Home monitoring participants

Baseline case report forms collected demographic information (age, gender, ethnicity, employment status), current use of technologies (smartphones, tablets/iPads, laptops), preference for text or e-mail reminders, clinical information about their glaucoma (e.g. glaucoma type and severity, treatment, surgeries), and baseline IOP and VF measurements. Data were collected by a trained member of site staff through participant report (demographics and technology use) and clinical information recorded in medical records. IOP and VF tests were only performed if they had not been conducted in clinic (with results available in medical records) within the last 3 months.

Site staff collected the follow-up case report form data on IOP and VF measurements conducted at follow-up and information on any changes in treatment. Data were also collected by site staff through participant report relating to satisfaction with the home monitoring intervention, satisfaction with training received, and overall preference between home and clinic monitoring. Participants were also asked to complete questions covering health resource use during the home monitoring period, which asked about their actual use of health services during that time, and if they considered contacting/using health services, to measure whether home monitoring led to any change in use of health services. Again, data were collected verbally from the participant by a trained member of site staff. IOP and VF tests were performed by a trained member of site staff at the follow-up visit.

Data were downloaded from the tonometer device via USB cable to dedicated software (iCare LINK) when the device was returned at the follow-up assessment. The data were transferred as an Excel sheet (.csv file) and added to the secure study database. The OKKO app data were stored on the OKKO Health portal and provided to the research team on request.

Site staff were asked to record any contact they had from the participant throughout the home monitoring period. Data collected included: date, method of communication, a brief description of the reason for the contact, and the approximate time the site staff member spent to resolve this issue.

Interviews

Data collection for the interviews with patient participants recruited to the home monitoring study was informed by a topic guide with questions framed around the constructs of the TDF. The TDF is an established behavioural framework that integrates 33 theories of behaviour into 14 domains that inhibit or enable behaviour (Knowledge, Skills, Social/Professional Role and Identity, Beliefs about Capabilities, Beliefs about Consequences, Optimism, Reinforcement, Intentions, Goals, Memory/Attention/Decision-making Processes, Environmental Context and Resources, Social Influences, Emotion, Behavioural Regulation).⁵⁰ For the purposes of I-TRAC, the target behaviour of interest was broad in its definition and was considered as use of the digital technologies for monitoring their glaucoma at home. Interviews were conducted by the Research Fellow over the phone.

Site staff focus groups and interviews were informed by a topic guide framed around key questions from the ADePT framework to assess pragmatics of running the study and the feasibility of progressing this study to a full-size trial.⁵¹ Focus groups and interviews were conducted via MS Teams.

All focus groups and interviews were recorded for verbatim transcription by an external transcription company.

Home monitoring

Frequency of patient participant contacts throughout the study period (for assessing resource impact) was collected via staff self-report and presented using frequencies. Adherence to intervention was calculated for each participant and for each device. This was calculated based on the number of weekly measurements they performed over the 12-week monitoring period. The start of the monitoring period was classified as the date the participant was due to perform their first measure plus each week for the subsequent 11 weeks (12 measures in total). A reported measurement was considered adherent to schedule if it was conducted ± 3 days of the scheduled measure. Overall adherence for each device was calculated as a percentage based on the number of measurements adherent to schedule, divided by 12 (total number of possible measures). Those scoring ≥ 80% were considered ‘adherers’ to the device. Combined adherence scores were categorised as ‘adherent’ based on ≥ 80% adherence to both devices individually with all other variations (i.e. adherence to only one device or to neither) classified as non-adherent or combined adherence.

Participant contact records were collated by the Research Fellow and the total number of contacts, mode of contact, reason for contact, and total time spent by site staff resolving any questions, for each participant, summarised.

Interviews

Interviews with patient participants were analysed using the TDF. A coding guide was developed to describe relevant data to be coded under each domain (see Appendix 4). Each transcript was reviewed for data relevant to any of the TDF codes but also concurrently for any utterances relevant to patient experience of glaucoma home monitoring but that did not fit TDF domains. Three members of the research team (KG, TC, CS) reviewed and double coded transcripts (KG n = 2, TC and CS n = 3). Coders met to review coding decisions with any disagreements resolved through discussion with an independent TDF expert for arbitrating unresolved disagreements. The coding guide was updated to reflect these discussions and then applied to the remaining transcripts.

Site staff focus groups and interviews were analysed using the items from the ADePT framework to help directly inform decisions about progression to full trial. A coding framework for the ADePT items was developed to describe the relevant data to be coded under each ADePT item (see Appendix 5). A second member of the research team checked coding of two of the four transcripts and disagreements were resolved through discussion between the two coders (CS and KG). Data under each ADePT item were reviewed to identify themes, subthemes and relationships between these. Analysis was further refined during writing with both coders reviewing and agreeing the final thematic framework.

Any changes from inception to the design of this phase of the research (and linked phases) are reported in Appendix 6.

Sample characteristics

Across the 3 sites, 42 patient participants were recruited. Differences in recruitment approaches were apparent in the screening logs received from each site. In sites 1 and 2, one strategy used was to prescreen clinic lists in advance to identify potentially suitable participants for the clinician to discuss with the patient at their clinic appointment. Another strategy involved the Research Nurses/other suitably qualified persons approaching patients in the clinic waiting area. Due to delays in opening and the requirement for time-efficient recruitment, site 3 purposively selected glaucoma patients who were known to the research team as being research active (e.g. having participated in one or more previous research studies). With regard to participation rates, for site 1, all patients approached by the PI at their clinic appointment agreed to participate in the study. For site 3, all patients approached by Research Optometrists agreed to participate. For site 2, where the Research Assistant approached patients waiting in clinic waiting rooms, the participation rate was 16% (18/116). While most did not provide a reason for declining participation, of those who did, 13 patients declined due to a reported lack of confidence with technologies, 10 due to a lack of interest in technologies, 5 due to the travel requirements to attend additional clinic visits and 4 due to concerns about burden related to multiple comorbidities. One person declined due to disagreement with the concept of home monitoring and another declined due to glaucoma testing being anxiety provoking.

