverse Events form: Psychological morbidity: any events relating to anxiety, depression or mood disorders Respiratory morbidity: any events relating to breathing or chest symptoms Musculoskeletal, Abdominal and chest pain: any events relating to pain in these systems unless known to be associated with an unrelated pre-existing condition.
The study nurses are advised to record any event for which there is uncertainty as to whether it is study related or not, and to discuss with the local PI or CI.

Serious AEs
All SAEs (including those that are expected and related) will be reported within 24 hours of the local site becoming aware of the event. The SAE form asks for nature of event, date of onset, severity, corrective therapies given or action taken, outcome and relatedness (i.e. unrelated, unlikely, possible, probably, definitely). The responsible centre Principal Investigator will assign the relatedness and expectedness of the event. Additional information will be provided as soon as possible if the event has not resolved at the time of reporting.
A flowchart is given below to illustrate reporting procedures:  GPs or nurses will be asked to notify us via an SAE form if a participant experiences any SAEs.  The Sponsor and main Research Ethics Committee (REC) will be informed of all related SAEs occurring during the trial according to the following timelines, where day zero is defined as the date the SAE form is initially received:  Events which are fatal or life-threatening will be reported no later than 7 calendar days after the sponsor is first aware of the reaction. Any additional relevant information must be reported within a further 8 calendar days.  Events that are non-fatal or non life-threatening will be reported within 15 calendar days of the sponsor first becoming aware of the reaction.  All Investigators will be informed of all related SAEs occurring throughout the trial. Local Investigators should report any SAEs as required by their Local Research Committee and/or Research and Development Office.

Follow Up and Post-study Serious Adverse Events
The reporting requirement for SAEs affecting participants applies for all events occurring up to the end of the last treatment. All unresolved adverse events should be followed by the local investigator until resolved, the participant is lost to follow-up, or the adverse event is otherwise explained. At the last scheduled visit, the investigator should instruct each participant to report any subsequent event(s) that the participant, or the participant's general practitioner, believes might reasonably be related to participation in this study. The investigator should notify the Centre for Applications of Health Psychology of any death or adverse event occurring at any time after a participant has discontinued or terminated study participation that may reasonably be related to this study.

ASSESSMENT AND FOLLOW-UP OF PARTICIPANTS
Measures will be administered for all participants at baseline, 3 and 12 months, unless otherwise stated (see Schedule of Observations and Procedures). Patient-reported outcome measures will be completed online. Non-respondents will receive two email reminders followed by a paper-based copy accompanied by a final telephone follow-up. All website usage (MBM measures) will be recorded automatically in Lifeguide.
An independent research nurse, blind to study allocation will complete the follow-up assessments. All patients (including withdrawn participants who have consented to follow-up appointments) will follow the procedure outlined below:  Approximately three weeks before the follow-up appointment is due, the patient will be sent a letter by the research team encouraging them to arrange an appointment with the research nurse. Appointments may be scheduled to take place in the patients' home or usual GP practice.  If the patient does not contact the research team to arrange an appointment within two weeks of the letter being sent the patient will be followed up by telephone.  The research team or research nurse will send confirmation of the follow-up appointment to the patient by telephone, text, email or letter (dependent on patient preference).  If the patient indicates that they would not be willing to complete a follow-up, no further contact will be made with the patient regarding the follow-up appointment.

DEFINITION OF END OF TRIAL
The end of the trial will be defined as date of entry of final data into database.

SAMPLE SIZE
A minimum of 40 participants from two centres will be recruited to each condition during the study.

RANDOMISATION
Participants will be randomised in a 1:1 ratio to receive either usual care (control) or MBM programme with optional support (intervention). The randomisation will be stratified block randomisation by the average value of AQLQ scores in the BREATHE Trial (Thomas et al).
Remote allocation will maintain allocation concealment from both the participant and the research nurse prior to allocation, however the outcome assessor will be blinded to participant allocation.

STATISTICAL PLAN INCLUDING INTERIM ANALYSIS
A detailed statistical plan will be developed prior to data lock. No interim analysis is planned. Data and all appropriate documentation will be stored for a minimum of 10 years after the completion of the trial, including the follow-up period, in a secure location at Southampton University.
Descriptive statistics of outcome variables will be calculated in order to inform a future randomized controlled trial, including examining levels of missing data, attrition rate, retention and completion. Statistical comparisons between and within intervention and control groups will be performed to estimate variability in primary and secondary outcomes within a linear regression framework. Further exploratory analysis of patient engagement, intervention usage and possible mediators of behaviour change will be performed.

QUALITATIVE EVALUATION
The qualitative study will be undertaken both alongside participation and after participants finish using MBM. Qualitative interviews will seek to provide an in-depth understanding of the perspective of patients, to inform intervention and trial acceptability and to generate hypotheses about intervention mechanisms of action that can be tested in a larger trial.
Participants and sampling Patients will be asked whether they give permission to take part in an interview at a later point in the study. Purposive sampling will be used to select patients from the intervention group to allow for a wide range of views and experiences of the MBM programme. It is anticipated that 20 patients will be interviewed during and following the study, and patients will be selected for interviews until saturation is reached.
Interviews and qualitative analysis Factors that may facilitate or diminish the acceptability of the MBM programme, and adherence to implementation will be explored across patient and health care professional interviews. Interviews will be conducted from after the 3-month assessment. Open-ended questions will be used to elicit user perspectives and experiences of the intervention, allowing participants to freely describe their experiences and views in their own way and to focus on whatever is most salient to them.
Interviews will be audio-recorded and fully transcribed. The findings will be used to inform any modifications needed to the MBM programme or the trial procedures for a potential future full RCT.
The transcriptions will be anonymised (identifiable data removed) and participants' transcripts will be given participant numbers so that they can be easily discussed between team members whilst protecting participants' identities. To ensure that we remain open to and grounded in users' perspectives we will carry out inductive thematic analysis [22] of all textual data,

CLINICAL TRIAL AUTHORISATION
This trial does NOT involve the testing of any Investigational Medicinal Products (IMPs) therefore approval from the Medicines and Healthcare products Regulatory Agency is not required.

ETHICS APPROVAL
The trial protocol will be submitted to a Research Ethics Committee (REC) recognised by the United Kingdom Ethics Committee Authority (UKECA) for review and approval. A favourable opinion must be obtained before commencement of any trial procedures (including recruitment of patients) occurs All substantial amendments must be approved by the REC responsible for the trial, in additional to approval by NHS R&D. Minor amendments will not require prior approval by the REC.
If the trial is stopped prematurely, it will not be recommended without reference to the REC responsible for the trial.
The outcome of the trial will be reported to the responsible REC within 90 days of completion of the last patient's final trial procedures. In the event of the trial being prematurely terminated a report will be submitted to the REC within 15 days. A summary of the Trial Report will be submitted to the responsible REC within one year of completion of the last participant's final study procedures.
The investigator must ensure that participant's anonymity will be maintained and that their identities are protected from unauthorised parties. On CRFs participants will not be identified by their names, but by an identification code.

3 CONFIDENTIALITY
Initial practice database searchers will be conducted on practice computer systems and subsequent study invitations will be the responsibility of the practice, to maintain the confidentiality of potential participants. Participants' identification data will be required for the registration process, which will be completed with the support provider at the baseline screening appointment. Participants will be informed that the research team will have access to their study data, and that personal information such as their name, telephone address and email address will be stored and used by the research team to stay in touch with them throughout the study.
The primary data outcomes measured by the support provider (such as asthma quality of life and current medication) will be stored on an independent, secure server. All other data (secondary self-report measures and intervention data) will be stored on dedicated secure spaces behind a firewall on password protected computer located in secure university buildings at the University of Southampton. The data is backed up daily. Confidentiality of all data entered into the DI will be maintained by following best practice in NHS IT protection and data security systems (e.g. regarding use of https, storing data behind the university firewall etc). Access to the website will be via username and password. The use of strong passwords will be enforced. The communication between the person entering data through a web browser and the server will be through a secure internet connection (HTTPS). The research team will adhere to the Data Protection Act 1998.
Digital records will be stored on a university computer in a password protected file. All data will be linked with the participant's study ID which does not include any personal or identifiable information, such as name or D.O.B. Study IDs will be linked with patient's names, email addresses telephone numbers in a separate, securely stored file.
At the end of the study, data will be anonymised and stored on a password protected computer located in secure university buildings and appropriately backed up. Confidentiality of all data entered into the DI will be maintained by following best practice in NHS IT protection and data security systems.

