Patient Acceptability of Home Monitoring for Neovascular Age-Related Macular Degeneration Reactivation: A Qualitative Study

Neovascular age-related macular degeneration (nAMD) is a chronic, progressive condition and the commonest cause of visual disability in older adults. This study formed part of a diagnostic test accuracy study to quantify the ability of three index home monitoring (HM) tests (one paper-based and two digital tests) to identify reactivation in nAMD. The aim of this qualitative research was to investigate patients’ or participants’ views about acceptability and explore adherence to weekly HM. Semi-structured interviews were held with 78/297 participants (26%), with close family members (n = 11) and with healthcare professionals involved in training participants in HM procedures (n = 9) (n = 98 in total). A directed thematic analytical approach was applied to the data using a deductive and inductive coding framework informed by theories of technology acceptance. Five themes emerged related to: 1. The role of HM; 2. Suitability of procedures and instruments; 3. Experience of HM; 4. Feasibility of HM in usual practice; and 5. Impediments to patient acceptability of HM. Various factors influenced acceptability including a patient’s understanding about the purpose of monitoring. While initial training and ongoing support were regarded as essential for overcoming unfamiliarity with use of digital technology, patients viewed HM as relatively straightforward and non-burdensome. There is a need for further research about how use of performance feedback, level of support and nature of tailoring might facilitate further the implementation of routinely conducted HM. Home monitoring was acceptable to patients and they recognised its potential to reduce clinic visits during non-active treatment phases. Findings have implications for implementation of digital HM in the care of older people with nAMD and other long-term conditions.


Introduction
Age-related macular degeneration (AMD) is a chronic, progressive condition and the commonest cause of vision loss in older adults [1]. Global prevalence of AMD is predicted to increase from 196 million in 2020, to 288 million in 2040 [2]. Neovascular AMD (nAMD) is a form of late AMD and is often associated with irreversible visual loss. It accounts for around

Materials and Methods
Qualitative methods were used to explore individual responses, views and experiences around HM acceptability, as well as to examine variations in contexts [28]. Semi-structured interviews were conducted face-to-face and via telephone. Participants were unknown to researchers conducting the interviews. The interview schedule (see Supplementary File S1) was based on the experience of the research team and on theories of technology acceptance. These included the Unified Theory of Acceptance and Use of Technology (UTAUT) [29][30][31], the Technology Acceptance Model (TAM) [32,33], the Theoretical Framework of Acceptability (TFA) [34,35] and the Senior Technology Acceptance Model (STAM) [36,37]. Members of the team that collected and analysed data (CT, SOC, MD) have extensive experience in the application of qualitative methods in healthcare research. The study followed the consolidated criteria for reporting qualitative research (COREQ) criteria [38]. Ethical approval was acquired from the National Research Ethics Service (IRAS ref: 232,253 REC ref: 17/NI/0235).

Participants
Recruitment to the MONARCH study began in July 2018 at five sites across the United Kingdom [24]. Apps were pre-installed on an iPod touch device given to participants who were asked to complete weekly HM for a minimum of 12 months, or until study completion in September 2021. Key characteristics of the three index tests for HM of nAMD reactivation are highlighted in Table 1. Recruitment to the qualitative component began three months after the MONARCH study. During the consent process for participation in the diagnostic accuracy study, individuals who consented to further contact to discuss participation in this qualitative study were approached via a telephone call from a qualitative researcher (CT, SOC). Informed consent was obtained prior to interviews and following explanation of procedures. Maximum variation sampling was used to ensure a range of perspectives were captured relating to age (young-old 50-69 years and older-old 70+ years), gender, laterality of nAMD (unilateral and bilateral) and time since first treatment (6-17 months, 18-29 months and 30-41 months). Usage data was assessed to classify participants based on adherence to HM as: 'Regular' (completed weekly HM without two or more gaps in testing of greater than three weeks), or 'Irregular' testers (stopped and started testing on more than two occasions, or stopped testing completely). Patients who declined to participate in MONARCH but consented to be contacted about the qualitative study were also approached. We approached informal 'carers', supporters or significant others in the lives of patients; and healthcare professionals who interacted with participants at study sites visits, to gather their perspectives about HM acceptability.