Figure 2 provides an overview of participant recruitment and retention.

FIGURE 2

In-home Tracking of glaucoma: Reliability, Acceptability, and Cost participant flow chart.

The patient participant-level characteristics shown in Table 8 indicate a mean age of 67 years in our cohort; the majority were white, and there was equal gender representation. The majority were familiar with some form of electronic device.

TABLE 8

Baseline demographics of patient participants conducting home monitoring

Baseline clinical characteristics

Median baseline IOP using GAT was 17 mmHg (IQR 12–20) and 16 mmHg (IQR 12–20) in right and left eyes respectively. Among the 42 patient participants, 26 (62%) had glaucoma in the right eye and 30 (71%) had glaucoma in the left eye. Other diagnoses are summarised by eye in Table 9.

TABLE 9

Baseline eye measurements, glaucoma treatment, past history and comorbidity

Median baseline VF mean deviation (MD) was −1.8 dB and −3.8 dB in right and left eyes. Median MD in the worse eye (by baseline MD) was −4.65 dB (IQR −10.2 to −1.5, range −30.5 to 0.75). Median MD in the better eye was −1.3 dB (IQR −4.2 to −0.1, range −30.5 to 1.2). The patient participant with a MD of −30.5 dB had VF data entered for only one eye. No other patient participants had VF data entered for only one eye. For almost all patient participants the VF test was deemed reliable by the clinicians (for 40 of 41 tests and 41 of 42 tests in right and left eyes, respectively).

Baseline topical IOP-lowering medications are summarised in Table 9. Out of 42 patient participants, 11 (26.2%) were not using regular eye drops for glaucoma; 9 patient participants had previously undergone laser trabeculoplasty in their right eye and 10 in their left eye; 16 patient participants had had previous surgery to their right eye and 20 previous surgery to their left eye. The majority of patient participants had a diagnosis of glaucoma (26 right eyes and 30 left eyes), but patient participants with suspected glaucoma, OHT and primary angle closure or primary angle suspect were also included in the study (Table 9). A total of 34 patient participants had glaucoma in at least 1 eye.

The majority of patients had mild to moderate glaucoma severity. The VF MD values also allow severity of glaucomatous VF loss to be determined. In all, 23 of 42 patient participants had mild to moderate glaucoma in their right eye and 24 mild to moderate glaucoma in the left eye. Ocular comorbidities including cataract and age-related macular degeneration were present in eight right eyes and eight left eyes.

Table 10 shows the clinical data obtained at the 3-month follow-up visit. Median Goldmann IOP was 17.0 mmHg in the right eye (IQR 13.5–19.0) and 17.0 mmHg in the left eye (IQR 13.0–19.5). VF MD was −2.0 dB (IQR −4.8 to −0.6dB) and −3.2 dB (IQR −7.3 to −1.2) in right and left eyes, respectively. Only 2 of 42 patient participants had a change in glaucoma treatment instigated by their clinician while participating in the study.

TABLE 10

Three-month follow-up measurements and change in treatment

Three patient participants withdrew from the study during the monitoring period, two from site 2 and one from site 1. One was due to health reasons (unrelated to glaucoma) and two were due to difficulties with the technology, with one withdrawing from the home monitoring intervention only and completing follow-up data collection and the post-intervention interview.

Adherence to intervention

Adherence to the tonometer alone was 67% (n = 28) compared to 60% (n = 25) of participants being adherent to the OKKO app (Table 11). Overall adherence to both devices (i.e. ≥ 80% adherence to both devices) was 55% (n = 23), with 31% (n = 13) considered as non-adherent to both devices and 14% adherent to only one device.

TABLE 11

Participant adherence with glaucoma home monitoring technologies

Additional patient participant contact with site

Of the 42 participants, 20 (48%) contacted their respective study site during their home monitoring period, averaging 1.7 contacts per participant who required support, and were spread across sites (Table 12). In total, 503 minutes (average of 15 minutes per contact) were estimated by site staff as being required to answer or resolve issues. The most common reasons for contact were problems with usernames and passwords for the OKKO app (n = 5), and patient participants having difficulties using both devices (n = 5) (Table 13).

TABLE 12

Unscheduled patient participant contacts during the home monitoring period

TABLE 13

Main reasons for patient participant contact during the home monitoring period (n = 20)

When asked about satisfaction with the home monitoring, the most frequent response (40%, n = 17) was that patient participants were ‘very satisfied’, and similarly, a majority (45%, n = 19) reported being ‘very satisfied’ with the training they received on the digital technologies (Table 14). When asked to report their preference for future monitoring, 50% of patient participants reported that they would prefer to be monitored at home.

TABLE 14

Satisfaction with monitoring and training and future monitoring preference

Decision to participate (goals; intentions; knowledge; social influences; social professional role and identity)

Patient participants reported that entry into the I-TRAC study was often precipitated by a discussion with their glaucoma specialist to participate in a research study that would explore the potential of home monitoring as an alternative to traditional monitoring. Participants did not report basing their decision to participate on the input of others, such as friends and family. Instead, they were often motivated by their own history of glaucoma to join the study or were interested in engaging with a new technology or research project, generally. Participants typically understood the rationale behind the trial (i.e. to test the feasibility of home monitoring as an alternative/supplement to in-clinic monitoring). However, some participants struggled with the concept that the devices themselves, particularly the tonometer, were also being refined through the I-TRAC study before they could be implemented into standard clinical practice.

