Respiratory Medicine 129 (2017) 39e45
Contents lists available at ScienceDirect
Respiratory Medicine journal homepage: www.elsevier.com/locate/rmed
Patient-perceived acceptability and behaviour change beneﬁts of inhaler reminders and adherence feedback: A qualitative study Juliet M. Foster, PhD a, *, Helen K. Reddel, MB BS, PhD a, 1, Tim Usherwood, BSc MB BS MD b, c, Susan M. Sawyer, MB BS, MD d, e, f, Lorraine Smith, PhD g, 1 a
Woolcock Institute of Medical Research, University of Sydney, Sydney, Australia Department of General Practice, Sydney Medical School e Westmead, University of Sydney, Sydney, Australia George Institute for Global Health, Sydney, Australia d Centre for Adolescent Health, Royal Children's Hospital Melbourne, Melbourne, Australia e Department of Paediatrics, The University of Melbourne, Melbourne, Australia f Murdoch Childrens Research Institute, Melbourne, Australia g Faculty of Pharmacy, University of Sydney, Sydney, Australia b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 March 2017 Received in revised form 10 May 2017 Accepted 25 May 2017 Available online 26 May 2017
Introduction: Little is known about patients' perceptions of electronic inhaler reminders, which have emerged in recent years as adherence promotion aids. This study explored asthma patients' attitudes toward the acceptability and utility of inhaler reminders. Methods: Participants from a 6-month cluster randomized controlled trial who received reminders for missed doses via SmartTrack adherence monitors (Adherium Ltd) were interviewed to explore their perceptions; interviews were audio-recorded, transcribed and analysed thematically. Results: 18 participants (50% male, mean age 39 years [range 17e68]) were interviewed. Three themes were identiﬁed. Acceptability and Feasibility: Interviewees found the monitor easy to use. For some, concerns about the monitor itself affected adherence, e.g. leaving it at home to avoid breakage. Positive features included that reminders played only for missed doses, and the choice of reminder tunes. Utility and Behavioural Impact: Interviewees described reminders as an effective “training” tool for adherence, encouraging habit-formation, behaviour change and attitude change. Reminders were considered less acceptable or useful by participants who preferred taking medication only when symptomatic or who doubted the necessity or safety of their medication. Sustainability: Some interviewees reported sustained behaviour change, supported by reminders, through the establishment of routine or via experiential learning that good adherence improved their asthma. Other interviewees wanted ongoing support (i.e. reminders or substitute adherence cues) after study end. Conclusion: Patients with asthma found 6-months' use of reminders and adherence feedback acceptable and useful for improving their adherence attitudes, adherence behaviours and conﬁdence in asthma selfmanagement. Some patients may beneﬁt from ongoing adherence support. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Medication adherence Asthma Behaviour change Patient satisfaction Ambulatory monitoring Reinforcement
Abbreviation list: RCT, Randomized Controlled Trials. * Corresponding author. Clinical Management Group, Woolcock Institute of Medical Research, PO Box M77, Missenden Road, NSW 2050, Australia. E-mail addresses: [email protected]
(J.M. Foster), [email protected]
sydney.edu.au (H.K. Reddel), [email protected]
(T. Usherwood), [email protected]
(S.M. Sawyer), [email protected]
(L. Smith). 1 Equal contributors. http://dx.doi.org/10.1016/j.rmed.2017.05.013 0954-6111/© 2017 Elsevier Ltd. All rights reserved.