Data Collection and Analysis
All interviews were audio-recorded and transcribed. A directed content analysis approach based on deductive and inductive coding was used [39]. An initial coding scheme was developed (CT, MD) thatwas based on a synthesis of relevant theoretical construct of technology acceptability [29][30][31][32][33][34][35][36][37]. see Supplementary File S2. The coding framework underwent iterative development as individual transcripts were reviewed and re-reviewed during data familiarisation (CT, SOC, MD). Following line-by-line coding of each transcript (CT, SOC), findings were summarised based on the coding scheme and these summaries were used to revise initial codes if necessary and develop new codes based on emerging data (See Supplementary File S3 for final coding scheme). A third researcher (MD) coded a random sample of 10% of the transcripts and subsequently discussed and compared coding with CT and SOC in order to ensure adequate rigour and reflexivity. Respondent validation was also undertaken for approximately 10% of the interviews. Related codes were then clustered and grouped into initial themes. Narrative summaries were written (SOC) for each theme, then reviewed (CT, MD) and discussed to refine main themes and sub-themes to ensure coherence. NVivo version 12 was used to manage data and facilitate the analysis process, which in summary included the following stages: i. Independent transcription, ii. Data familiarisation, iii. Independent coding, iv. Development of an analytical framework, v. Indexing, vi. Charting and vii. Interpreting data.

Results
Thirty percent (89/297) of participants recruited to the MONARCH study agreed to be contacted about the qualitative component. When contacted-3/89 did not wish to take part, and 8/89 could not be reached. The remainder (26%; 78/297 of MONARCH participants) were interviewed (see Figure 1). This included participants categorised as "regular" (n = 63) or "irregular" testers (n = 14) and "non-testers" who declined to take part in HM (n = 1). Characteristics of patient participants (n = 78) were comparable to those not taking part in the qualitative study (see Table 2). In addition to the 78 patient participants, 11 informal 'carers', and 9 healthcare professionals were interviewed (6/11 informal 'carer' interviews took place in the presence of the patient participant). A total of 98 interviews were completed (patients, carers and health professionals). Interviews were conducted, in-person at patients' homes (n = 51), at clinical sites (n = 4), or via telephone (n = 45) between October 2018 and September 2020, lasting 36 min on average (range: 25 to 78 min). derwent iterative development as individual transcripts were reviewed and re-reviewed during data familiarisation (CT, SOC, MD). Following line-by-line coding of each transcript (CT, SOC), findings were summarised based on the coding scheme and these summaries were used to revise initial codes if necessary and develop new codes based on emerging data (See Supplementary File S3 for final coding scheme). A third researcher (MD) coded a random sample of 10% of the transcripts and subsequently discussed and compared coding with CT and SOC in order to ensure adequate rigour and reflexivity. Respondent validation was also undertaken for approximately 10% of the interviews. Related codes were then clustered and grouped into initial themes. Narrative summaries were written (SOC) for each theme, then reviewed (CT, MD) and discussed to refine main themes and sub-themes to ensure coherence. NVivo version 12 was used to manage data and facilitate the analysis process, which in summary included the following stages: i. Independent transcription, ii. Data familiarisation, iii. Independent coding, iv. Development of an analytical framework, v. Indexing, vi, Charting and vii. Interpreting data.