INDEMNITY
The sponsor of the trial is University of Southampton. University of Southampton insurance will apply. The lead authors of the protocol are the Chief Investigator (Prof Lucy Yardley), and Prof Mike Thomas, who are both university employees and as such will be covered by the University insurance.

SPONSOR
University of Southampton is acting as the sponsor for this trial. The My Breathing Matters coordinating team has been delegated duties by the Sponsor relating to: submissions to regulatory authorities and GCP.

DEVIATIONS AND SERIOUS BREACHES
Any trial protocol deviations/violations and breaches of Good Clinical Practice occurring at sites should be reported to the MBM trial coordinator and the local R&D Office immediately. The MBM trial coordinator will then advise of and/or undertake any corrective and preventative actions as required.
All serious protocol deviations/violations and serious breaches of Good Clinical Practice and /or the trial protocol will immediately be reported to the regulatory authorities and other organisations, as required in the Medicines for Human Use (Clinical Trials) Regulations 2004, as amended.

AUDITS AND INSPECTIONS
The trial may be participant to inspection and audit by University of Southampton, under their remit as sponsor, the trial coordinating centre as the Sponsor's delegate and other regulatory bodies to ensure adherence to ICH GCP, Research Governance Framework for Health and Social Care, applicable contracts/agreements and national regulations.

TRIAL MANAGEMENT
The Trial Management Group (TMG) is responsible for overseeing progress of the trial. The day-to-day management of the trial will be co-ordinated through the trial coordinating centre and oversight will be maintained by the Trial Steering Committee, with an experienced independent chair and representation of patient representatives, Asthma UK and Blood Pressure UK.
The Steering Committee will meet twice a year throughout the programme to provide strategic guidance and independent monitoring of progress and professional conduct. We will encourage in person attendance at all these meetings where possible, but will also provide for attendance by teleconference when necessary, and will circulate papers and minutes before and after meetings for communication with those who cannot attend for any reason.

PUBLICATION POLICY
All publications and presentations relating to the trial will be authorised by the Trial Management Group and will follow an agreed publication policy. Dissemination of our work will be via multiple pathways: a) to the scientific community through presentation at national & international conferences and regular publication in highly cited and open access peer reviewed journals. b) to clinical and academic colleagues via professional societies: links with the following societies will be exploited to raise the profile of this work: Royal College of GPs and Physicians, Society of Behavioural Medicine, British Sociological Society Medical Sub-group. c) to patients via patient groups: We will work with Asthma UK to disseminate our results. d) to participants: All participants will be sent an accessible summary of the findings from the study that they took part in within six months of study completion. e) to relevant NHS organisations and healthcare providers (e.g. Clinical Commissioning Groups, NHS Choices, UCL partners).  Methods 3a Description of pilot trial design (such as parallel, factorial) including allocation ratio 7 Trial design 3b Important changes to methods after pilot trial commencement (such as eligibility criteria), with reasons -4a Eligibility criteria for participants 7 Participants 4b Settings and locations where the data were collected 7 4c How participants were identified and consented 7 Interventions 5 The interventions for each group with sufficient details to allow replication, including how and when they were actually administered 8 6a Completely defined prespecified assessments or measurements to address each pilot trial objective specified in 2b, including how and when they were assessed 10 Outcomes 6b Any changes to pilot trial assessments or measurements after the pilot trial commenced, with reasons -6c If applicable, prespecified criteria used to judge whether, or how, to proceed with future definitive trial -7a Rationale for numbers in the pilot trial 8 Sample size 7b When applicable, explanation of any interim analyses and stopping guidelines -Randomisation: 8a Method used to generate the random allocation sequence 8 Sequence generation 8b Type of randomisation(s); details of any restriction (such as blocking and block size) 8 Allocation concealment mechanism 9 Mechanism used to implement the random allocation sequence (such as sequentially numbered containers), describing any steps taken to conceal the sequence until interventions were assigned 8

Strengths and Limitations of the Study (Summary: up to 5 bullet points)
1. This pragmatic randomised controlled feasibility trial examined 'My Breathing Matters' (MBM), a digital asthma self-management intervention that supported both pharmacological and non-pharmacological management of asthma symptoms.
2. MBM was developed using theory, evidence and person-based approaches, and compared to standardised usual care (a booklet) with successful blinding and randomisation.
3. Not all patients engaged with the intervention, and although numeric improvements in patient reported asthma outcomes were larger in the active arm, improvements were observed in both arms. show that sub-optimal control is common and that the majority of people with asthma in the UK frequently experience potentially avoidable symptoms and quality of life impairment [2].
Proactive self-management of asthma has been convincingly shown to improve clinical outcomes and have been advocated in guidelines for 25 year [3]. Guidelines are not always well implemented [4] and consequently some people with asthma do not receive evidence-based interventions which are known to impact positively on outcomes. Recent large-scale systematic reviews demonstrated that supported asthma self-management can reduce healthcare utilisation and increase asthma control, without increasing healthcare costs [5,6]. For example, people with asthma without a management plan are four times more likely to have an asthma attack needing emergency care in hospital, yet only 44% of people with asthma in the UK report having a self-management plan [7]. Self-management recommendations for asthma have also encompassed non-pharmacological strategies to improve control. These include lifestyle interventions, such as smoking cessation, allergen avoidance, weight reduction in those with obesity, and breathing retraining interventions [8].
Digital interventions (DIs) are increasingly recognised as a possible approach to achieve the aims of supporting chronic diseases such as asthma. DIs can be convenient, easily accessed and may provide cost-effective tools by automating routine aspects of patient education, monitoring and support [9]. There is accumulating evidence that DIs are feasible and may be effective in the context of asthma. The SMASHING trial compared usual care with web-based educational resources, self-monitoring and automated feedback on medication titration, plus some group and email nurse support for patients with asthma. After 12 months the intervention group had better quality of life and lung function and more symptom-free days, at no extra cost [10]. The RAISIN pilot trial indicated that self-management interventions that included nonpharmacological (behavioural and psychological) components could be effective at improving quality of life and asthma control, with improvements to 'reach' and response rate, by catering to patients with mild asthma but impaired quality of life [11]. A recent systematic review and meta-analysis indicated that self-management DIs may be able to improve asthma control and reduce asthma-related quality of life impairment [12] however there is limited evidence of benefit for other outcomes and larger confirmatory trials are required.
In the current randomised controlled feasibility trial, we developed and evaluated a digital selfmanagement intervention, that incorporated pharmacological and non-pharmacological selfmanagement support for adults in primary care with impaired asthma-specific quality of life ('My Breathing Matters'; MBM), using evidence, theory and person-based approaches [13] and in line with Medical Research Council (MRC) guidance for developing and evaluating complex interventions [14].

Aim
The aim of the MBM study was to assess the feasibility of a trial to evaluate a digital intervention in primary care to improve quality of life and other clinical outcomes (such as asthma control, health resource use, lung function) of people with asthma, in comparison to usual care (with provision of standard patient information materials produced by the charity Asthma UK).

Research Objectives
1. To assess feasibility of trial procedures including recruitment strategy, eligibility criteria, consent, withdrawal, randomisation and blinding.
2. To assess feasibility of the MBM digital intervention including usage and engagement.
3. To assess feasibility of data analysis, including data collection, data quality and management of trial data across trial endpoint measures to inform sample size calculations for a larger phase 3 randomised controlled trial.

Method Design
We conducted a pragmatic feasibility RCT of the MBM digital intervention in primary care.

Setting
Eligible participants were identified from seven general practices from the Wessex, UK primary care research network to facilitate recruitment of people with varied socio-economic status. To ensure we evaluated the intervention across a spread of socio-economic deprivation, practices were purposively selected to be both rural (N = 4) and urban (N = 3), with mean practice deprivation index of 20.60% (SD 10.5); practice socio-economic deprivation deciles = 2, 4, 4, 5, 8, 10, 10, in which lower deciles indicate more deprivation [15]).