Digital technologies which promote illness management are predicted to expand in the coming years. In the adherence space, novel technologies to promote medication use have emerged, such as electronic monitors with in-built reminders [1,2]. Reminders for missed doses have been shown to improve adherence with inhalers and clinical outcomes in asthma patients in randomized controlled trials (RCT); in a pragmatic 6-month RCT in adults with moderate-
J.M. Foster et al. / Respiratory Medicine 129 (2017) 39e45
severe asthma, we showed that inhaler reminders for missed doses led to improved adherence (average 73% vs 46%, p < 0.0001) and reduced severe exacerbations , similar results have been observed in children after ED presentation . There has been a concerted call for greater involvement of patients in the design of new technologies , since the ultimate success of innovations depends on their acceptability and perceived usefulness by patients . While effective implementation and long-term persistence with inhaler monitors and reminders will similarly be driven by patient acceptability, there is little research about this. A few studies have explored patient attitudes to pill box adherence monitors and reminders in patients with HIV [7,8], or kidney transplants , but these cannot necessarily be extrapolated to inhalers or to patients with respiratory disease. Two studies exploring patient-acceptability of monitors and reminders for inhalers were restricted to a 7e30 day window of exposure to the monitor and reminders [10,11], and implemented within a research setting. These data provided little information about the feasibility of longer term use in real world settings nor about the process of patient behaviour and attitude change, nor about barriers to change in response to monitoring, reminding and adherence feedback. The aim of this study was to explore asthma patients' attitudes toward the acceptability of electronic adherence monitoring and the usefulness of medication reminders for missed doses and adherence feedback after six months' use. 2. Methods
website using an internal mobile phone SIM card. Patients could customize reminder ringtones and reminder times, cancel individual reminders, or switch reminders off completely. The monitor screen displayed the time since the last dose in minutes, hours, days or weeks . Each month, patients received an automated email to remind them to view a graph of their daily preventer use on a website (Fig. 1); a similar email was sent to their GP. The patient also received telephone contact from the research team every two months to collect study outcomes for the RCT. 2.4. Study design This was a qualitative study utilizing in-depth, semi-structured interviews to explore asthma patients' experiences of using an adherence monitor with reminder and feedback capability. 2.5. Interviews Interviews of approximately 45 min duration were conducted by telephone by JMF. An interview guide (Appendix A) was used to explore patients' perceptions of barriers, facilitators, usefulness and impact of using the monitor and its reminders. Participants were assured that the content of the interview would be kept strictly conﬁdential. After each interview, ﬁeld notes were recorded of emergent themes, the situational context, and/or new topics for follow up in subsequent interviews. Interviews continued until thematic saturation occurred (i.e. when three interviews had been conducted with no new themes emerging) .
2.1. Recruitment of participants for the randomized controlled trial 2.6. Analysis Patients aged 14 years and older with moderate-severe asthma (prescribed twice daily ICS/LABA for 1 month prior to enrolment; no asthma exacerbation in the last month; and not currently participating in an asthma study) and sub-optimal asthma control (Asthma Control Test 19) ; were enrolled by their GP in a pragmatic 6-month cluster randomized controlled trial (RCT) of GPdelivered interventions for improving patients' adherence with inhaled ﬂuticasone propionate/salmeterol xinafoate (hereafter referred to as ‘preventer’) therapy [3,13]. 2.2. Recruitment of participants for the interview study At the end of the RCT, a sub-sample of the participants who received an inhaler-based ‘reminder and feedback’ intervention (see below) were invited to take part in an interview. Participants were selected using theoretical sampling to represent a range of: gender, age, education and self-reported satisfaction with monitoring and reminders, the latter recorded by participants at RCT exit on a 100 mm long visual analog scale from ‘not at all satisﬁed’ to ‘extremely satisﬁed’. We also aimed to include one or two patients who dropped out of the study to capture any negative views on the monitor that may have contributed to their decision to drop out. For half of the participants selected for interview, their GP had also received training to deliver an ‘adherence counselling’ intervention (an intervention that was found to be ineffective for improving medication use or adherence attitudes) [3,13]. 2.3. Reminder and feedback intervention Patients received twice-daily reminders for missed preventer doses from the SmartTrack (Adherium, Auckland, New Zealand) adherence monitoring device (hereafter referred to as “monitor”) that we have previously validated and ﬁeld tested . The monitor clipped onto the patient's preventer inhaler, recorded the date/time of all actuations and uploaded the data each month to a secure
Recordings were transcribed verbatim and transcripts were anonymised. We adopted the thematic approach to analysing qualitative data . Interviews were analysed systematically by JMF, with the ﬁrst step being familiarisation (reading and rereading transcripts and ﬁeld notes) to gain an immediate impression of data, and analytic memoing to document the researcher's reﬂections. Initial descriptive coding was then carried out which later progressed to pattern coding , and identiﬁcation of themes and subthemes. The analysis was carried out iteratively as each new interview was completed. All codes and themes were continuously checked and re-checked by JMF, with regular in-depth discussion of coding and themes with co-authors HKR and LS to ensure they represented the content of the interviews. All co-authors commented on the themes prior to ﬁnalising the analysis. 2.7. Ethical considerations The Human Research Ethics Committee (HREC) of the University of Sydney approved the interview study protocol (ref. no. 05-2009/ 11731). Written information was provided, and written informed consent was obtained from all participants. 3. Results Eighteen participants (50% female; mean age 39 [range 17e68] years) were interviewed (Table 1). Of patients receiving reminders in the RCT , 7 were lost to follow up and 2 dropped out, one of whom agreed to be interviewed. The latter participant had not provided a score on satisfaction with monitoring at study withdrawal. The mean participant-reported satisfaction (scored 0e100) with the monitor was 82 (range 32e100). Interviews lasted an average of 47 min (range 21e90 min). Analysis of the transcripts revealed three principal themes: ‘Feasibility and Acceptability’, ‘Utility and Behavioural Impact of