R – In terms of your confidence, it sounds like you felt quite confident at that first contact about the study that this would be something that you would be able to do.

P – Yes, but I probably didn’t quite understand what you’re trying to achieve. I thought what you were trying to do was sort of find out… not quite sure. I didn’t realise that it was actually sort of testing the equipment. I thought it was more the idea of regular pressure checks, that sort of emphasis. I suppose I got the emphasis wrong.

Pt 12007

Use of the devices (behavioural regulation; environmental context and resources; skills)

When considering participants’ experiences using the devices, several patient participants discussed their impressions of the intervention training. Most patient participants reported that training for the use of the device was generally well received and believed to instil the skills necessary for participants to monitor on their own. However, delivery of training was not consistent across the sample, with some participants stating that the session was ‘rushed’ and did not cover all aspects of monitoring to their satisfaction. At times, this appeared to be attributable to time and space pressures within the I-TRAC sites, which either did not allow sufficient time or a suitable environment to conduct training. Some participants also voiced concerns about the difficulties staff conducting the training were having aligning the tonometer to take a measurement. Lastly, some participants felt that the interval between the training session and active monitoring was too long, leading to them forgetting many of the skills they had acquired.

Well, I just had met up with this one nurse and she was struggling to find a room where we could do it, so … but then found one, and we had a limited amount of time, but it probably wasn’t ideal for her. […] I think it’s somehow sort of a … I think in order to go from a situation where you’re shown something to you’re doing it, you need a sort of continuity, and I don’t think that was there in the session, and because of the holiday break, I wasn’t following it through so there wasn’t a sort of trans- … a smooth transition from trying it to using it.

Pt 12007

Regardless of training, many participants felt that the devices were accessible enough to engage with through step-by-step, trial-and-error perseverance, assisted by the manual and demonstration videos, until they were reasonably confident they were conducting the monitoring as intended. Engaging with the devices through practice also allowed participants to individualise their process, such as performing it in certain rooms or with additional materials (e.g. stacks of books), so that they could reliably achieve readings in less time and/or fewer attempts. Remembering to monitor was not described as difficult, with participants successfully incorporating the process into their established routines, most often in tandem with taking medications or self-care activities.

R – How easy or difficult was it to remember to do your weekly monitoring each week?

P – I found it pretty easy but I’ve got a reasonably decent memory. And I also put it near the eye drops so when I was doing my eye drops, I would see it and think, ‘Is it Tuesday? No, it’s not.’ But it would be there to be seen, it was somewhere it could be seen and not just away in a cupboard out of sight.

Pt 13005

Potential challenges to home monitoring (environmental context and resources; memory, attention, and decision processes; reinforcement; skills)

Challenges to using the devices were predominantly focused on the use of the tonometer. Difficulties opening and mounting probes, as well as positioning the tonometer in front of the eye, were described often and appeared to be a persistent frustration for some, particularly those with dexterity issues. Feedback from the tonometer on accurate positioning and the pressure readings themselves was deemed insufficient or believed to be absent. Issues performing the visual function tests on the iPad were comparatively fewer, as use of a tablet was more familiar to most participants. However, a frequent criticism of the VF testing was that it was ‘boring’ or ‘tedious’ and required efforts to maintain concentration on the task. Overall, however, use of the devices appeared to be a surmountable challenge for most participants.

Then the tonometer. I normally looked at the instructions to make sure I was doing the thing correctly. I found it very difficult to have the piece of equipment correctly on the eye because it was so sensitive to every little movement, but I did eventually manage to get the green circle of light, but then often when I went to press the button to record the measurement, the piece of equipment moved, so you lost the [measurement]. That was one of the little difficulties. Then it went on. I can think of all sorts of things that didn’t happen that shouldn’t have done.

Pt 13001

Perceived advantages and disadvantages of home monitoring (beliefs about consequences; emotion; environmental context and resources; knowledge; social professional role and identity)

Participants could readily volunteer both potential advantages and disadvantages to home monitoring. Speaking first of advantages, there were perceived benefits to the individual as well as to the larger healthcare system. The most frequently mentioned benefit of home monitoring was its promise to relieve pressures on clinics and eye-care providers for providing all aspects of routine monitoring. Participants could appreciate that taking on some aspects of that monitoring could free up appointments for those with more acute health concerns or otherwise allow providers to reallocate their time to improve standards of care. Having control over how often one could monitor was also seen as advantageous, particularly when waiting times exceeded several months in clinic settings. Participants believed that more frequent monitoring could be beneficial to catch any early indicators of advancing disease and seek treatment before their condition worsened. Finally, the logistics of travelling to in-person appointments were described as a practical motivator to engage in home monitoring to reduce the number of clinic visits necessary.

It’s certainly much more convenient to sit here than have to trundle up to the eye clinic. I know that’s necessary because they need to see you as well, but maybe not just as often. I think it’s a good idea and having undertaken it, I would recommend it.

Pt 14002

Conversely, several of the perceived disadvantages were related to deviating from the standard treatment pathway. Firstly, several participants were concerned about the overall accuracy of the home monitoring device and how it compared to monitoring within professional settings. These concerns appeared to be predicated either on the accuracy of the home monitoring devices themselves or in the participant’s belief in their capability to use the equipment as intended. In either case, the potential for inaccurate results drew concerns that an individual’s eye health could be deteriorating without their, or their provider’s, knowledge. Even when concerns over accuracy were not as relevant, delays between recording high pressures and being able to access care were seen as a source of anxiety. Even nominal readings were felt to be of little value if they were not contextualised by the knowledge of the healthcare provider, who could provide reassurance that values were trending normally or that a plan of care needed to be enacted.