Participants
Patients were included in the trial if they were aged 18 years or more, had physician-diagnosed asthma in their medical record, had received one or more anti-asthma medication prescription in the previous 12 months, had impaired asthma-related health status (Asthma Quality of Life Questionnaire score of less than 5.5 as assessing using a self-completed postal questionnaire), provided informed consent, were able to understand English and had access to the internet.
They were excluded from the trial if 1) their GP considered it inappropriate for them to take part (such as having an additional terminal condition), 2) they were attending a secondary care asthma clinic, or they were receiving either maintenance oral corticosteroids or injected biological treatments to control their asthma, 3) they were diagnosed with COPD, 4) they had a household member already enrolled on the study, or 5) they were judged by the research nurse to have 'unstable asthma' according to clinical assessment and spirometry data at the baseline assessment (in which case they were referred back to their GP), or were diagnosed with 'difficult asthma' defined by British Thoracic Society (BTS).

Recruitment
Electronic searches of the computerised primary care medical record were conducted, and and freepost return envelopes were posted to participants who returned them if they were interested in taking part. Patients who met screening criteria (AQLQ score of less than 5.5) were contacted by research team staff and attended a baseline appointment at their practice with a trained research nurse. Recruitment began in March 2017 and was completed in August 2017.

Sample size
The target for this trial was to recruit 80 patients overall (40 per arm), in order to assess primary feasibility outcomes and to assess intervention engagement and acceptability.

Randomisation and Blinding
After completing outcome measures at their baseline appointment participants were randomised (block randomisation stratified by an average primary care AQLQ score [4.3] taken from a previous trial using the same inclusion criteria [8]). Information packs were given to participants after randomisation with instructions on signing up to MBM (if randomised to intervention group) or just usual care materials (if randomised to control). Research nurses conducting baseline appointments were blinded throughout the study until the final questionnaire which was only delivered to the intervention group.

Intervention Group: Usual Care with MBM and Asthma UK booklet.
Patients in the intervention group continued to receive usual care but were also given a code that allowed free unlimited access to MBM. MBM is a digital asthma self-management intervention that supports asthma self-management using both pharmacological and nonpharmacological approaches, developed using the LifeGuide Software [17] and described below according to the TIDieR checklist [18]. A demonstration version of the intervention is available here: http://www.mybreathingmatters.co.uk. After signing up and completing quality-of-life The pharmacological section provided information on different medication classes and inhalers, the use of personalised asthma action plans (PAAPs), encouraged medication adherence, and gave information to facilitate and inform an effective asthma review with their GP.
Pharmacological content was initially based on 'Living Well with Asthma', an asthma selfmanagement intervention that previously demonstrated feasibility for self-management [11,19] and was developed in collaboration with people with asthma and with input from Asthma UK (a national asthma charity). This section was designed to answer common concerns about medication, incorporating a strategy described as 'the 4-week medication challenge' that encouraged participants to realise the benefits of adherence to regular medication, by selfmonitoring their symptoms during 4 weeks of continuous inhaler use. The non-pharmacological support included sections on a number of strategies to improve asthma control, such as Breathing Retraining, stress reduction, and additional healthy lifestyle resources (physical activity, weight reduction, hand hygiene and smoking cessation). Optional nurse support was available by Asthma UK who provide a dedicated nurse helpline that was advertised through the intervention.
The intervention was developed using the person-based approach [20] which places patients at the heart of the development process. Evidence from primary mixed methods research (such as [19]) and qualitative and quantitative reviews was used to develop guiding principles. A prototype intervention was piloted using 'think aloud' interview studies in which patients with asthma used prototype versions of the MBM website and provided feedback on intervention acceptability and feasibility as they used it. In 46 interviews with 30 patients (purposively selected across a range of age and gender), the intervention was iteratively modified and updated to address patient feedback until participants indicated no further modification was required, confirming the intervention was as acceptable and engaging as possible (for more details on this process see [21]). Due to the digital nature of the intervention, participants could engage with components of the intervention as much or as little as they wished. Tailored advice was offered according to participants' preference to find out more about pharmacological or non-pharmacological selfmanagement techniques (patients selected a check box option of "I'd like to find out more about how my asthma could be helped by i) making the most of my asthma medicine, or ii) 'nonmedicine' ways to help my breathing."), with automated reminders whenever patients had not accessed the intervention for several weeks, or when content was made available that they had not previously seen. The intervention was not modified during the study.
Intervention usage was monitored through digital usage metrics (reported below).Nonengagement with the website was not addressed, in line with the pragmatic nature of the feasibility study. Participants were sent one email and received one phone call in which they were offered technical support if they had not logged onto the intervention at all for one month following their baseline appointment.

Control Group: Usual Care with Asthma UK Booklet
To provide 'good quality' usual care to participants allocated to this arm, as well as usual care from their practice, participants were given an Asthma UK booklet 'Live Well with Asthma' at their baseline appointment. The booklet was created by a multidisciplinary team and expert patients, and aimed to provide essential information and advice to enable effective selfmanagement to occur. It is available to anyone via the Asthma UK website [22]. The booklet was provided in hard copy and provided information about asthma symptoms and triggers, medication adherence and usage techniques, PAAPs and support from families. The booklet also advertised the Asthma UK support line. Booklet usage was not monitored.

Outcome Measures
In line with objectives, trial outcomes are reported below as 1) feasibility outcome measures, 2) intervention usage outcome measures, and 3) trial endpoint measures to inform a larger trial.  -Health-related quality of life. Measured using the EQ-5D-5L [24] in which participants select their functioning level across five dimensions (immobility, self-care, usual activities, pain and discomfort, and anxiety and depression) on a five-point scale, with higher scores indicating greater problems with functioning.
-Health-related capability. Measured using the Icepop Capability measure for adults (ICECAP-A [25]), in which participants select their capability across five dimensions (stability, enjoyment, achievement, attachment and autonomy) on a four-point scale, with higher scores indicating better capability.

11
-Anxiety and depression. Measured using the Hospital Anxiety and Depression Scale (HADS [26]), a 12-item questionnaire in which higher scores on depression and anxiety subscales indicate higher anxiety across two subscales (anxiety and depression).
-Enablement, measured using a modified version of the patient enablement instrument (PEI [27]) that has been validated in previous RCTs [28]. The modified PEI is a 7-point scale consisting of 6 items, in which higher scores indicate more enablement.
-Patient satisfaction was measured by asking patients whether they saw any benefits or disadvantages to using MBM, and whether they would recommend it to friends and family based on the NHS Friends and Family Test (FFT [29]). Physiological measures of lung function were taken at baseline and 12 month appointments: forced expiratory volume (FEV 1 ), ratio of forced expiratory volume to forced vital capacity (FEV 1 /FVC) and peak expiratory flow rate (PEFR).
At 12-month follow-up, we also monitored health resource use GP consultations, A&E visits, hospital admissions, asthma medication use and use of antibiotics for chest infections using GP practice patient notes.
Healthcare utilisation data were collected via retrospective notes review conducted by practice staff. Staff were provided with a template for reviewing data, and an instruction manual to ensure correct data were provided. Initial notes reviews were completed within two months from completion of primary data collection. Nine patients (10% of the total patients in each practice) were also reviewed by research nurses to assess data quality.

Data Analysis
Primary analysis of the study was a description of key feasibility outcomes including patient eligibility, recruitment rates, withdrawals, 3 and 12-month follow-up response rates and digital intervention usage, as reported in the trial protocol (see Supplementary File 1).
Descriptive statistics were used to identify any floor or ceiling effects. For continuous measures, means, standard deviations (SD) and 95% CIs were reported at baseline, 3-month and 12month for each group, as well as for the sample as whole.
Exploratory analysis explored group differences in continuous primary endpoint measures (AQLQ, ACQ, HADS, PEI) using linear regression models that controlled for baseline values.
Participants were analysed in the group to which they had been randomised and comprised complete cases only.
Proportions of patients achieving a minimal clinically important difference (MCID) was described for asthma quality of life (the AQLQ MCID is 0.5 [16]).
Sensitivity analysis explored missing data at 3 and 12 months.
Healthcare utilisation outcomes were explored using a negative binomial model of group count data.
Health economic analysis was descriptive, reporting estimates of cost and outcomes measures and baseline and follow-up. The completeness and suitability of EQ-5D-5L and ICECAP were Intervention engagement was descriptive.