J.M. Foster et al. / Respiratory Medicine 129 (2017) 39e45
Reminders’ and ‘Sustainability’. Illustrative quotes are presented below with a pseudonym, gender and age of interviewee. The nature of iterative, semi-structured interviews means that every topic will not necessarily be discussed with every interviewee, but to provide an impression of reporting frequency we use the terms “a few” “some”, and “most” in the results to indicate that “1 or 2”, “3 to 4”, and “14 or more” participants respectively reported an experience. 3.1. Feasibility and acceptability Interviewees generally found the monitor easy to use (“It was easy to use, it was just there on my dressing table and when the old inhaler ﬁnished I just opened it up and popped the new one in.” Michelle, F, 65 yrs) and likeable (“I liked it just the way it was.” Tania, F, 28 yrs). 3.1.1. Reminder feasibility Some interviewees missed reminders due to not hearing a ringtone, forgetting to take their inhaler and monitor with them (e.g. on weekends away), or due to changes in routine e.g. because the reminder rang before returning home from work. A disadvantage for some was seen as an advantage by others; i.e. the fact that the inhaler and reminder were conjoined prompted interviewees to take their treatment “straight away” and could be a superior method to an SMS-based reminder (“If I hear a text message come in you're like ‘Oh, I'll just ﬁnish what I'm doing, then I'll read it”. And you realize it's four hours later and the text message is still sitting
there.” Beth, F, 33 yrs). The monitor's on-screen feedback stating ‘time since last puff taken’ was considered a helpful back-up for missed reminders and to avoid double-dosing (“If I couldn't remember whether I'd taken it or not, I could go and have a look at when I'd taken it last.” Tania, F, 28 yrs). 3.1.2. Social compatibility and privacy A few participants thought that the reminder was accepted by people sharing their home while some were concerned that it might disturb them. There was also a concern that the monitor and/ or reminder might draw attention to asthma in public places (“The reminder would go off, and you'd be looking for somewhere to hide in a corner to take it if you're out.” Jane, F, 57 yrs) or that at home it could provoke unwanted interference in the patient's asthma care by housemates or partners (“Somebody in the house would usually say ‘Your thingo's going off’ (laughs) and I was like ‘Yeah, yeah’ ...I understand that they are doing it to help out but I will get there.” Sam, M, 38 yrs). 3.1.3. Monitor physical characteristics A few found the monitor size satisfactory (“It just ﬁt in my pocket.” Sarah, F, 17 yrs) and some found it bulky. An older individual (F, 65 yrs) with arthritis thought the buttons on the monitor too small and a younger interviewee (F, 24 yrs) found the monitor unattractive and suggested providing a range of ‘skins’ so users could choose the monitor colour.
Fig. 1. Example of the preventer use graph on the website. Feedback, in the form of a graph of daily preventer use could be viewed by the patient on a password-protected website. The graph showed the number of inhalations prescribed (dotted line) and the number of inhalations the patient took each day over a 14-day period. A reminder email containing a copy of the most recent 14-day graph was e-mailed to the patient, and their GP, every 30 days.
J.M. Foster et al. / Respiratory Medicine 129 (2017) 39e45
3.1.4. Monitor software features Features that were liked included the range of 14 reminder ringtones which was considered good by some and limited by a few. Interviewees liked that a favourite ringtone could be selected, that ring times could be personalised for weekday/weekends and that ringtones only rang for missed doses. The ability to switch reminders off completely was also seen as a useful option. Suggested features, which were not provided, included a ringtone volume adjustment, a ringtone snooze button, a seven day ring time option (as opposed to weekday and weekend provided), and a time zone change option for travel. 3.1.5. Monitor feasibility One participant (Michael, M, 27 yrs) mentioned two monitor failures during the six month period (one monitor counted more puffs than he had taken and the screen of another monitor froze) but despite this he still thought “reliability was good”. Three participants had to do manual uploads from the monitor (using the onboard menu) in areas of poor network coverage. Amira (F, 21 yrs) reported putting a reliever inhaler inside the monitor at times in order to record how often she was using her reliever inhaler (which would have resulted in an inﬂated preventer adherence record for the RCT). A few participants were concerned that the monitor might be tampered with by others sharing their home, or damaged by everyday wear and tear. This resulted in moving the inhaler (and monitor) from its usual location e.g. in the bathroom, or leaving it at home resulting in changes to usual medication-taking routines (“A couple of times when I [travelled] just over night I didn't take it. I just had [my reliever] with me.” Sam, M, 38 yrs). 3.1.6. Feasibility and acceptability of web-based adherence graph Most participants viewed the adherence graph and found the graphs easy to understand but three had difﬁculty fully deciphering it, which limited its perceived usefulness and may have created misunderstandings. One interviewee suggested replacing the graph and simplifying its information by showing a simple target number of prescribed puffs for each day and the proportion of puffs achieved for each day. 3.2. Utility and behavioural impact of reminders 3.2.1. Effective reminding Most interviewees found the monitor to be effective for reminding them to take their preventer. They integrated it into their daily life, for example, by using the reminder as an alarm clock as well as a medication reminder. A few recommended the
reminder to others due to its usefulness (“I tell [many people]‘Why you don't do the program?’[sic]...because for me it was helping me. For me it worked.” Catarina, F, 68 yrs). Although individuals tended to choose ringtones they liked, some talked about the importance of hearing an ‘annoying’ ringtone for increased effectiveness (“It needed to be annoying! It's not necessarily a negative thing. You need to be annoyed to do things sometimes!” Peter, M, 45 yrs).