Results
Thirty percent (89/297) of participants recruited to the MONARCH study agreed to be contacted about the qualitative component. When contacted-3/89 did not wish to take part, and 8/89 could not be reached. The remainder (26%; 78/297 of MONARCH participants) were interviewed (see Figure 1). This included participants categorised as "regular" (n = 63) or "irregular" testers (n = 14) and "non-testers" who declined to take part in HM (n = 1). Characteristics of patient participants (n = 78) were comparable to those not taking part in the qualitative study (see Table 2). In addition to the 78 patient participants, 11 informal 'carers', and 9 healthcare professionals were interviewed (6/11 informal 'carer' interviews took place in the presence of the patient participant). A total of 98 interviews were completed (patients, carers and health professionals). Interviews were conducted, in-person at patients' homes (n = 51), at clinical sites (n = 4), or via telephone (n = 45) between October 2018 and September 2020, lasting 36 min on average (range: 25 to 78 min). . RT = regular testers (completed weekly testing without significant gaps in testing during the study period); IT = irregular testers (completed one, two or three tests, or stopped-and-started testing, or withdrew from diagnostic accuracy study); NT = non-testers (declined to take part in home monitoring); * Declined . RT = regular testers (completed weekly testing without significant gaps in testing during the study period); IT = irregular testers (completed one, two or three tests, or stopped-and-started testing, or withdrew from diagnostic accuracy study); NT = non-testers (declined to take part in home monitoring); * Declined to participate (n = 3), not contactable by telephone (n = 8); ** Not contactable by phone (n = 3); *** Moved post (n = 1), not contactable (n = 5).
Views about HM acceptability appeared to be represented by five overarching themes (and nine associated sub-themes): 1. The role of HM; 2. Suitability of procedures and instruments; 3. Experience of HM, and 4. Feasibility of HM in usual practice; 5. Impediments to home monitoring. Relationships between themes and coding of data are shown in Supplementary File S4. Each theme is presented in the section below. Illustrative quotes are provided in Table 3. Views of informal 'carers' and healthcare professionals are summarised in Supplementary Files S5 and S6. * Calculations are based on n = 77 as overall qualitative sample includes n = 1 participant who declined to take part in home monitoring but consented to take part in the qualitative part of the study. ** For patients with two involved eyes, better seeing eye is used.    The role of HM was clear from the perspective of participants who viewed it as facilitating 'self-measurement' of visual acuity. Monitoring was viewed as providing 'ownership' or 'control' over visual health, in that patients 'knew their own eyes better than anyone else'. Participants described how they 'self-monitored' vision before the study, noting ' . . . changes when looking out at a street sign visible from my window', or checking for difficulties when reading or watching television. This practice is referred to clinically as an 'environmental Amsler test'.

Sub-Theme 2: Perceived Impact on Eye Care
Home monitoring was seen as supporting usual care by providing further assessments of possible visual deterioration. Participants did not regard HM as a 'replacement', but as part of their care, particularly during non-active treatment periods. It was recognised as potentially reducing the frequency of clinic visits. This was an acceptable and positive outcome. It was highlighted this could save time and effort without inconveniencing family and friends often required to transport patients to clinics. Potential for HM to reduce health service costs and relieve burden on services was noted. It was recognised this might allow appointments to be targeted based on clinical need, specifically, when risk of deterioration is higher. Concerns were expressed that reliance on HM might delay treatment depending on the degree to which results were reviewed by clinic staff.

Theme 2. Suitability of Procedures and Instruments
The iPod touch and index tests were viewed as novel innovations but also as reflecting the increasing pervasiveness of technology. Participants' pre-HM use of similar devices varied but even those with minimal exposure viewed HM as realistic and suitable. Overcoming unfamiliarity or hesitancy with technology was regarded as 'something that needed to be done' as part of their care. Participants suggested demonstrations of the tests, prior to, and separate from the formal 'training' in HM procedures provided reassurance that it was 'simple enough to do'. This increased belief in their capacity to undertake HM. In addition to technological apprehension, participants reflected that being older and unfamiliar with technology might limit engagement, with concerns that HM was too complex or 'burdensome'.