Recruitment and retention
Six practices were initially recruited and after monitoring recruitment rates a seventh practice added. In this additional practice only half the list (randomly selected) were offered participation in the study to avoid over recruiting. Across the 7 practices, 68478 patients were assessed for eligibility with 3199 meeting initial eligibility criteria (asthma diagnosis, >1 asthma medicine prescription in last 12 months, screened by practice). 266 patients completed postal screening measures before the recruitment period finished, of whom 125 were eligible to take part (impaired asthma-related quality of life, AQLQ score less than 5.5). Ninety patients responded to further contact. Two patients did not attend their baseline appointment leaving a final sample of 88 patients (intervention N = 44, usual care N = 44) who were recruited into the study (13.5 per practice) and were randomised over a 5 month period. Figure 1 presents the study CONSORT diagram.
During the study, 2 patients withdrew before 3-month follow-up and 2 before 12-month followup. All were in the intervention group. Patients withdrew for several reasons including lack of time (N = 1), illness (N = 1), death of family member (N = 1), and lack of perceived benefit (N = 1). 3 of 4 participants who withdrew had used the intervention; one had not. One participant was withdrawn from the study prior to 12-month follow-up they were no longer eligible (i.e. they were referred to secondary care).

Patient characteristics
Demographics and baseline characteristics of participants are presented in Table 1 and were reasonably well-balanced between arms across all measures.  Table 1 compares those lost to follow-up to those who remained in the study at 3 months in a sensitivity analysis. Those lost to follow up were slightly more likely to be female, have a higher BMI, a longer time since diagnosis, a lower AQLQ score, a higher HADS-A and HADS-D score and to be from a more deprived postcode.

Trial Endpoint Measures
The full data of the trial endpoints is set out in Table 2.
Both the intervention group and control group improved from baseline to 3-month and 12-  Table 2.

Adverse Events
Adverse events were reported by GPs and nurses who contacted the study team to report both adverse and serious adverse events. Nine adverse events were reported (intervention N = 6, usual care N = 3). These were assessed by research team clinicians and all were considered unlikely to be related to the study. Three were related to participant asthma (asthma exacerbation not leading to hospital admission, upper respiratory tract infection, sinusitis).
Three serious adverse events were reported (intervention N = 2, usual care N = 1). These were considered unlikely to be related to the study and the condition (atrial fibrillation, open distal radius fracture, cardioversion).

Healthcare Utilisation Outcomes
Data were collected from retrospective notes reviews (conducted by practice nurses) from 83 participants, reported in Table 3. Data was collected from 84 practices for 7 participants, with 4 participants from 1 practice incomplete. The data quality check, and subsequent examination by research team clinicians (MT) found that reviews completed by the practice nurses varied substantially in quality with varied levels of detail, and the quality of data achieved in this way was insufficient for a health economic analysis.  Note: (*) Incidence rate ratio vs. intervention group, controlling for 12 months before study period.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  Comparisons between group count data were reported using a negative binomial model but given the issues with the reliability of the data, should be interpreted cautiously. The prescription rate was approximately 8% higher in for both SABA (IRR 1.08, 95% CI 0.82, 1.43) and ICS (IRR 1.08, 95% CI 0.86, 1.35) in the control group compared to the intervention group.
Both groups had a low number of prescriptions for oral steroids, oral steroids and antibiotics, with only 15 prescriptions in total for either of these medications, making between group comparisons unreliable.
Due to unreliability of data, frequency of GP consultations, A&E admissions and hospitalisations have not been reported.

Health Economic Outcomes
Both EQ-5D-5L and ICECAP-A had the same completion rates as other secondary measures completed at follow up (see Table 2).
Patients reported several programmes across both groups including gym, walking, yoga, sewing, language courses, physio and signing (see Appendix 2). There were no substantial differences in terms of numbers or costs although the sample size was small.

Discussion
In line with our main research objectives, findings from our randomised controlled feasibility trial demonstrate that a full-size confirmatory trial to confirm effectiveness of MBM, a digital self-management intervention for adults in primary care with asthma is likely to be feasible and acceptable. Our trial procedures, intervention usage and data management were all feasible.
Our sample varied in age with a relatively high mean (56 years) indicating that our digital intervention can provide benefit to older adults. Both male and female adults were wellrepresented in our trial. A notable proportion of our sample was obese (41%), in line with previous findings [11]. Given that obesity is a risk factor for asthma, a larger trial could further improve effectiveness by providing more specific behavioural content for obese adults with asthma (such as tailored content to increase motivation to use weight-loss related lifestyle components in obese patients). Our sample was also predominantly white. Underrepresentation of minority ethnic groups in medical research in the UK is an ongoing issue [12] and should be addressed in recruitment procedures in the full trial.
The feasibility of a full trial is supported by the effective completion of trial procedures. All patients who completed baseline measures were randomised. Completion of measures was good at both follow-up points (3-month via post and 12-month at participants' practice). Where participants were not able to attend a follow-up appointment at practices, they were satisfactorily followed up via post or telephone for main trial measures. 8 patients were lost to follow-up (4 withdrew and 4 no longer responded to attempts to contact them). Notably, all 8 patients lost-to-follow were in the intervention group. It is possible that patients in the control group were more likely to maintain contact as they were only able to access the intervention upon completion of 12-month follow up measures. Although loss to follow-up is low, it is important to consider whether that loss is differential. Those lost to follow-up were more likely to be more socio-economically deprived, female, have a higher BMI, a longer time since diagnosis and a higher HADS-A and HADS-D score. It is possible that these patients would benefit from using MBM more than most, and therefore we have proposed several ways to further increase trial efficacy. Automatic email intervention registration at baseline (patients cannot attend baseline appointment without enrolling on the intervention) would increase initial engagement and engagement with trial procedures throughout duration of study. Online questionnaire completion during screening process would i) screen patients who are unable to interact with online trial/intervention and therefore unable to benefit from the intervention (feasibility trial estimate = 2%), and ii) streamline baseline/follow-up procedures.
Both health economic outcomes had high completion rates but did not suggest substantive change, similar to EQ-5D measures in previous non-pharmacological self-management trials (such as [8]). It is possible that an alternative measure such as the Short Form 12-item Survey (SF-12 [31]) in which participants consider the previous two weeks (whereas in the EQ-5D they consider the immediate present) may be better suited to measure small yet valuable changes in well-being over a full trial. Our detailed mixed-methods process analysis explored issues of trial acceptability in more detail, and will be reported in a subsequent paper. A full trial of this nonpharmacological intervention should accurately capture 'non-medical' costs (such as gym membership) that are likely to impact disease-specific quality of life, as well as medical costs that would be affected by changes in healthcare utilisation.
Healthcare utilisation data were collected by practice nurses whose main role was to provide usual clinical care at the practices, using a manual to guide data collection, rather than by trained research nurses, and our quality check demonstrated that the data collection process used was unreliable some centres. We conclude that in a full subsequent study, these data should be collected from the medical record by a trained member of the study team (such as a trained research nurse), as has been successfully used in previous studies [8].
Engagement with the intervention was slightly increased compared to a previous similar digital asthma self-management intervention [19] at initial sign up (82% vs 76%) as well as maintaining a higher number of ongoing engagement throughout the follow up period (median 3 additional log ins vs. 1), although our study used a broad primary care population while the RAISIN protocol primarily recruited from areas of high deprivation. This finding demonstrates that the use of the person-based approach to develop the intervention resulted in an intervention that was acceptable and engaging to patients, even using a pragmatic methodology in which patients self-registered at home instead of being registered by a GP during their baseline appointment. Participants accessed both pharmacological and non-pharmacological self-management content. We further explored the acceptability of the intervention to people with asthma in a mixed-methods process analysis which will be published separately.
Estimates of effect size demonstrated that participants who received the intervention and completed follow-up measures showed improved and clinically relevant quality of life and asthma control. The order of magnitude of the mean between-group improvements in the patient reported measures of control (ACQ) and asthma-related QOL (AQLQ), although not statistically significant with the sample size of this feasibility study, was comparable to that reported in controlled studies of pharmacological [32] and non-pharmacological [8] interventions in asthma, and so justify a confirmatory study with a fully-powered sample.
There was no suggestion of an effect on physiological measures of lung function. These results are in line with previous studies of behavioural self-management interventions in primary care adults with asthma (such as BREATHE, RAISIN), and demonstrate the importance of interventions targeting outcomes that incorporate elements of functional wellbeing (disease specific quality of life, subjective symptoms), rather than solely focusing on objective, physiological measures that are not correlated with quality of life.
The effectiveness of our intervention could be further increased according to findings from our process evaluation. This analysis, which will be reported separately, broadly agrees with previous research [19] in finding that many patients consider their asthma to be 'wellcontrolled' despite having important levels of symptoms and quality of life impairment on validated questionnaire. This implies that many people had become accustomed to their ongoing symptoms and had altered their life to try to reduce their impact, using denial as a coping mechanism. As a consequence, the means of appropriately targeting and framing selfmanagement interventions should be carefully considered in future work, focussing on  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  maintaining good health rather than improving poor health. Some of our findings (such as the association between quality of life improvement and ongoing intervention engagement) demonstrate that framing content as positive and not focusing on illness -for example 'How to keep your breathing healthy' rather than 'How to reduce asthma symptoms' may lead to an acceptable, engaging intervention that benefits this patient group.
There were some limitations to this small feasibility study. Although our researchers and statisticians were blind to group allocation, patients would have known that they were allocated to the intervention rather than the usual care control. This is common in complex behavioural interventions. Furthermore, although we endeavoured to recruit participants across a broad demographic range, the reach of our intervention could be improved. While the reach of digital interventions improves as digital literacy increases nationally, care must be taken to ensure that 'digital transformation' of NHS services does not entrench healthcare inequality, by facilitating a 'digital divide' that fails to provide adequate health and social care to those who do not have the digital skills to benefit.