I think for me at first, it was just because I’d be scared, thinking it’s really high. Say it was like eight o’clock at night when you measured and for me, it would be something like 37 and I would have to massage my eye. I would then think, ‘Now I’ve got to wait until nine in the morning to contact them if it goes up.’ It just affects you psychologically that way. Sometimes I’d think I’d rather not know. I mean, you could phone a doctor at the hospital, but you know it’s not an acute emergency so you’re actually then very anxious. That was probably the bit that scared me the most at the beginning.

Pt 12011

Recommendations to improve monitoring (behavioural regulation; environmental context and resources; reinforcement; social influences)

Several aspects of the devices and the home monitoring process were identified from this sample. The coordination required to align the tonometer for readings was cited as a barrier for those with dexterity or other hand–eye coordination issues. Participants believed that this could be streamlined in some way to promote easier alignment of the tonometer with the eye that was not as susceptible to fluctuations in hand steadiness. Some participants also mentioned difficulties in accessing their pressure readings from the device, possibly due to a display time-out. It was suggested that readings should be available in an easier-to-read format that preserves the history of recordings, as well.

Potential improvements to training were mentioned by several participants. As mentioned above, some participants felt there was too long an interval between the training session and the first ‘live’ monitoring session. They suggested that training be held as near as possible to this first session to increase the likelihood that skills practised during training will transfer to real-world use. ‘Top-up sessions’ (i.e. condensed training focused on reiterating or improving skills already learned) were seen as a useful addition to the course of home monitoring if participants were finding it difficult to apply their skills from one session. Additionally, some participants suggested that ongoing technical support would have been helpful, but their attempts to solicit that help within this feasibility trial were not addressed. As for any other additional content of the training, some participants were unaware of the demonstration videos and suggested those be emphasised as an available resource in the course of training. There were also mentions that the training manual provided, while helpful for most, should be reviewed for readability, as some participants complained of not being able to utilise it.

R – So, one suggestion that’s come in from a couple of participants is that it might be useful to have a lead-in period, so you get maybe a week or two just to practise at your leisure with the equipment and then commence the study. What would you think of that suggestion?

P – I think that’s a very good idea because when you’re only doing it weekly, from one week to the next, you can forget how you managed to do it correctly. So, if you had a week where you could just use it anytime, say every day, just to get into the swing of it, I think that would be a good idea.

Pt 13009

Finally, some improvements to the process of monitoring were suggested. In order to help build confidence in their ability to conduct monitoring, participants suggested the inclusion of a ‘lead-in’ period. This period would be an opportunity for participants to practise with the device without fear of ‘getting it wrong’ and affecting trial results. Once in active monitoring, as mentioned above, participants would have also liked easier-to-read and easier-to-understand feedback from the devices so they could track the progress of their readings. Further resources on how to interpret results may also prove valuable in this sense. A final comment on the active monitoring period was the anxiety that may be precipitated by too-frequent measurements. To pre-empt this, participants suggested that monitoring should be restricted to only what is necessary, for example once a week.

I think it would be valuable if you could do it monthly or something like that, just once, no more than that if it was stable. Then that way it could be fed into some computer to see the trends or something. But any more, I’m just not sure, just like my blood pressure. Psychologically, you just might get neurotic.

Pt 12011

Site staff knowledge, beliefs and feelings towards home monitoring

All site staff reported positively when asked to reflect upon how they felt when first approached to help deliver the I-TRAC study. All staff were optimistic about a novel approach for glaucoma care which they believed could improve patient care.

Always positive about doing the stuff you know, there’s something new coming it’s always nice to have something new isn’t it. But particularly I think the it’s the fact it’s a device study and you’re interested in how the devices will actually improve patient care. So it’s about excitement of being able to introduce something new that’s going to hopefully make a difference really, yeah.

P037, Research Nurse, Site 2

Several site staff members reported having one or two concerns prior to the site opening. One concern related to patient participants asking about clinical implications of data collected, as this was not something being assessed within I-TRAC. Another found some of the flexibilities in the protocol (e.g. eligibility criteria, what actual tests were done for assessing IOP and VFs in clinic prior to the study) a bit worrying to start, due to past experiences on tightly controlled trials (compared to the I-TRAC feasibility study).

Yeah, there is also the fear, going back to feelings or thoughts, that if you’re doing this as part of a study as well, the patient wants to know, ‘All that effort I put in taking the measurements, what does it mean? Am I okay?’ that creates a challenge as well particularly when you’re just looking at feasibility and you’re actually not that interested in the actual data they’re collecting.

P036, PI, Site 1

Site staff perceived a number of benefits, particularly for patients, arising from home monitoring. These included a sense of control of their glaucoma through self-monitoring, greater convenience from not having to attend clinic as often, and increased chances of detecting significant changes more quickly, leading to better outcomes in terms of vision. Across site staff accounts, there was a strong sense of buy-in and a desire to see the home monitoring interventions work for patients.

Well that’s the thing, I mean it could be useful for people who live far from the hospital where there is a remote… access issue, it might be helpful, people perhaps with some mobility issues about them getting to the hospital but not stopping them from using the home monitoring device. Yeah, I think there are niches where it would be helpful, but I’ve always had in my mind that this would be a device for continuous monitoring for some reason, and that ultimately the aim is to keep the patient out of the hospital.

P035, PI, Site 2

The ease of running the I-TRAC study and the compatibility of its components

The findings around study conduct were themed in relation to ease of running the study and were composed of two subthemes related to study deliverability: (1) easy-to-run and low-burden study; and (2) quality and usefulness of training.

1. Low-burden, easy-to-run study

Overall, all participating site staff felt that I-TRAC was easy to run and its components (i.e. study processes) worked well together. This was qualified by reports that the I-TRAC study was well explained, supported and was perceived to be relatively low burden for participants and sites.