Sub-Theme 1: Training for Home Monitoring
Participants described training as essential for 'getting going' with HM but as 'information heavy'. 'Refresher' sessions were suggested to help overcome difficulties recalling training information. Experiential learning was important as participants reported HM became easier and routine with practice. Participants described as valuable 'informal' support and advice that health professionals provided during clinic-based study visits.

Sub-Theme 2: Test Preferences
Participants viewed the iPod as easy to use but suggested a larger screen could increase usability. Tests were referred to, positively, as 'feeling like a game' that reduced 'boredom' of repeated testing. A preference was expressed for MultiBit (MBT) relative to MyVisionTrack ® (mVT). This was linked to 'feedback' (represented as a percentage score) (discussed under sub-theme 3 below). Other factors influenced participants' views. For MBT, there was lack of clarity about the purpose of the later stages of the test, described as 'disheartening' as they became progressively 'too fast'. The mVT test was viewed as too subtle (when discriminating between the 'distorted' circles), and as being difficult to complete, leading to participants 'guessing' responses. In addition, the mVT test and the paper-based keepSightJournal (KSJ) were perceived as less engaging compared to the MBT. Overall, MBT test scores were viewed as providing valuable feedback to 'self-monitor' changes in vision. This was apparent despite participants not being encouraged to use the scores, or being provided with information on their purpose. Participants described noting results, making comparisons over time and using feedback to 'beat my last score'. This sense of 'self-competition' was seen as helpful for maintaining engagement. However, perceived ambiguity about meaning of scores produced uncertainty over how to respond to changes, and doubts about how results were used, and regarding their role as a patient. 'Retesting' was described as a useful way of confirming changes in scores. Negative changes were attributed sometimes to tiredness or becoming distracted during testing. Consistently lower results, even when small in magnitude, were interpreted as a concern about their eye health. Participants highlighted how their views about HM improved as they became more familiar with testing. Weekly HM was seen as realistic and feasible, taking between 10-15 min. Establishing a HM 'routine' was considered important. Participants used several methods to help continue regular HM, including written reminders or prompts. It was described how family members would visit at set times each week and this functioned as a reminder when to test, and a means of getting help with setting up devices. Participants referred to using other forms of digital 'self-monitoring', including blood pressure and respiratory rate measurements; and said this made it easier to set up HM routines, acting as a reminder to complete vision tests. Participants acknowledged that HM would need to be continued in the longer-term and would need to be 'easy, and not a burden' to achieve sustainability and high adherence.

Sub-Theme 2: Use of Ongoing Support
Support provided by the study telephone helpline was perceived as key. Occasionally, advice were provided by health professionals during clinic visits and this was deemed valuable.
3.5. Theme 5. Impediments to Home Monitoring 3.5.1. Sub-Theme 1: Practical Issues Practical or technical issues with HM were relatively infrequent. These tended to be minor, including devices needing to be recharged before testing (see Supplementary File S7).
Occasionally, issues stopped participants from testing and returning to HM-restarting even after a brief period of non-testing could be challenging. Participants reported, in general, that issues could be overcome via 'problem solving'. Regarding the MBT, participants addressed the requirement for dark conditions during testing by changing their physical environment, such as placing a blanket over their head or testing in windowless spaces. Sometimes, adaptions were described as making the test challenging and 'awkward'. A more substantial issue was that participants were sometimes unsure if tests had been completed successfully and data had been transmitted. It was suggested that an improvement might be to provide instantaneous confirmation of both aspects.

Sub-Theme 2: Personal Health and Social Factors
Other issues that negatively affected HM were related to health concerns or functional limitations such as fine motor tremors, fatigue and concentration problems. Participants with caregiving responsibilities, such as providing childcare, or caring for a partner or spouse, reported difficulties finding time to test consistently.