3.2.2. Behaviour change and habit-formation Reminders changed the adherence behaviour of interviewees during the six months of monitor use (“I was never particularly stringent about taking the preventative...But being on the study I had to take the preventative every day and that little thing went off at me all the time telling me when.” Tania, F, 28 yrs). Interviewees described this change occurring through habit-formation (“To get into the habit of taking it morning and night. Those reminders were fabulous, I would've forgotten it if I didn't have it.” Tania, F, 28 yrs). Formation of these new adherence habits, through the use of reminders, was reported to occur as early as 1 month and as late as 6 months (study end). In the RCT, on a group level, adherence improved within one month of receiving the reminder device . The reminder was thought to work as a “training programme” for good adherence (“I wouldn't have done it if [my GP] had sat me down and said ‘Right you've got to take this every day and every night’. I still would never have got there! And the key was the [reminder] … I called it a training programme. Like I feel like I'm trained now, to take [my preventer].” Beth, F, 33 yrs). Another positive impact of reminders was the improved adherence with other non-respiratory medications, reported by two interviewees (“If I didn't get into the routine of taking the [preventer] I probably wouldn't be taking that [neck and back medication] properly neither!” Jane, F, 57 yrs).
3.2.3. Change in health and quality of life Participants noticed improvements in asthma symptoms (e.g. breathlessness, cough), reaction to triggers (e.g. dust), exacerbations, need for reliever medication and positive impacts on their ability to work, exercise and take part in family life. Family members were perceived to have become less concerned about the interviewee's asthma (“My ﬁanc e was very, very, very worried about my asthma…me diligently taking my medication has assuaged his worry a lot.” Tracey, F, 24 yrs). Tania (F, 28 yrs) said that she could now live more like a healthy person (“I can just go and do stuff like everybody else, like people without asthma.”).
Table 1 Interviewee characteristics. Patients
n ¼ 18
Female (%) Age, years (mean: range) Living in socially disadvantaged areaa (%) Highest level of education: Post-secondary qualiﬁcation/Secondary/Still at school (%) ICS daily dose ﬂuticasone propionate equivalent, mg (mean ± SD) Asthma Control Testb (mean ± SD) Satisfaction with the monitorc, mean: range
50 39: 17-68 33 72/22/6 694 ± 291 14.6 ± 3.8 82: 32-100
All data were collected at entry into RCT except ‘Satisfaction with the monitor’ which was collected at RCT study end. Months between study end and interview: mean 6.2 (range 2e10). Months between RCT entry and interview: mean 11.9 (range 7e16). a Social disadvantage at home address: ‘Disadvantaged’ SEIFA Quintile 3, ‘Advantaged’ SEIFA Quintile: 4e5 . b Asthma Control Test (scored worst:5ebest:25) 19 or less indicates suboptimal control . c “How satisﬁed are you with the [monitor] overall?” Scored not at all satisﬁed ¼ 0 to extremely satisﬁed ¼ 100, data (n ¼ 17) collected at RCT study end (data missing for one patient who dropped out of the RCT but agreed to interview); Mean (range) for total RCT sample receiving reminders was 83 (32e100).