A lot easier than I thought it was going to be. I think as well, I did sit down the day before and go through the paperwork and just have it in my head, ‘This is what I’m going to say, this is how I’m going to do it’, and have it laid out. I think you have to do that to seem confident in front of the patient. Otherwise, if I’m faffing about, they’ll be like, ‘You don’t know what you’re doing’, and I’ll not know what I’m doing. I think the combination of the training plus the written material that was there too, it was fine.

P038, Research Optometrist, Site 3

Site staff reflected on I-TRAC being surprisingly easy to recruit for and felt this was due to patient participants having a keen interest in something that was novel and that had a clear benefit for them in terms of reassurance, control and burden (being able to monitor from home instead of travelling to clinics).

think maybe for this particular study, home monitoring, I don’t know if that’s such a big issue because we found that recruiting was really easy and we recruited just from our face-to-face clinic, and really we recruited the majority of the patients for each phase in a day. It was literally asking consecutive patients and most people said, ‘Yes, I’d be very interested in this’, and that was surprising, that’s certainly not an experience that is very common in studies. I know there were… I mean it was a very wide inclusion criteria, but nevertheless you still expect perhaps a fair few to decline.

P036, PI, Site 1

Site staff reported few challenges in terms of study conduct. Those challenges raised involved logistics (equipment return and appointments), an issue relating to accessing data from the tonometers (i.e. one site forgot to download the data from the tonometers before reissuing the devices to the next set of participants), and research and information governance issues prior to site opening.

I think we may have had some challenges in getting the equipment back and how the patients… how we got patients back into the system, because I think that was potentially an issue in terms of clinicians’ time because there’s also the idea of getting… picking up all the other data like the visual fields and things like that, and these things all… the organisation of those things and organising when the patients were coming back about picking suitable times where clinicians were available and the actual patient was available, the logistics of that with lots of people.

P037, Research Nurse, Site 2

1. Quality and usefulness of training

The training of patient participants to use both pieces of home monitoring equipment was a critical and central component of this intervention. The consensus among site staff was that the time to train patient participants reduced with practice, initially requiring more time than had been allocated (sometimes up to 2 hours) but reducing to the allocated time (30–45 minutes) after they had trained their first few patient participants. Site staff reflected that they felt the quality of their training improved after the first few participants, as they gained in confidence and developed the skills and tips to help patient participants learn to use the devices. They reported that the patient participant-facing training materials (a step-by-step handbook with images, and several YouTube videos) were a good resource for patient participants. However, some site staff reported that they felt not all participants were using the provided support, resulting in a number of contacts from patient participants requiring additional support during the monitoring period. Site staff reported that they found the site training materials and site initiation visit training (protocol and study instruction manuals) to be useful, coherent and generally to cover all that was needed, ‘Yeah, the step-by-step guide was very, very detailed, the fact that there was one for the patient and one for the practitioner as well was really handy’ (P033, Research Optometrist, Site 3).

However, one aspect of site training, the instructional training for using the tonometer, was delivered by a representative from the commercial partner. The consensus across site staff was that this training did not help them to teach patient participants how to use the tonometers. Several mentioned that they felt the representative was there in a sales capacity, not to deliver training, and had no practical experience of using the device so could not answer a number of questions the site staff had.

We need somebody who’s thorough, who knows all about it, who knows from a patient perspective as well. I’ve had some fantastic training in the past with other companies and real trainers but I kind of feel there was a sales element there. So no, that wasn’t very good for explaining.

P039, Research Nurse, Site 1

Many mentioned that the YouTube videos and written instructions were more helpful than the commercial partner training in terms of helping them master how to use the equipment and train patient participants to use the equipment.

Yes, because I mean I actually learned more from watching that YouTube training than I did actually from the sale rep side of things, so, yeah. Yeah, certainly the YouTube thing helped once we’d got to grips with the YouTube address.

P037, Research Nurse, Site 2

The feedback from sites was that equipment training should be developed and delivered by those with experience of using the equipment and training others (preferably patients) to use the equipment, so that they have knowledge and awareness of common problems patients have and how to overcome these.

Eligibility, recruitment and representativeness of sample

All three sites reported that they found it easier to recruit than they had anticipated at the outset of the I-TRAC study. Across the sites, staff generally felt, however, that their samples were not representative of the typical glaucoma population. The three sites reported how they approached recruitment differently.

Site 1 recruited from a general face-to-face clinic and participants were first approached by the clinician/PI and if agreeable, referred to the Research Nurse. The PI reported that the selected participants were really keen to participate. The Research Nurse in one site, however, noted a trend where they felt older participants appeared more determined to master the technology, and the younger participants, particularly those still working, had other priorities. This was echoed by the PI from site 2. Conversely, some staff from other sites believed younger participants were easier to train, perceiving older participants as being more time-intensive to train and support with new technologies.

I must say though, when I was phoning some people, maybe one or two of the younger ones, they said, ‘We’ve not been doing it, just not got round to it’, etc., so that was a wee bit of a disappointment… They were more elderly in the first cohort but they were determined they were going to do it. I don’t know if it’s that generation, I’m not sure.

P039, P039, Research Nurse, Site 1

Site 2 recruited participants through their virtual clinic, which typically deals with straightforward glaucoma cases (i.e. glaucoma patients considered at low risk of progression due to stable disease and few comorbidities which may complicate glaucoma management). Site 2 staff reported that while recruitment was easy given the access to a large pool of suitable participants, the approach to recruitment involved the Research Associate spending time in the waiting area of the clinic with patients not pre-notified about the study by the clinician/PI. The PI discussed how while time-efficient (a high volume of patients to speak with in one half-day session), recruiting from this clinic only, as they did, would not represent all glaucoma patients. This is because those whose condition may be more complex would be seen in a different clinic and, conversely, those patients with straightforward glaucoma and OHT would be discharged to community services.