Protocol Information
This protocol describes the My Breathing Matters trial and provides information about procedures for entering participants. The protocol should not be used as a guide for the treatment of other participants; every care was taken in its drafting, but corrections or amendments may be necessary. These will be circulated to investigators in the trial, but sites entering participants for the first time are advised to contact the Centre for Applications of Health Psychology to confirm they have the most recent version.

Online registration and randomisation:
• Standardised advice given to all patients (Asthma UK information leaflet)

Intervention group (DI ; N = 40) Patient has access to My Breathing Matters
• Online training in appropriate asthma medication use and access to self-management support • Online breathing retraining • Access to online lifestyle modification programs within intervention (physical activity, weight reduction, smoking cessation) • Resource use monitored.
• Qualitative interviews to discuss experience of intervention Practices identified by the PCRN: • 6 practices diversely sampled by size, rural-urban location and Index of Multiple Deprivation

Questionnaire assessment (3 months)
• To be completed via post; second pack will be sent out after 2 weeks if no response; AQLQ completed over the phone with blinded researcher if still outstanding after a further 2 weeks.

BACKGROUND
The UK has one of the highest prevalence's of asthma in the world; nearly 6% of the UK population have asthma, comprising 5.4 million people, most of whom are managed in primary care [1] . Hospital admission and mortality rates for asthma showed improvements in the last decades of the last century, but these improvements have stalled since the millennium. Premature mortality from asthma was 1.5 times as high in the UK as in the rest of the EU in 2008, with around 1000 to 1200 deaths a year recorded since 2000. It is estimated that 90% of deaths are associated with preventable factors. Asthma is associated with high numbers of admissions and Emergency Department attendances, and it is estimated that 70% of these could have been prevented by appropriate early intervention and self-management [1] . Surveys of asthma symptoms and health status impairment continue to show that sub-optimal control is common and that at any given time the majority of asthmatics in the UK suffer potentially avoidable symptoms and quality of life impairment [2] .
Although the UK leads the world in providing guidelines for asthma management, these have been poorly implemented and people with asthma do not receive evidence-based interventions, particularly individual action plans, which are known to impact positively on outcomes [3] . Patient education and proactive self-management have been convincingly shown to improve clinical outcomes in asthma [3] and have been advocated in guidelines for 20 years [4] . People with asthma without a management plan are four times more likely to have an asthma attack needing emergency care in hospital [5] . However, a representative (Ipsos-MORI) survey carried out by Asthma UK in 2010 [6] suggested that only a quarter of people with asthma in the UK have a self-management plan. Self-management in asthma can also encompass nonpharmacological interventions to improve control and empower the patient, such as breathing exercises or lifestyle changes such as smoking cessation and weight reduction (since smoking and obesity are associated with worse prognosis in asthma [7] ).

RATIONALE AND RISK BENEFITS FOR CURRENT TRIAL
Increasingly widespread access to the internet and mobile phones [8,9] means that healthcare Digital Interventions (DIs) are accessible to the majority of patients, and can be used to provide information and support at any time the patient needs it [10] . DIs can empower patients by providing better access to personalised information, and support for active involvement in treatment and self-management [10] . A large meta-analysis found a small but significant positive effect of DIs on health-related behaviours [11] , whilst a Cochrane review found evidence that computer-based health interventions for those with chronic health conditions significantly improved knowledge, health behaviours and clinical outcomes [12] . DIs have the potential to make significant savings by automating routine aspects of patient education, monitoring and support, freeing up health professional resources for when patients most need them [13] . These savings can play an essential part in meeting the NHS QIPP agenda to achieve increased efficiency gains despite the growing demand created by an expanding and ageing population. There is accumulating evidence that DIs can deliver better and more efficient healthcare in the context of asthma. A recent systematic review by our group found that self-management DIs could improve asthma control and reduce asthma-related quality of life impairment but called for larger, more robust trials [14] . We identified several existing self-management DIs for asthma, but half of these were for children, and many were not in English. While there are numerous commercial DIs for asthma, only one has been evaluated [15] . The RAISIN pilot trial demonstrated that self-management interventions could be effective at improving quality of life and asthma control, with improvements to 'reach' and response rate by catering to patients with mild asthma but impaired quality of life through non-pharmacological means [16,17] . A Danish trial compared GP and specialist care with web-based self-monitoring with automated feedback and a stepped care medication plan (with GP advice when required) [15] . After six months those allocated to the web-based self-monitoring had greater improvement in symptoms, quality of life and lung function. The SMASHING trial in the Netherlands [18] compared usual care with web-based educational resources, self-monitoring and automated feedback on medication titration, plus some group and email nurse support. At one year the intervention group had better quality of life and lung function and more symptom free days, at no extra cost [19] .

TRIAL OBJECTIVES
The primary aim of the My Breathing Matters trial is to assess the feasibility, acceptability, effectiveness and cost-effectiveness of a DI in primary care for the self-management of asthma, in comparison to usual care (with provision of standard patient information materials produced by the charity Asthma UK).
To assess the feasibility and acceptability of a trial of My Breathing Matters, an intervention designed to assess improvements in clinical outcomes (e.g. quality of life, health resource use, lung function) of people with asthma.
Secondary research questions 1.
To assess feasibility of trial procedures including: clinical research governance, recruitment strategy, trial documents (e.g. PIS), eligibility criteria, consent/withdrawal, randomisation and blinding.

2.
To assess feasibility and acceptability of MBM intervention including: usage and engagement, adherence and completion, fidelity of providers.

3.
To assess feasibility of data analysis, including data collection, data quality, management of trial data and estimates of effect size across trial outcome measures (see primary/secondary endpoints) to inform sample size calculations for a larger phase 3 RCT.
Tertiary research questions 1. To examine intervention usage, progress and engagement to inform well-powered usage analysis in larger trial My Breathing Matters is a digital intervention for the self-management of asthma, consisting of pharmacological support, (advice about asthma reviews and personal asthma action plans, information about medication and side effects) and non-pharmacological components (stress reduction, online versions of breathing retraining courses shown to be acceptable and feasible [20] ) and optional user-selected lifestyle modifications. 80 participants will be randomised either to the My Breathing Matters programme or to a control group receiving usual care with provision of an Asthma UK information leaflet. This trial will be coordinated from UK facilitated by the PCRN's in these areas with researchers and research nurses employed at each centre.