J.M. Foster et al. / Respiratory Medicine 129 (2017) 39e45
3.2.4. Change in adherence and self-management attitudes Participants described a shift in their adherence attitudes during the six-month use of the reminder because it helped them create a clearer link between poor adherence and poor asthma control (“I, quite quickly into the piece, noticed that by taking [the preventer] much more frequently and by having the reminder feature I was actually a lot more healthy.” Tim, M, 38 yrs). Jane (F, 57 yrs) reported that a concern about the monitor ringing in public led her to switch the reminders off, followed by a period of poor asthma control, and that this experience had led her to switch the reminders back on, and ultimately to a positive change in attitude toward adherence (“I got all tight chested and wheezy and I'd be thinking [I haven't taken my preventer] and then it would dawn on me.”). A few had become more conﬁdent in managing their asthma “I can control it myself now. Like, I don't have to go to the doctors when I feel like it's coming on.” Jane F, 57 yrs). 3.2.5. Adherence barriers limiting the utility of reminders Reminders were less effective for prompting behaviour change in those who had doubts about the necessity or safety of their preventer treatment (“I didn't take [my preventer] as much because I didn't think it was necessary to be taking it at the time. But I was also hesitant to take it ‘cos I don't know if it's necessarily good to be taking medication all the time.” Brett, M, 29 yrs), or if they did not consider their asthma serious or life threatening. Interviewees who disliked routine, had an existing routine or preferred to take medication in their own time considered reminders unnecessary (“I didn't need it to remind me, because I knew I was gonna take it every day. So it was more the fact that I just didn't want it going off and me having to do it at the time. I'd rather do it to my own time.” Jamie, M, 62 yrs). Reminders were also considered unnecessary in individuals who wished to use their preventer only in symptomatic periods (“It was not necessary for the reminder… Once the symptoms go away -they can go away for a month at a time with no issues- I didn't see the point in taking it.” Brett, M, 29 yrs). Interviewees such as these tended to dislike and/or switch off reminders. Three interviewees reported switching their reminder off; two of these did so immediately after study start. 3.2.6. Interaction with health professionals Use of reminders was inﬂuenced by the GP knowing the interviewees' medication use (“[After switching the reminders off for 3 weeks] I started to think about if the doctor sees my results, then I thought that maybe I have to use this… So then I was like ‘I'll just use it again and see how I go’… And to tell you the truth I did see a difference. I actually maintained my asthma.” Amira, F, 21 yrs). Individuals with attitudinal barriers to adherence such as concern about side effects considered that talking to a doctor about asthma and/or their medication concerns provided more useful information than a reminder alone (“[With the reminder there] wasn't really much advice given. After I'd ﬁnished the study I went back and spoke to a female [GP] and she recommended me a new [preventer] since I had problems with the old one. I thought that was actually more helpful to be honest because she seemed to have more information available [as] to what the problem was.” Brett, M, 29 yrs). Reassurance from a health professional about treatment safety was also considered key for supporting reminder use (“If I could be assured that in the long term it's a risk-beneﬁt, that there would be no risk, then I would take my asthma [preventer] more often… If I was assured then I would use the reminders to remind me.” Peter, M, 45 yrs). 3.2.7. Utility and behavioural impact of web-based adherence graph For a few individuals the graph was not of any interest, or was too de-contextualised to be helpful (“[I wanted to know] how my
asthma had improved over that six month period or just some recommendations.” Brett, M, 29 yrs). For others the graph was helpful for identifying temporal patterns of preventer use or visually reinforcing adherence behaviour, which was important for accepting their poor adherence (“I used to miss [my preventer] all the time, even though I swear I was taking it every time the alarm went off. So [the graph] made me realize.” Beth, F, 33 yrs). 3.3. Sustainability 3.3.1. Sustained adherence After six months of use, RCT participants were required to return their monitor. Positive changes in adherence behaviour persisted for some, even after the monitor had been returned, through the establishment of routines (“The [monitor] got me into a routine of using a [preventer]. Something I'd not done in the past. So that's a routine that I've been able to follow post project, which has deﬁnitely beneﬁted me.” Tom, M, 42 yrs) or via experiential learning that good adherence was beneﬁcial (“It's like I've been trained to do that and I keep doing that now… Because it made me realize it's better for me you see.” Catarina, F, 68 yrs). Interviewees reporting sustained behaviour change, supported by reminders, said they had made a commitment to adhere, and thus noticed the link between adherence and well-controlled asthma (“So I put the effort in and the rewards are shown in my health.” Tania, F, 28 yrs). 3.3.2. Ongoing adherence support Other interviewees talked about needing ongoing support to prevent relapsing after removal of reminders at six months, as new adherence habits had not yet stabilized, or they had become somewhat reliant on the reminder as a medication-taking cue (“But even now I still have to think whether I've taken it or not… so I've sort of gotta [look after] the habit myself.” Tania, F, 28 yrs). Around half wanted to keep the monitor and use it, either longterm or periodically when their adherence went “off track”. Two interviewees mentioned being willing to purchase a monitor; one said he would pay between AUD30-100 and one between AUD200400. A few interviewees used other medication-taking cues in place of reminders after returning the monitor; one put the preventer in a conspicuous place and one set an alarm on their phone. Beth (F, 33 yrs) described collaborating with her partner on replacement adherence strategies (“I said ‘How am I gonna remember to take this?’