[W]e have discharged quite a few of our patients to the community, the very, very straightforward ones like ocular hypertensions, so they weren’t… there might’ve been one or two but they’re usually… if they’re ocular hypertensions in the hospital they’re usually there for a reason. So actually quite a lot of them… I’m sorry, that group would be missing.

P035, PI, Site 2

The Research Assistant discussed the difficulties of having around half of participants decline participation, with reasoning related to the person approached having no interest in a technology study.

But yeah, generally I think my group is diverse-ish, but obviously there’s that inherent lack of diversity you get by just sitting down outside a clinic saying, ‘We want you to take some technology home’, and then half of the patients say, ‘Oh, I don’t do tech’, and just don’t want to talk to you ever again.

P031, Research Assistant, Site 2

The Research Assistant also discussed how they had found it challenging to recruit a diverse sample of participants, especially in relation to ethnicity. An issue reported by several members of site staff was that of language, perceiving language issues as being a significant barrier not just to accessing the information leaflets but also learning to use the devices in a second language.

Yeah, I did recruit, I did very happily manage to recruit a lovely I think Caribbean lady who’s part of the study, but other than that I have approached a lot of people and I think sometimes there have been language issues, the fact that it’s all in English, the information sheet’s in English, that’s been an obstacle I think a couple of times, although it hasn’t always been explicitly stated that that is the issue, yeah.

P031, Research Assistant, Site 2

The PI from this site questioned whether it would be more helpful to have a more diverse study team, or at least a more diverse team of recruiters, as he felt a shared ethnicity may go some way towards the study having a more inclusive approach to recruitment of glaucoma patients.

Well I think there is a certain racial thing, I mean if you had a black guy, black doctor asking black people to participate in a study I think that’s much better than some old white fella like me asking them to do it, and I think that just makes it much more accessible.

P035, PI, Site 3

For site 3 the seven participants were specifically targeted as they were known as individuals with the educational and technology familiarity needed to participate and would enhance recruitment efficiency. Site 3 staff reported they felt their participants were younger than typical glaucoma patients. The following excerpt presents this site’s views when asked about how representative their sample was.

Well I don’t… well I expected ours [when considering the I-TRAC sample of patients] to be older….

P032, Research Optometrist, Site 3

Yeah, not in terms of age and probably also not in terms of familiarity with electronic devices.

P033, Research Optometrist, Site 3

But then I suppose it’s good to have a range and different types of glaucoma as well as part of the study, but on the whole they’re not.

P032, Research Optometrist, Site 3

Unintended consequences arising from the intervention

Three members of site staff (one from each site) discussed how they had initially been concerned about the intervention causing patient participant anxiety arising from increased focus on their disease. Despite these concerns, there were no reports from any site staff of any actual or suspicions of induced anxiety within their patient participants. One site reported two patient participants who required treatment, an outcome that was triggered by patient concerns relating to the tonometer readings which resulted in contact with their consultant. In both cases, the site staff member reported that the patient participants did require treatment and that both patient participants were glad they had participated so it could be detected.

There were two participants who actually had to come in as a result of the pressure that they measured. One of them went to the optometrist and then she came in to us, and the other one just came directly to us because… if they hadn’t had the equipment, they wouldn’t have known their pressure had gone up…. Both these people liked the equipment, they liked using it, they thought it was worthwhile.

P039, Research Nurse, Site 1

Adaptations or considerations required for a future trial

Site staff reflected on several aspects of study conduct that would require careful review prior to progressing to a full trial. This included: (1) the additional resources that would be required; (2) aspects of study design to be improved (i.e. determining the ideal follow-up duration); (3) participant selection criteria; and (4) improvements to intervention training. These subthemes are presented below.

1.

Increased human and financial resources

In relation to resource use, sites reported that to feasibly run I-TRAC as a full trial with more participants, increased logistical and technical support would be required to manage the devices and assist participants with any queries. There were also concerns about the perceived (significant) costs of acquiring a large number of devices and accessories (e.g. batteries), and how a large trial would operate if equipment had to be rotated between participants in an effort to reduce technology costs.

Does it mean you’re going to follow patients for a year? Does that mean you’re going to follow patients for… if you have to follow them for a year, well essentially you need equipment for every patient in the trial, that’s costly I expect, and you need backup to manage the technical aspects that they may have, repairs, all of those sorts of things. I suppose it’s all manageable, it’s just understanding it before you do it, which I guess is what this is about.

P035, PI, Site 2

While it was not conducted as part of the I-TRAC study, it was recognised that the human resource requirements for reviewing and acting on the data received from the devices in a future trial would have implications for delivery. Site staff were concerned about the substantial time required to process this amount of data. Suggestions to make this achievable included exploring methods for efficient data review (e.g. making use of automated algorithms that would trigger a message to the clinician only if a test is abnormal, or AI technologies) and integrating data into electronic patient records. There were also concerns about the impact on resources required to address readings from the home monitoring devices that suggest the need for further intervention.

I guess at the moment it would be… perhaps it’s not very efficient because it’s all very innovative. But I guess, you know, you can incorporate this into electronic records and there will be, let’s say, the potential of having the information displayed in a different way, that makes the clinicians’ tasks easier. It depends, you know? At the moment it will be time-consuming because we don’t have a good system to process this information. It would need to be well thought to make it efficient.

P034, PI, Site 3

2.

Changes to study design

Most suggestions related to study design were improvements to the intervention, such as involving carers/family members in the monitoring, offering home visits instead of patients coming to the clinic for training, improving the electronic reminders to avoid spam filters, and to consider restricting the study to evaluating only one technology rather than multiple technologies at the same time.