TRIAL DESIGN
The MBM study will comprise: 1) Feasibility trial: 80 participants (40 per arm) will be recruited from practices to confirm the acceptability and feasibility of the intervention, full trial protocol and study procedures. 2) Qualitative analysis will also be embedded in to the My Breathing Matters study.

TRIAL OUTCOME MEASURES
The primary outcome of the trial will be the feasibility and acceptability of a trial of My Breathing Matters, an intervention designed to improve clinical outcomes (e.g. symptom control, quality of life) of people with asthma.
Secondary outcomes will be to assess the feasibility of measuring (and estimates of effect size to generate hypotheses and perform sample size calculations with which to test them) in the following trial measures: 3-month questionnaire packs will be mailed out with a freepost return envelope. Baseline and 12-month follow up questionnaires will be taken using paper questionnaire packs and clinical measures will be taken by a research nurse.
For 3 month and 12-month measures, a second pack will be sent out after 2 weeks if no response; followed by questionnaire measures (AQLQ) completed over the phone if still outstanding after a further 2 weeks.
Further detail regarding clinical measures: • Lung function will be measured using spirometry by an appropriately trained research nurse. Measures of lung function will be: FEV1 (forced expiratory volume in 1 second), FEV1/FVC (ratio of FEV1 to forced vital capacity), PEFR (peak expiratory flow rate). • Medication prescriptions, any changes and support provision (consultations) will be identified from patient notes reviews after completion of the 12-month follow-up. Medication use will be converted into defined daily doses of medication.
Further detail regarding patient reported measures: • Asthma-specific quality of life (AQLQ) at 12-month follow-up • Asthma symptoms will be measured using the Asthma Control Questionnaire (ACQ) • Anxiety and depression will be measured using the Hospital Anxiety and Depression inventory (HADS) • Patient enablement (PEI): will be measured using the Patient Enablement Instrument [31].
• Patient medication adherence: all participants will complete the MARS as a measure of medication adherence [2] .
In the event that the results from any of the questionnaires are of concern (e.g. demonstrate significant anxiety or depression), the research team will inform the participant of their scores and the normal range of scores, and suggest that they may wish to discuss this result with their GP.
Economic measures: • Health resource use (professional contacts, referrals, prescriptions) • Patient time/burden needed to take part in intervention • Health-related quality of life: the EuroQol (EQ-5D 5L) provides a measure of quality of life for economic analysis [5] and the ICEACAP-A measures non-health-specific quality of life in adults [21] .
• Patient costs (e.g. additional costs incurred if patient engages in extra activities beyond intervention, such as gym membership)

Tertiary outcomes (intervention group only) will include:
• Lifestyle change choice, usage and progress • Adherence to recommendations • Engagement with program and reasons for such engagement All measured over the 12 months of the study. Assessed at baseline, after 3 months and after 12 months of online training.
Socio-demographic data • Participants' age, gender, internet experience, disease status (time diagnosed), smoking habits, education level, and social deprivation indices based on postcode will be recorded at baseline Eligible participants will be identified from around 6 general practices located by primary care research networks (PCRN); Practices will be selected in order to confirm feasibility for a wider trial. Principal Investigators will be identified at the regional level, with lead GPs nominated at each site.

CENTRE SELECTION
The following documents must be in place and copies sent to the MBM Clinical Trial Coordinator (CTC) (see contact details on page 2): • The approval letter from the relevant R&D Department, following submission of the Site Specific Information (SSI) form (where required) • A signed Study Agreement (PI and sponsor signature) • Completed Signature List and Roles and Responsibilities document • Completed contacts list of all site personnel working on the Study Upon receipt of all the above documents, the MBM CTC will send a confirmation letter to the lead GP. This letter must be filed in each centre's Site File. Along with this confirmation letter, the practice should receive their trial supplies and a study pack holding all the documents required to recruit a patient into the MBM Trial.

Screening
• Electronic database searches will be conducted, and records screened by general practitioners to remove participants who meet exclusion criteria. The latest version of the participant information sheets along with the consent and screening questionnaire (including a freepost response envelope) will be posted with a letter inviting the patient to take part in the study; • Eligible patients may also be identified and referred opportunistically during routine consultations with the practice nurse or GP.
• Study advertisements will be displayed in participating practices; • Patients responding to the study advertisements will phone the research team who will conduct a minimal screening of participants by telephone using a standard set of questions (included as a supporting document) to establish whether they meet the basic inclusion criteria, e.g. diagnosis of asthma, access to internet. The research team will then send copies of patient study materials (consent form and questionnaire) to confirm eligibility.
Patients who do not respond within one month of the practice invite will be followed up by telephone. This procedure will be explicitly outlined in the invite letter sent from the practice. In addition, all invited participants will be provided with a response form to return should they wish to decline to take part in the trial. This will ask for basic demographic information and their reasons for declining, including an option not to give a reason if they prefer not to (see patient opt-out).

EXCLUSION CRITERIA
6. Asthma judged at the baseline assessment to be dangerously unstable and in need of urgent medical review (if unstable asthma is found, the patient will be referred back to usual primary care clinician for review) 7. Terminal disease or other condition which in the opinion of the family doctor makes them inappropriate to take part 8. Diagnosed with 'difficult asthma' as defined by BTS. 9. Documented diagnosis of Chronic Obstructive Pulmonary Disease (COPD) 10. Household member already enrolled on the study

Informed consent
All patients will receive a copy of the consent form in their invitation letter pack, and will be asked to complete it when contacting the research team. All patients will receive a face-to-face baseline consultation with the practice nurse where additional informed consent (see consent form) will be recorded before clinical measures, such as confirmation of asthma diagnosis, are undertaken (see primary outcomes data document).

Randomisation
Patients will be randomised at the practice where if in the intervention group they will be given access to the My Breathing Matters programme, or if in the control group will be given Asthma UK information leaflet. (see section 9.2 for further details of randomisation). Participants in the intervention group will be instructed to sign up to the My Breathing Matters programme upon which they will be notified by email of their involvement in the study. Only one patient per household will be randomised.

CONSENT
Consent to enter the trial must be sought from each participant only after a full explanation has been given, an information leaflet offered and time allowed for consideration. Signed participant consent will be obtained. The right of the participant to refuse to participate without giving reasons will be respected. After the participant has entered the trial the clinician remains free to give alternative treatment to that specified in the protocol at any stage if he/she feels it is in the participant's best interest, but the reasons for doing so should be recorded. In these cases the participants remain within the trial for the purposes of follow-up and data analysis. All participants are free to withdraw at any time from the protocol intervention without giving reasons and without prejudicing further treatment. Participants will give informed consent (including explicit consent to access relevant data from the DI and their medical records) before they first log onto MBM. They will be free to withdraw at any time simply by contacting the research team or practice staff  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  Patients will be withdrawn from the trial if: • They are no longer eligible (specialist management of Asthma or illness precluding participation) • They choose not to continue All patients who withdraw from the intervention will be asked if they are prepared to attend the final follow-up appointment with the research nurse and/or complete self-report follow-up measures. If they agree, they will be invited according to the follow-up procedure and will have the option to answer follow up questionnaires without using the website. .