… I said ‘I wanna do this regularly’. And that's when he suggested ‘Well why don't you just put it next to your toothbrush?’” Beth, 33 yrs). Continued access to the online adherence graph (with a nonreminding monitor) was also suggested as a formal follow-up support tool after removal of the reminders. 3.3.3. Peer mentoring A few interviewees who had undergone a transformation in their own adherence beliefs shared their discovery with others creating a snowball effect on individuals who had not taken part in the RCT (“[My cousin's wife said] ‘I've gone on [to a preventer] like you recommended. My asthma is tenfold better’…And I said: ‘Are you taking it regularly?’ and she said: ‘Yep! Taking it every day!’.” Beth, F, 33 yrs). 4. Discussion This qualitative study showed that people with moderate to severe asthma considered six months use of inhaler monitoring and adherence feedback acceptable, feasible and useful; it improved their adherence attitudes and behaviours through habit formation, routine and experiential learning, and resulted in increased conﬁdence in asthma self-management and asthma control. Speciﬁcally,
J.M. Foster et al. / Respiratory Medicine 129 (2017) 39e45
participants described a shift in their adherence attitudes during their six-months’ use of the reminder device because it helped them create a clearer link between poor adherence and poor asthma control; this realisation may be particularly important for sustained behaviour change. For some, ongoing adherence support beyond 6 months was needed to maintain adherence. Reminders were considered less useful, less acceptable and were more often switched off by participants who disliked routine, had an existing medication-taking routine, or who preferred to take their preventer only in symptomatic periods. Change in adherence attitudes and behaviours often did not occur in these participants, nor in some who doubted the seriousness of their asthma or the necessity or safety of their preventer treatment. Participants with such attitudinal barriers to adherence articulated that the information they needed to change their adherence was not provided by the inhaler reminders and adherence graph alone. In clinical settings, it would appear that such patients need a different level of engagement with their GPs prior to considering, or in conjunction with adherence reminders. Our ﬁndings ﬁt well with the social cognitive and health belief models of health behaviour , which have been shown to be effective for designing interventions to improve adherence . For example, interviewees reported changes in their outcome expectancies (e.g. Taking my preventer will improve my asthma), selfefﬁcacy (I can control my asthma myself), acquisition of behavioural skills (I have been trained now), and behavioural intention (I want to take my preventer to stay well). But many who had negative evaluations of the risk of taking medications or who thought they had low need for medication or low disease risk did not change adherence attitudes or behaviours. Interestingly, when the reminders were turned back on after a decline in asthma control, the personal experience of better asthma control reinforced positive adherence beliefs. Further, the greater awareness of asthma at home appeared to have encouraged collaboration with family or friends around adherence strategies. The impact of reminders on family members could be a fruitful avenue for future research. Participants identiﬁed both positive and negative aspects of inhaler monitoring. The feasibility of monitoring could be hampered when fear of breakage led to the monitor (and inhaler) being deliberately left at home. While some participant comments were speciﬁc to the model and data network used (e.g. monitor bulk, perceived fragility, upload delays) and may not be relevant to current models, other comments were generalizable. For example, privacy concerns, e.g. that reminders might draw attention to asthma, led some to switch reminders off. For some participants, a graph displaying daily preventer use over the previous two weeks was considered superﬂuous, but for others it visually reinforced adherence behaviour, and was important for coming to accept their poor adherence. Strengths of the study include the broad range of participants interviewed (e.g. age, education and end-of-study device satisfaction), and the rich data obtained about the satisfaction, acceptability and effectiveness of reminders as well as concerns about their social compatibility; ﬁndings which mirror those of previous studies [10,11]. In addition, our research provides novel insights into patients' perceptions of the attitude and behaviour change mechanisms (and barriers to these) which occurred during six months use of medication monitoring, reminding and feedback in a real-life study. Our ﬁndings are important for supporting the successful implementation of these interventions in clinical settings. Limitations include that patients had been informed that the RCT would include reminders and only one of 18 interviewees used the monitor for less than six months (due to dropping out from the RCT). While we deliberately selected participants with a wide range of satisfaction scores (from 32 to 100 out of a possible 100), and a
range of concerns were articulated during interviews, this could have reduced our ability to capture negative responses. The average gap of 6 months between RCT end and interview may have resulted in reduced recall by interviewees, but it allowed us to explore their experiences of reminder cessation. Further, all interviewees had moderate-severe asthma and 72% had a post-secondary qualiﬁcation, so our ﬁndings may not fully apply to people with milder asthma and/or lower education. Our ﬁndings suggest a number of opportunities for future research. These include exploring: the duration of behavioural beneﬁt gained from inhaler reminders, and the impact of medication reminders on participants' family and friends. More information is needed on the most effective length of reminder interventions, the best format for displaying adherence data to patients and the impact of different types of health professional support on reminder outcomes (e.g. from reviewing adherence data to discussing patients' treatment concerns or uncertainties). Finally more data on the acceptability of sharing adherence data between clinician and patient is needed.