Views of Informal 'Carers' and Healthcare Professionals-Summary
There was a high level of concurrence between perspectives of informal 'carers' and healthcare professionals and views of patients. Healthcare professionals suggested that increasing age might be linked to reluctance or hesitancy about HM, acknowledging that this view may be related to assumptions about older people and use of technology. Healthcare professionals agreed with patients' views that training was essential to uptake and use of HM, and that training needed to be adapted to individual patients and their technology experience. There was a concern that HM could be resource intensive in terms of technical support needed (See Supplementary File S6). Informal 'carers' or family members saw their role in HM as supportive and facilitative; and valued feedback from test performance to assess response to treatment received by their family member (See Supplementary File S5).

Discussion
This qualitative study investigates views of patients, informal 'carers' and healthcare professionals about acceptability of home monitoring for nAMD reactivation. Home monitoring was acceptable to participants and key factors such as patient's understanding of HM, and how it could be integrated into usual care appeared to influence these views. According to relatively younger patients, older peers might find HM to be a challengea perception also reflected in views of healthcare professionals. However, the factors that appeared to have a greater impact on positive views about HM acceptability were participants' perceptions around the usefulness of HM for eye care, how easy it was to complete weekly HM, and their experience of undertaking HM.
Inexperience with using technology did not seem to limit or affect HM, or a participant's intention to use it, and experience relating specifically to other forms of digital monitoring of health symptoms (e.g., blood pressure monitoring or medication reminder apps) may have been a facilitating factor. Establishing the 'habit' of HM and integrating it into a participant's routine seemed to be important in terms of ensuring regular use. Weekly HM was feasible though more frequent monitoring (e.g., daily testing) may be too burdensome and, therefore, less acceptable. In general, the HM tests were reported to be easy to undertake and non-burdensome. The time commitment required to undertake HM was also acceptable and, although technical difficulties were relatively infrequent, access to ongoing support was regarded as essential to successful HM, and for overcoming any unfamiliarity with use of technology. Support included 'formal' training and assistance from healthcare professionals with technical aspects of HM, and 'informal' support primarily from partners and family members in the form of encouragement and facilitation of HM, and to help manage in situ any technical issues.
It was recognised that there was potential for HM to reduce the frequency of clinic visits, particularly during non-active treatment phases. This positively viewed aspect was contingent upon the expectation and availability of an easily accessible route back to usual service arrangements if further treatment was required. The use of test performance feedback was perceived by participants as a way to 'self-monitor' vision, even though 'feedback' was provided by only one of the tests (the MBT). This was despite information having been given at the outset to participants that the HM test information would not influence their care, and that they should not contact the hospital on the basis of test results, but should do so if they noticed a deterioration in their vision.
A slight preference for the MBT test was evident, mainly due to its perceived 'feedback' feature, and this was reported despite the test needing to be completed in darkness, which presented an inconvenience and a practical impediment to testing. In addition to the lack of performance feedback, the mVT was also perceived as being 'too difficult' to complete effectively. This may have been due to the staircase procedure used to determine threshold, hence there would have been trials in which the patients struggled to see an odd one out as it honed in on the final threshold. The mVT, along with the paper-based KeepSight Journal, were also reported as less 'engaging' than the MBT. Performance feedback alongside motivational or supportive messages can be key components of digital interventions; and they tend to be associated with higher intervention usage and acceptance [40]. Various forms of 'meaningful' feedback about a patient's health status can provide a sense of support, guidance and information, as well as giving reassurance to patients [41]. However, feedback by itself may also be an intrinsic motivator to sustained HM beyond any 'clinically meaningful' relevance of change in performance scores [42][43][44].
While test complexity and technical or reliability problems are common barriers to technology acceptance [45][46][47], these issues were relatively infrequent in the present study. Persistence and 'problem solving', as well as external support were needed to overcome them in the relatively few occasions when they did occur. Uncertainty about whether or not the digital index tests had been completed successfully and if test data had been transmitted to the study team appeared to influence views about acceptabilityParticipants suggested strongly that confirmation of successful test completion and data transmission should be provided to HM patients. As one of the participants noted, ' . . . it (HM) isn't worth it if you don't know if the test is working properly or if no one knows what is happening on the other end'. In summary, HM was acceptable to participants. Only five of 78 participants in the qualitative sample (approximately 6%) discontinued HM. Facilitators to regular HM appeared to revolve around concerns about eye health, perceptions that HM provided a sense of control through test feedback, perceived ease of test completion and the ability to form a routine or HM habit.