INTERVENTION ARM
The online My Breathing Matters programme will comprise three main components:

Intervention components
The MBM programme will use behavioural techniques to improve functional quality of life of primary care patients with asthma, by supporting illness self-management by pharmacological and non-pharmacological means (thereby reducing risk of asthma exacerbation). The program has three main design objectives: i) to engage people who do not view themselves as having active asthma, ii) to persuade and educate users to implement appropriate pharmacological management and iii) to encourage users to employ nonpharmacological methods of improving QoL. Key features to address these objectives include: i) maintaining positive illness context throughout (i.e. promote health rather than manage illness) and offering a simple, unobtrusive interface to provide optional (and flexible) support only when needed, ii) focusing on persuading/educating users regarding the necessity, efficacy and safety of preventative asthma medication, and facilitating easy completion of an action plan with primary care support, and iii) educating users on benefits and offer psychological methods to improve quality of life (e.g. cognitive behavioural techniques for symptom management), tailored access and address patient concerns about relevant positive lifestyle changes, such as weight-loss if overweight, smoking cessation if current smoker, physical activity if inactive.
Patients in the intervention group will be offered optional support from Asthma UK helpline -as a source of asthma advice and support for those who would like personal contact outside of that offered by MBM or their healthcare team, to support selfmonitoring and lifestyle modification. This support would be the same as that offered to anyone who calls (outside of MBM) so the Asthma UK nurse team would not require additional training.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y Motivating and reassuring support messages will be sent by email to all patients in the intervention group every 2 to 4 weeks; these will encourage and reinforce patient adherence to lifestyle changes (where applicable).

USUAL CARE
All patients will be able to access the Asthma UK information leaflet following randomisation. Patients within the usual care condition will be informed that they should simply follow their usual instructions for asthma management. After the patients have completed the 12-month follow-up they will be given access to the intervention, in line with recommendations from PPI representatives.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o n l y

CAUSALITY
The assignment of the causality to trial procedures of any serious event should be made by the investigator responsible for the care of the participant using the definitions in the table below.
If any doubt about the causality exists the local investigator should inform the trial coordinator who will notify the Chief Investigator. Other clinicians may be asked for advice in these cases.
In the case of discrepant views on causality between the investigator and others, all parties will discuss the case. In the event that no agreement is made, the Ethics Committee will be informed of both points of view.

Relationship Description Unrelated
There is no evidence of any causal relationship Unlikely There is little evidence to suggest there is a causal relationship (e.g. the event did not occur within a reasonable time after administration of the trial medication). There is another reasonable explanation for the event (e.g. the participant's clinical condition, other concomitant treatment).

Possible
There is some evidence to suggest a causal relationship (e.g. because the event occurs within a reasonable time after administration of the trial medication). However, the influence of other factors may have contributed to the event (e.g. the participant's clinical condition, other concomitant treatments).

Probable
There is evidence to suggest a causal relationship and the influence of other factors is unlikely.

Definitely
There is clear evidence to suggest a causal relationship and other possible contributing factors can be ruled out.

REPORTING PROCEDURES
All adverse events should be reported. Depending on the nature of the event the reporting procedures below should be followed. Any questions concerning adverse event reporting should be directed to the SCTU in the first instance.

Pre-existing Conditions
A pre-existing condition should not be reported as an AE unless the condition worsens by at least one CTCAE grade during the trial. The condition, however, must be reported in the pre-treatment section of the CRF, if symptomatic at the time of entry, or under concurrent medical conditions if asymptomatic.

Non serious AEs
All adverse events that may be related to the study will be recorded in the relevant case report form and Adverse Event form and sent to the SCTU within one month of the form being due. As adults on average see their GP approximately 5 times per year for a variety of routine and unscheduled appointments (e.g. for medication review, self-limiting minor illnesses and long-  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y term conditions unrelated to asthma), many medical encounters are of no relevance to the study. Events that will be recorded include any judged by the study nurse to be possibly related to the study. In particular, all medical encounters related to the following medical areas or symptoms will be recorded in the Adverse Events form: Psychological morbidity: any events relating to anxiety, depression or mood disorders Respiratory morbidity: any events relating to breathing or chest symptoms Musculoskeletal, Abdominal and chest pain: any events relating to pain in these systems unless known to be associated with an unrelated pre-existing condition.
The study nurses are advised to record any event for which there is uncertainty as to whether it is study related or not, and to discuss with the local PI or CI.

Serious AEs
All SAEs (including those that are expected and related) will be reported within 24 hours of the local site becoming aware of the event. The SAE form asks for nature of event, date of onset, severity, corrective therapies given or action taken, outcome and relatedness (i.e. unrelated, unlikely, possible, probably, definitely). The responsible centre Principal Investigator will assign the relatedness and expectedness of the event. Additional information will be provided as soon as possible if the event has not resolved at the time of reporting.
A flowchart is given below to illustrate reporting procedures: • GPs or nurses will be asked to notify us via an SAE form if a participant experiences any SAEs.
• The Sponsor and main Research Ethics Committee (REC) will be informed of all related SAEs occurring during the trial according to the following timelines, where day zero is defined as the date the SAE form is initially received: • Events which are fatal or life-threatening will be reported no later than 7 calendar days after the sponsor is first aware of the reaction. Any additional relevant information must be reported within a further 8 calendar days.
• Events that are non-fatal or non life-threatening will be reported within 15 calendar days of the sponsor first becoming aware of the reaction.
• All Investigators will be informed of all related SAEs occurring throughout the trial. Local Investigators should report any SAEs as required by their Local Research Committee and/or Research and Development Office.

Follow Up and Post-study Serious Adverse Events
The reporting requirement for SAEs affecting participants applies for all events occurring up to the end of the last treatment. All unresolved adverse events should be followed by the local investigator until resolved, the participant is lost to follow-up, or the adverse event is otherwise explained. At the last scheduled visit, the investigator should instruct each participant to report any subsequent event(s) that the participant, or the participant's general practitioner, believes might reasonably be related to participation in this study. The investigator should notify the Centre for Applications of Health Psychology of any death or adverse event occurring at any time after a participant has discontinued or terminated study participation that may reasonably be related to this study.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59

ASSESSMENT AND FOLLOW-UP OF PARTICIPANTS
Measures will be administered for all participants at baseline, 3 and 12 months, unless otherwise stated (see Schedule of Observations and Procedures). Patient-reported outcome measures will be completed online. Non-respondents will receive two email reminders followed by a paper-based copy accompanied by a final telephone follow-up. All website usage (MBM measures) will be recorded automatically in Lifeguide.
An independent research nurse, blind to study allocation will complete the follow-up assessments. All patients (including withdrawn participants who have consented to follow-up appointments) will follow the procedure outlined below: • Approximately three weeks before the follow-up appointment is due, the patient will be sent a letter by the research team encouraging them to arrange an appointment with the research nurse. Appointments may be scheduled to take place in the patients' home or usual GP practice.
• If the patient does not contact the research team to arrange an appointment within two weeks of the letter being sent the patient will be followed up by telephone.
• The research team or research nurse will send confirmation of the follow-up appointment to the patient by telephone, text, email or letter (dependent on patient preference).
• If the patient indicates that they would not be willing to complete a follow-up, no further contact will be made with the patient regarding the follow-up appointment.

DEFINITION OF END OF TRIAL
The end of the trial will be defined as date of entry of final data into database.

SAMPLE SIZE
A minimum of 40 participants from two centres will be recruited to each condition during the study.

RANDOMISATION
Participants will be randomised in a 1:1 ratio to receive either usual care (control) or MBM programme with optional support (intervention). The randomisation will be stratified block randomisation by the average value of AQLQ scores in the BREATHE Trial (Thomas et al).
Remote allocation will maintain allocation concealment from both the participant and the research nurse prior to allocation, however the outcome assessor will be blinded to participant allocation.

STATISTICAL PLAN INCLUDING INTERIM ANALYSIS
A detailed statistical plan will be developed prior to data lock. No interim analysis is planned. Data and all appropriate documentation will be stored for a minimum of 10 years after the completion of the trial, including the follow-up period, in a secure location at Southampton University.
Descriptive statistics of outcome variables will be calculated in order to inform a future randomized controlled trial, including examining levels of missing data, attrition rate, retention and completion. Statistical comparisons between and within intervention and control groups will be performed to estimate variability in primary and secondary outcomes within a linear regression framework. Further exploratory analysis of patient engagement, intervention usage and possible mediators of behaviour change will be performed.