5. Conclusions Patients with moderate-severe asthma found 6 months' use of reminders for missed doses and adherence feedback acceptable and useful for improving their adherence attitudes, adherence behaviours and conﬁdence in asthma self-management. Ongoing adherence support may be needed to maintain adherence for some. In addition to reminders, greater health professional-patient interaction and mutual understanding may be needed to appropriately motivate adherence in patients who hold beliefs which act as barriers to adherence.
Declaration of funding sources This study was funded by the National Health and Medical Research Council, Australia (ID 571053) and by an independent educational research top up grant from GlaxoSmithKline (GSK) Australia. These funders had no involvement in the study design, analysis, interpretation of data or the decision to submit this manuscript. The funders have not seen this manuscript prior to submission.
Declaration of interest JMF, TU, SS and LS have no ﬁnancial interest to declare with respect to this work. In the past 12 months HKR received fees from AstraZeneca (AZ), GSK, Merck & Novartis for data safety monitoring board participation; from AZ, GSK and Novartis for participation in advisory boards; and from AZ, Boehringer Ingelheim, GSK, Mundipharma and Teva for providing independent educational presentations, all unrelated to this work.
Acknowledgements The authors thank the patients who participated in the study; our research assistants: Pamela Kidd, Andreas Kruse, Hannah Openshaw, Louise Harkness, Kanika Jetmalani; Adherium (Auckland, New Zealand), from whom the monitors were purchased, and who provided some technical support during the RCT; and the funders of this study: National Health and Medical Research Council (which also funded the RCT), and GlaxoSmithKline Australia who provided an educational research top up grant.
J.M. Foster et al. / Respiratory Medicine 129 (2017) 39e45
Appendix A. Interview guide
Secondary questions and probes
Let's start by talking about your asthma in general. How has your asthma has been over the When was it and what happened? last 12 months? Let's talk now about the black [monitor] which was ﬁtted to your [preventer] during the asthma project What do you think of the [monitor]? What was good/not so good about it? What happened? Why? How would you change the [monitor]? Can you describe some advantages/ If reminders have not come up ask: How do you feel about the [monitor] disadvantages of the reminders? reminders? What happened? Why? Did you experiment with switching reminders off? What did you usually do when you heard a reminder? Can you give me an example of when What happened? In what way do you think the the reminders were useful or not reminders could be improved? useful? Do you remember looking at the website which showed a graph of how much [preventer] you used each day? What did you think of it? In what ways, if any, was the website helpful or unhelpful? Do you think the [monitor] reminders or the website changed how you view or take care of your asthma?
What were the positive/negative things about the website?
Can you think of any other ways your asthma or your opinion about asthma has changed in the last 6 months?
Can you give an example; What happened as a result of those changes? How have those changes affected you and people around you? Do you see any link between these changes and taking part in the MICA study?
On a scale of 1e5 how easy was it for you to use the [monitor] and [preventer] use website? 1 ¼ very difﬁcult 2 ¼ quite difﬁcult 3 ¼ neither difﬁcult nor easy 4 ¼ quite easy 5 ¼ very easy
Why did you choose that answer? Was one of these easier to use than the other? Can you give me some examples? What other experiences did you have when you were using these things?
Can you give me an example? What happened?