Comparisons with Previous Literature
Our finding that HM was acceptable to patients and could be used to assess the stability of nAMD and monitor associated changes in vision is in keeping with other related research about the monitoring of visual function using digital testing [48][49][50][51]. For example, a pilot study of a tablet-based programme for detecting progression in patients with intermediate AMD indicated good agreement between HM and clinic-based testing [49]. A recent, prospective cohort study concluded that HM, using the Alleye App, was a useful adjunct to usual eye care, with high specificity and predictive value for identification of disease progression (93.8% (95% CI: 86.2-98.0%) and 80.0% (95% CI: 59.3-93.2%), respectively [50].
Despite information security having been highlighted as an issue with the use of home-based digital technology in ophthalmology care [52], participants in the current study did not express this as a concern. An important observation was that MBT scores were 'self-monitored' by patients and to an extent, informal 'carers' or relatives. These participants described noticing a 'pattern' in scores which they perceived as indicating a decline during the time leading up to clinic appointments to receive anti-VEGF injections and which improved in the weeks following injection. Similar observations were also reported in an earlier study where patients with nAMD using the MBT undertook more frequent testing than was suggested, and independently recorded their test scores [53].
Adherence with HM in some studies appears to be associated different factors including age and treatment status. For example, ophthalmology service patients including patients with nAMD who took up the offer of self-monitoring using the Alleye App [54] tended to be significantly younger, male, living with family and not currently in active treatment. However, within this group, actual compliance with HM was higher in those who were older, and in active treatment. It should be taken into account when making comparisons with previous literature that follow-up periods in these papers have tended to be relatively short (less than six months), meaning that the long-term impact of factors influencing HM are not certain.

Strengths and Limitations
Strengths of the study include the use of maximum variation sampling to recruit participants based on a range of important characteristics including gender and time since initiation of treatment. Recruitment took place across different clinical sites and geographical locations, and was successful in enrolling participants who were reflective of the overall sample from the diagnostic accuracy study. Participants' views about acceptability were similar across these locations. The study incorporated the views of informal and professional carers and, generally, there was a high level of concurrence across these perspectives. A potential limitation of the study is that a mixture of telephone and face-to-face interviews were used. This was to allow flexibility for participants, but was also a result of the COVID-19 pandemic and the required social distancing measures that precluded home visits. Despite this, there did not appear to be any substantial variances between interviews conducted using these different methods. Although an attempt was made to recruit more participants who stopped HM, or who completed it irregularly, the majority of the sample included participants who showed good levels of adherence to weekly HM. This might suggest the sample could be described as having characteristics of 'early adopters' of technology, and this may have influenced the results.

Conclusions
This qualitative study provides important insights into the perspectives of patients, 'informal' carers and healthcare professionals about the acceptability of HM for assessing reactivation in nAMD. Home monitoring was acceptable and non-burdensome but initial training and ongoing support are essential to successful implementation. Reasons for stopping testing related to unresolved technical issues or changes in personal circumstances such as a partner becoming unwell. Patients recognised the potential for HM to reduce clinic visits, particularly during non-active treatment phases and the related cost savings. Feedback scores from index tests (MBT) were valued by participants as a way to 'selfmonitor' visual acuity. However, there was ambiguity about use of performance scores to assess any changes in vision, and uncertainty about what scores meant in terms of visual health. These findings have important implications for the design and use of digital HM in the care of older people with nAMD as well as in other long-term health conditions.   Data Availability Statement: Study data are available from the corresponding authors upon request.