One way that might help simplify it, and also perhaps help identify the key research question would be to just test one thing rather than testing visual fields and pressure. I don’t know which would be best. The technology for pressure is more developed but visual fields are probably more important.

P036, PI, Site 1

Exploring the potential for utilising participants’ social support within both the training and use of the devices to be more inclusive of those at greatest risk of vision loss was also suggested.

The other thing I suppose is that we focused on self-measurement, but I think that looking at carers and how they can help and support the patient is really important too because I find it hard to put eye drops in my own eyes, now if I had glaucoma I’d be asking my partner to do it for me, I’m terrible, and I think, well patients do that too, and I think having someone to help you and support you really to take the measurements at home would help because one of my worries about this sort of technology is it’s the vulnerable groups, the people who are probably at highest risk of losing vision, the people who can’t put eye drops in themselves, and they’re the people who might need it most and they might not be able to do it. So I think looking at, if there was further work with this, I think looking at carers as well would be useful.

P036, PI, Site 1

Within I-TRAC, digital reminders for home monitoring devices (such as e-mails and text messages) were provided to enhance adherence. Site staff highlighted the need to ensure these reminders are received as intended and not filtered out by spam filters or blocked by mobile phone providers. One site staff member reported that the need for patient participants to travel to the clinic for the training and provision of equipment acted as a barrier for several participants, undermining the potential benefit for patients who cannot travel. For this site staff participant, they suggested that providing study visits at home may improve recruitment of those who are often less represented in research.

I think it would be entirely possible. I think you… I think definitely within the capacity of going and visiting patients there’s no reason it couldn’t be done at the patient’s home, it’s just whether that’s a reasonable use of time I think.

P031, Research Assistant, Site 2

The importance of selecting research sites and staff purposively was also commented upon, due to the importance of confidence with technologies and the patience required to help teach others. A further recommendation in relation to design was the potential for a further study to determine the duration of follow-up.

I wonder if you almost need another… a pilot study to see how long do you need to… do patients need to keep them for to get full results, because that would be a key question in terms of how many devices you need and then the cost of the study….

P036, PI, Site 1

3.

Recruitment of participants

When asked about their expectations for recruiting a larger number of participants for a future trial, there were mixed views. Although site staff felt recruitment for this feasibility study, with very broad criteria for participant selection, had been easy to recruit to, there were concerns about how easy recruitment of a more tightly defined eligible population might be.

I think the fact that we have a virtual clinic where there is a large pool of what would’ve been considered suitable patients for this all coming to the same place in a non-stressful environment made a massive difference to our ability to recruit patients, and that has two real implications for me. Going ahead to a randomised controlled trial, which I guess is what you’re hoping to do, I would be very much looking at recruiting centres which run virtual clinic services because the patients are quite easy to access or more easy to access. But it also creates a barrier as well because in a way they are generally more straightforward patients and it depends who you’re trying to evaluate these devices on. If you want more advanced patients then you’re going to have to look elsewhere and then it will be harder to recruit. This was particularly in relation to the typical age demographic of glaucoma patients, and assumptions about older patients abilities to use the technologies.

P035, PI, Site 2

Especially at the moment, we didn’t have any restrictions, the availability was anyone with glaucoma, so that was basically… if in the context of a trial, you have a particular population, it might be more difficult… I can imagine from my impression, this technology might not be for everybody. I guess if there is a trial, there will be an inclusion and exclusion criteria, there would be a particular population of glaucoma people, people with glaucoma. I guess, you know, I’m just thinking that perhaps recruitment will be difficult.

P034, PI, Site 3

Site staff discussed how there were uncertainties about which glaucoma patients home monitoring would best meet the needs of in terms of usefulness for the patient, usefulness for the clinician, and from a cost-effectiveness perspective.

Then yeah, maybe those ones who find it difficult, who are stable but also find it difficult to come to appointments, and I agree with [other participant] that those younger ones that are familiar with technology and have busy lives.

P032, Research Optometrist, Site 3

That’s the biggest thing because I think… yeah, it’s a bit… it’s a chicken and egg situation isn’t it because you don’t really know what sort of patients it might benefit… But yeah, I think it’s really hard because it’s just speculating and guessing because we don’t really know what kind of patients or what scenarios it would be of use.

P036, PI, Site 1

For one PI, the difficulty arose because the two technologies being utilised together as the home monitoring intervention in the I-TRAC study are measuring different outcomes. For them, each of those outcomes independently would help very different subgroups of glaucoma patients, and those who would benefit from both were a very exclusive group. Clarifying who this intervention would be most appropriate for was indicated to be an important priority prior to any full-size trial.

But I think separating the pressure and the visual fields and looking at the different indications for those is really important because it’s very different situations where you might do one or the other. So yeah, I think there’s almost three situations, there’s three questions, how can we use home glaucoma monitoring, but how can we use home fields and home IOP because they’re different, different groups.

P036, PI, Site 1

4.

Improvements to intervention training (for staff and patient participants)

Many site staff participants reported suggestions as to how the training for patient participants could be improved. The first was to identify and develop ‘good trainers’ at research sites – those who are patient and able to promote confidence and reassurance among patient participants. It was considered helpful if the person delivering the training had some awareness of the patients’ clinical and disease status, as this provided insightful information as to how best to help the patient learn to use the devices.

I think also the other thing that I noticed was that it’s helpful if the person teaching the patient has an understanding of the patient’s condition, because we had a patient where they had poor vision on one eye and it was really hard to get them to measure the pressure in that eye, but they could do the other eye, and we didn’t realise initially, and then once the notes were opened and you’re like, ‘Oh, the vision’s quite poor in that eye’, it became apparent why there was that big difference. I think just having some… I think if you just dump a patient with a technician who knows nothing about that patient, that might not be the best way to teach them.