QUALITATIVE EVALUATION
The qualitative study will be undertaken both alongside participation and after participants finish using MBM. Qualitative interviews will seek to provide an in-depth understanding of the perspective of patients, to inform intervention and trial acceptability and to generate hypotheses about intervention mechanisms of action that can be tested in a larger trial.
Participants and sampling Patients will be asked whether they give permission to take part in an interview at a later point in the study. Purposive sampling will be used to select patients from the intervention group to allow for a wide range of views and experiences of the MBM programme. It is anticipated that 20 patients will be interviewed during and following the study, and patients will be selected for interviews until saturation is reached.
Interviews and qualitative analysis Factors that may facilitate or diminish the acceptability of the MBM programme, and adherence to implementation will be explored across patient and health care professional interviews. Interviews will be conducted from after the 3-month assessment. Open-ended questions will be used to elicit user perspectives and experiences of the intervention, allowing participants to freely describe their experiences and views in their own way and to focus on whatever is most salient to them.
Interviews will be audio-recorded and fully transcribed. The findings will be used to inform any modifications needed to the MBM programme or the trial procedures for a potential future full RCT.
The transcriptions will be anonymised (identifiable data removed) and participants' transcripts will be given participant numbers so that they can be easily discussed between team members whilst protecting participants' identities. To ensure that we remain open to and grounded in users' perspectives we will carry out inductive thematic analysis [22] of all textual data,  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n l y triangulated where appropriate with quantitative self-report measures and web usage data, and constant comparison and discussion among team members to reach inter-rater agreement on themes and interpretations.

CLINICAL TRIAL AUTHORISATION
This trial does NOT involve the testing of any Investigational Medicinal Products (IMPs) therefore approval from the Medicines and Healthcare products Regulatory Agency is not required.

ETHICS APPROVAL
The trial protocol will be submitted to a Research Ethics Committee (REC) recognised by the United Kingdom Ethics Committee Authority (UKECA) for review and approval. A favourable opinion must be obtained before commencement of any trial procedures (including recruitment of patients) occurs All substantial amendments must be approved by the REC responsible for the trial, in additional to approval by NHS R&D. Minor amendments will not require prior approval by the REC.
If the trial is stopped prematurely, it will not be recommended without reference to the REC responsible for the trial.
The outcome of the trial will be reported to the responsible REC within 90 days of completion of the last patient's final trial procedures. In the event of the trial being prematurely terminated a report will be submitted to the REC within 15 days. A summary of the Trial Report will be submitted to the responsible REC within one year of completion of the last participant's final study procedures.
The investigator must ensure that participant's anonymity will be maintained and that their identities are protected from unauthorised parties. On CRFs participants will not be identified by their names, but by an identification code.

3 CONFIDENTIALITY
Initial practice database searchers will be conducted on practice computer systems and subsequent study invitations will be the responsibility of the practice, to maintain the confidentiality of potential participants. Participants' identification data will be required for the registration process, which will be completed with the support provider at the baseline screening appointment. Participants will be informed that the research team will have access to their study data, and that personal information such as their name, telephone address and email address will be stored and used by the research team to stay in touch with them throughout the study.
The primary data outcomes measured by the support provider (such as asthma quality of life and current medication) will be stored on an independent, secure server. All other data (secondary self-report measures and intervention data) will be stored on dedicated secure spaces behind a firewall on password protected computer located in secure university buildings at the University of Southampton. The data is backed up daily. Confidentiality of all data entered into the DI will be maintained by following best practice in NHS IT protection and data security systems (e.g. regarding use of https, storing data behind the university firewall etc). Access to the website will be via username and password. The use of strong passwords will be enforced. The communication between the person entering data through a web browser and the server will be through a secure internet connection (HTTPS). The research team will adhere to the Data Protection Act 1998.
Digital records will be stored on a university computer in a password protected file. All data will be linked with the participant's study ID which does not include any personal or identifiable information, such as name or D.O.B. Study IDs will be linked with patient's names, email addresses telephone numbers in a separate, securely stored file.
At the end of the study, data will be anonymised and stored on a password protected computer located in secure university buildings and appropriately backed up. Confidentiality of all data entered into the DI will be maintained by following best practice in NHS IT protection and data security systems.

INDEMNITY
The sponsor of the trial is University of Southampton. University of Southampton insurance will apply. The lead authors of the protocol are the Chief Investigator (Prof Lucy Yardley), and Prof Mike Thomas, who are both university employees and as such will be covered by the University insurance.

SPONSOR
University of Southampton is acting as the sponsor for this trial. The My Breathing Matters coordinating team has been delegated duties by the Sponsor relating to: submissions to regulatory authorities and GCP.

DEVIATIONS AND SERIOUS BREACHES
Any trial protocol deviations/violations and breaches of Good Clinical Practice occurring at sites should be reported to the MBM trial coordinator and the local R&D Office immediately. The MBM trial coordinator will then advise of and/or undertake any corrective and preventative actions as required.
All serious protocol deviations/violations and serious breaches of Good Clinical Practice and /or the trial protocol will immediately be reported to the regulatory authorities and other organisations, as required in the Medicines for Human Use (Clinical Trials) Regulations 2004, as amended.

AUDITS AND INSPECTIONS
The trial may be participant to inspection and audit by University of Southampton, under their remit as sponsor, the trial coordinating centre as the Sponsor's delegate and other regulatory bodies to ensure adherence to ICH GCP, Research Governance Framework for Health and Social Care, applicable contracts/agreements and national regulations.

TRIAL MANAGEMENT
The Trial Management Group (TMG) is responsible for overseeing progress of the trial. The day-to-day management of the trial will be co-ordinated through the trial coordinating centre and oversight will be maintained by the Trial Steering Committee, with an experienced independent chair and representation of patient representatives, Asthma UK and Blood Pressure UK.
The Steering Committee will meet twice a year throughout the programme to provide strategic guidance and independent monitoring of progress and professional conduct. We will encourage in person attendance at all these meetings where possible, but will also provide for attendance by teleconference when necessary, and will circulate papers and minutes before and after meetings for communication with those who cannot attend for any reason.

PUBLICATION POLICY
All publications and presentations relating to the trial will be authorised by the Trial Management Group and will follow an agreed publication policy. Dissemination of our work will be via multiple pathways: a) to the scientific community through presentation at national & international conferences and regular publication in highly cited and open access peer reviewed journals. b) to clinical and academic colleagues via professional societies: links with the following societies will be exploited to raise the profile of this work: Royal College of GPs and Physicians, Society of Behavioural Medicine, British Sociological Society Medical Sub-group. c) to patients via patient groups: We will work with Asthma UK to disseminate our results. d) to participants: All participants will be sent an accessible summary of the findings from the study that they took part in within six months of study completion. e) to relevant NHS organisations and healthcare providers (e.g. Clinical Commissioning Groups, NHS Choices, UCL partners).  Methods 3a Description of pilot trial design (such as parallel, factorial) including allocation ratio 7 Trial design 3b Important changes to methods after pilot trial commencement (such as eligibility criteria), with reasons -4a Eligibility criteria for participants 7 Participants 4b Settings and locations where the data were collected 7 4c How participants were identified and consented 7 Interventions 5 The interventions for each group with sufficient details to allow replication, including how and when they were actually administered 8 6a Completely defined prespecified assessments or measurements to address each pilot trial objective specified in 2b, including how and when they were assessed 10 Outcomes 6b Any changes to pilot trial assessments or measurements after the pilot trial commenced, with reasons -6c If applicable, prespecified criteria used to judge whether, or how, to proceed with future definitive trial -7a Rationale for numbers in the pilot trial 8 Sample size 7b When applicable, explanation of any interim analyses and stopping guidelines -Randomisation: 8a Method used to generate the random allocation sequence 8 Sequence generation 8b Type of randomisation(s); details of any restriction (such as blocking and block size) 8 Allocation concealment mechanism 9 Mechanism used to implement the random allocation sequence (such as sequentially numbered containers), describing any steps taken to conceal the sequence until interventions were assigned 8   *We strongly recommend reading this statement in conjunction with the CONSORT 2010, extension to randomised pilot and feasibility trials, Explanation and Elaboration for important clarifications on all the items. If relevant, we also recommend reading CONSORT extensions for cluster randomised trials, non-inferiority and equivalence trials, non-pharmacological treatments, herbal interventions, and pragmatic trials. Additional extensions are forthcoming: for those and for up to date references relevant to this checklist, see www.consort-statement.org.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46