Why did you choose that answer? If you could keep the [monitor] and [preventer] use website, how likely would it be that you would use them in the future? 1 ¼ deﬁnitely not 2 ¼ unlikely 3 ¼ possibly 4 ¼ quite likely 5 ¼ very likely
 S.D. Fenerty, C. West, S.A. Davis, S.G. Kaplan, S.R. Feldman, The effect of reminder systems on patients' adherence to treatment, Patient Prefer. Adherence 6 (2012) 127e135.  T. Charles, D. Quinn, M. Weatherall, S. Aldington, R. Beasley, S. Holt, An audiovisual reminder function improves adherence with inhaled corticosteroid therapy in asthma, J. Allergy Clin. Immunol. 119 (2007) 811e816.  J.M. Foster, T. Usherwood, L. Smith, S.M. Sawyer, W. Xuan, C.S. Rand, H.K. Reddel, Inhaler reminders improve adherence with controller treatment in primary care patients with asthma, J. Allergy Clin. Immun. 134 (2014) 1260e1268.  A.H.Y. Chan, A.W. Stewart, J. Harrison, C.A. Camargo, P.N. Black, E.A. Mitchell, The effect of an electronic monitoring device with audiovisual reminder function on adherence to inhaled corticosteroids and school attendance in children with asthma: a randomised controlled trial, Lancet Resp. Med. 3 (2015) 210e219.  Institute of Medicine, Crossing the Quality Chasm: a New Health System for the 21st Century, 2017 [cited Jan 18, 2017]. Available from: https://www. nationalacademies.org/hmd/Reports/2001/Crossing-the-Quality-Chasm-ANew-Health-System-for-the-21st-Century.aspx.  A. De Vito Dabbs, B.A. Myers, K.R. Mc Curry, J. Dunbar-Jacob, R.P. Hawkins, A. Begey, M.A. Dew, User-centered design and interactive health technologies for patients, Comput. Inf. Nurs. 27 (2009) 175e183.  A.W. Wu, C.F. Snyder, I.C. Huang, R. Skolasky, H.F. McGruder, S.A. Celano, O.A. Selnes, A.S. Andrade, A randomized trial of the impact of a programmable medication reminder device on quality of life in patients with aids, AIDS Patient Care STDS 20 (2006) 773e781.  J.E. Haberer, G.K. Robbins, M. Ybarra, A. Monk, K. Ragland, S.D. Weiser, M.O. Johnson, D.R. Bangsberg, Real-time electronic adherence monitoring is feasible, comparable to unannounced pill counts, and acceptable, Aids Behav. 16 (2012) 375e382.  L. De Bleser, B. Vincke, F. Dobbels, M.B. Happ, B. Maes, J. Vanhaecke, S. De Geest, A new electronic monitoring device to measure medication adherence: usability of the helping hand (tm), Sensors-Basel 10 (2010) 1535e1552.  S. Howard, A. Lang, S. Sharples, D. Shaw, See I told you I was taking it! Attitudes of adolescents with asthma towards a device monitoring their inhaler use: implications for future design, Appl. Ergon. 58 (2017) 224e237.  J.M. Foster, L. Smith, T. Usherwood, S.M. Sawyer, C.S. Rand, H.K. Reddel, The reliability and patient acceptability of the smarttrack device: a new electronic monitor and reminder device for metered dose inhalers, J. Asthma 49 (2012) 657e662.  M. Schatz, C.A. Sorkness, J.T. Li, P. Marcus, J.J. Murray, R.A. Nathan, M. Kosinski, T.B. Pendergraft, P. Jhingran, Asthma control test: reliability, validity, and responsiveness in patients not previously followed by asthma specialists, J. Allergy Clin. Immunol. 117 (2006) 549e556.  J.M. Foster, L. Smith, T. Usherwood, S.M. Sawyer, H.K. Reddel, General practitioner-delivered adherence counseling in asthma: feasibility and usefulness of skills, training and support tools, J. Asthma 53 (2016) 311e320.  J.J. Francis, M. Johnston, C. Robertson, L. Glidewell, V. Entwistle, M.P. Eccles, J.M. Grimshaw, “What is an adequate sample size? Operationalising data saturation for theory-based interview studies, Psychol. Health 25 (2010) 1229e1245.  M.B. Miles, A.M. Huberman, J. Saldana, Qualitative Data Analysis: a Methods Sourcebook, third ed., SAGE Publications, Los Angeles, 2014.  J. Saldana, The Coding Manual for Qualitative Researchers, second ed., SAGE Publications, California, 2013.  Australian Bureau of Statistics, Socio-economic Indices for Areas (technical paper), Australian Bureau of Statistics, Canberra, 2006.  C.J. Armitage, M. Conner, Social cognition models and health behaviour: a structured review, Psychol. Health 15 (2000) 173e189.  V.S. Conn, M. Enriquez, T.M. Ruppar, K.C. Chan, Meta-analyses of theory use in medication adherence intervention research, Am. J. Health Behav. 40 (2016) 155e171.