P036, PI, Site 1

In terms of the specific tips and stuff, it’s generally been stuff like knowing that some patients find it far easier with one eye closed or knowing that actually sometimes I just need to go in and adjust the device a bit myself and help them get it in the right position, and then once they know how that should feel, a lot of them just… they can remember that and it helps a lot.

P031, Research Assistant, Site 2

A number of site staff wondered if group training for patient participants would be helpful.

For me, they’d be in an environment where the equipment would be there for them to play with. As I say, internally here, there was issues about getting space so I didn’t think that initially helped. If I had a conference room for them, get the videos up where they can see it, huge, not make it too long because they’ll maybe get a wee bit tired, provide a wee provision of some fluids and what-have-you, cups of tea, they’d like that, it would be a social thing for them. Have the equipment there that they can play with at the time, each one get a wee shot themselves, see how they feel with it, ‘How do you feel? Go to that table, have a wee shot’. Somebody like myself always there on hand to explain to them and also demonstrate. I think a huge presentation behind us, like the videos you get on YouTube showing them, big enough for them to see, take questions, play with it, I’m always there to help them.

P039, Research Nurse, Site 1

This group approach was perceived to be a potentially efficient approach of delivering training for sites, but the additional social aspect of the training environment was also expected to be helpful for patient participants.

It would probably be easier if we just had an I-TRAC demo and just brought everybody on the same day to do training, or maybe did group training with the patients rather than individually. That would be something that would save time, so one person can maybe teach three or four people at the same time and maybe those people have a bit of support as well because they’re all part of the same study… People with conditions can feel isolated sometimes if they think they’re alone, but if they have a group where maybe they come to the clinic once a month, like you say, it’s a social thing. I know older patients love coming, getting tea and biscuits and a chat, that’s why it takes so much time!

P038, Research Optometrist, Site 3

Several aspects of training for site staff, directly related to the intervention devices themselves, were also suggested as areas of improvement for the future. Increased provision of more detailed technical information on the devices, what they are measuring and how they work, were suggested as ways of improving site staff confidence not just in the devices but in their ability to answer patient participant questions about the technologies being explored: ‘I think more clarity in what that test is actually doing would be good, there maybe wasn’t that much information on that’ (P032, Research Optometrist, Site 3).

Site staff were generally in agreement that they would like to see more training opportunities to increase the time they could spend practising using the devices and practising training another person to use the device, so as to increase their familiarity and confidence with the devices. Site staff also suggested several benefits of having spare equipment on site, such as that staff can continue to practise with the device while participants are home monitoring, and it would be helpful in supporting participants when they contact for help with their device over the phone.

I don’t know if this is doable in terms of resources, but to have a device, one of the [tonometers] on site that we get to keep to solely use as a practice device, and then if we’ve got new staff coming on board or whatever we’ve always got something there that we can do in-house training for people.

P033, Research Optometrist, Site 3

Pragmatic considerations for the delivery of site training also included ensuring a suitable training environment is made available at site for the training session; as was reflected on by one site, space can be limited at NHS sites.

There wasn’t much space for them [a suitable room for the commercial trainers to deliver training at site] here. It didn’t really hit off very well initially. Do you know what I would have liked? If we could have all gone to a wee training course, you know, meeting up with the other people. I know the budget is only so much but if we’d done a training course with the other people where we could all ask questions, take our time and been shown how to use it, we could take notes and things like that as well and then try and start a wee bit earlier.

P039, Research Nurse, Site 1

Multisite training days, where staff from across sites get together off-site to complete their device training, was mentioned as a preference by several site staff participants.

I think it’s something I wouldn’t necessarily think would need to be an organised event, but it’s something I’d recommend for future sites is familiarising themselves with the devices, maybe just letting them use some of the probes, just actually go through, ‘This is how the device works. If I’m going to use this, this is how I use it’, show it, get all of the members of the team used to how it works so that then they can explain it to patients quickly, because when I learnt how to use it, it was just the one meeting, I think was the only person who had it demonstrated on, and then from then it was just, ‘Now you’ve got to show patients how this works, and actually having a bit more experimentation in the meantime might’ve been useful, yeah’.

P031, Research Assistant, Site 2

Impact and influence of COVID-19 pandemic on perceptions of acceptability to patients and clinicians

Site staff were asked to discuss whether they believed the COVID-19 pandemic had any influence on recruitment to I-TRAC. No instances were reported where the pandemic appeared to influence a participant’s decision to participate. It was noted, however, that the pandemic may have made some patients more amenable to the use of technologies, through increased familiarity gained by using them during the pandemic to stay in touch with friends and family and use of remote health consultation services. Our PI in site 2 reported that more patients were now being seen in virtual clinics post pandemic, as patient care pathways have changed due to the pandemic, and as a result this could be said to have made recruitment easier as the more stable patients were available to be recruited from one clinic, as opposed to searching multiple clinics.

I wonder, just add to that, whether patients might be more happy to try technology and might be… I think there were certainly some people that had never used an iPad and Zoom before and now they do to talk to relatives and maybe, so maybe it helped, and people might’ve been worried about coming to clinics and, so the idea to monitor disease at home might, they might’ve been more open to that than they would’ve been before the pandemic.

P036, PI, Site 1

Table 19 provides a summary of the strengths and weaknesses of I-TRAC perceived by site staff. Table 20 reports a summary of site staff suggestions regarding a future evaluation of digital technology for home monitoring glaucoma.

TABLE 19

Site staff perceptions of strengths and weaknesses in conducting the I-TRAC feasibility study

TABLE 20

Site staff suggestions for future evaluation of digital technology for home monitoring glaucoma