Enthusiasm tempered with uncertainty

When asked how they felt about the I-TRAC home monitoring interventions for glaucoma patients, expert glaucoma clinicians were generally enthusiastic about the opportunities this intervention presented, described through terms such as ‘excited’, ‘delighted’, and ‘glad’.

I think it’s a great idea, I think we need to know this. We’ve got real problems with capacity in glaucoma and if we can have a trial that helps us better understand the patient acceptability, clinicians’ views on acceptability and the meaningfulness of the data and so on and so forth then it’s to be welcomed.

P002, Optometrist, Focus Group 1

These feelings were often reported in relation to home monitoring being a potential solution to national capacity problems across HES. However, enthusiasm was tempered with caution – as indicated by a number of conditional statements such as ‘I think it would be a great thing if it worked well and it was reliable’ (P008, Consultant, Focus Group 2) – demonstrating the connection between affective attitude and perceived effectiveness.

Concerns about negative affect for patients and staff

Expert glaucoma clinicians also reported that they felt many patients would be welcoming of this approach, highlighting how this could reduce patient stress and provide comfort, ‘some patients genuinely like writing the numbers down and they like to know what they were before and they get great comfort in knowing there are numbers written down that they have control over’ (P018, Consultant, Interview).

However, among the positive feelings, some clinicians reported feelings of anxiety and frustration as a risk of such interventions, which could be experienced by both staff and patients.

Digital data, I mean it does sound like this panacea and it’s wonderful, but actually when it doesn’t go quite so right it can be extremely difficult and frustrating, and I’m thinking about the age group of our patients, how hard it might be for them.

P004, Consultant, Focus Group 1

Ethical risks of remote and commercial data collection technologies

When asked about ethical issues this intervention may pose, many clinicians discussed aspects of data security. There were concerns about how to protect these systems from external threats, and the need for secure data exchange. Many felt that NHS systems of governance for IT should be adequate to adopt this; however, one clinician raised the need to consider how commercial entities also manage data governance:

[O]ne thing we’ve not necessarily quite touched on is the link with the commercial sector. So OKKO Health are a commercial sector organisation, whoever’s come up with [home visual field intervention] presumably is as well. So who owns the data, who’s responsible for the governance of the data, all of that.

P002, Optometrist, Focus Group 1

Also related to the commercial side were concerns about what would happen to data should a private enterprise withdraw from service, posing the risk that patient data could be lost and no longer governed by NHS data governance. There were concerns that these data could then be misused. There were also some concerns about safety, with calls for regular data auditing being required to establish the safety of this service. One issue raised by a number of clinicians connected to data control is the risk that patients may allow the device to be used by others. The potential risks from this include poor treatment decisions from non-patient data, issues about consent if researchers hold data that are not from the consented patients, and the potential for detecting eye diseases in others (not the patient). However, having user logins to prevent non-patients from using the devices was suggested as an approach that could overcome this, suggesting addressing this concern was not insurmountable.

I wonder… the temptation would be to try it on your family and friends. I mean, I don’t know. But they might say, ‘Have a go. This is what I have. Look at what the hospital’s given me’ which will mess up your data.

P010, Consultant, Interview

Intervention fit with principles of good care

There were a number of references to how this intervention fits with the principles of good care. For some, home monitoring fits well within the self-management framework, where it is considered appropriate to empower patients to manage their own health care. For others, there was concern that home monitoring could lead to reducing glaucoma clinics to little more than data monitoring hubs, leading to a loss of personalised and holistic care.

. . . there’s the reduction potentially of… it could be that glaucoma clinics are seen to be reduced down to data collection on instruments. We know that many of these patients have things they want to talk to us about, we know that they have things wrong with their eyes other than glaucoma and there’s a loss of the holistic approach to patients when things are reduced down to maybe a few questions on a proforma and some data supplied intermittently versus a face-to-face interaction.

P002, Optometrist, Focus Group 1

For others, limitations in the current system are contributing to the overtreatment of glaucoma, whereby clinicians, knowing it could be some time before they review a patient again, tend to provide medical treatment, on a ‘just in case’ basis. Being able to offer home monitoring was viewed as a solution to this and could reduce unnecessary medical treatment.

I think actually virtual clinics you often overtreat and also I guess many of you have these, what we call follow up pending lists, and certainly COVID’s made them go much higher, and you think a patient may have a problem and you say, ‘Okay, we’ll see you again in six months’, but six months maybe 18 months, and, so on that basis I think you sometimes overtreat because you just don’t know when you’re going to see this patient again. So the fact you could have a facility which can help with monitoring patients that you haven’t got capacity to see within the hospital eye service may enable you to treat less patients and therefore get less morbidity from their treatments.

P005, Consultant, Focus Group 2

For one clinician, while supportive of home monitoring, they wondered if this focus upon already diagnosed patients was missing a bigger problem in glaucoma care, specifically, the low level of detection of the disease in its earlier stage.

I think it’s fine this sort of study, but there is a bigger picture that there are lots of people in the UK and particularly in the developing world who don’t get picked up in the community and there has to be solutions to that that could involve these mobile technologies or whatever.

P009, Consultant, Focus group 2

Managing equity in patient selection

Clinicians frequently reported concerns related to equity and equality, particularly in relation to making decisions as to who should obtain home monitoring equipment, and what fallout there may be from those decisions. Frequent references to factors which may make it difficult for some to participate include accessibility, language, education, and technical abilities (discussed in detail in Perceived effectiveness). However, these led to ethical consequences: how to select and prioritise patients for home monitoring (particularly where there will be resource constraints), impact upon those who are not selected for home monitoring and the risk of creating a two-tier system. The latter was in relation to an expectation that patients may have to contribute financially towards equipment, which opens the door to one system for those who can afford and one for those who cannot.

Well actually, prioritisation. If say someone’s neighbour got this and someone else didn’t get it, word goes around in the glaucoma community: ‘Why did she get it? How come I didn’t get it. Am I more less important? Are you more important?’ That’s not going to go well, I suppose. Ethically how do you choose because you have a limited resource, so that limited resource ethic problem. Even if you gave them the option to buy it, then it’s also another ethical dilemma, because the rich are getting better monitoring. But, having said that, it frees up space, frees up one iCare for another person who can’t afford it, so I mean it’s that ethical dilemma of privatisation versus, you know, but… so that same story. But if they want to buy it, I think they should be able to buy it.

P010, Consultant, Interview

However, some argued that such a two-tier system would be inevitable and necessary to accommodate those for whom home monitoring is not suitable regardless of affordability. Clinicians’ accounts report the need for clear guidance as to how to select patients for home monitoring.

I suppose the other ethical issue is, and we’ve kind of touched on it with patients that are able to use these devices and patients that aren’t able to use these devices, if this becomes the gold standard of treatment for patients and then you have a patient that is unable to do it, how can you ethically not let them have treatment, do you then pass(?) that on to go and take the measurements as regularly in-house. That’s obviously years down the line.

P015, Consultant, Focus Group 3

Autonomy to determine the who, when and where of home monitoring

When asked about the advantages and disadvantages of this intervention, several attributes became apparent. The first was that there was a desire among clinicians for autonomy to decide how best to integrate home monitoring into usual care.

So I don’t think we could decide or should decide as to who can use it. I guess if the underpinning evidence is that these are meaningful measurements, then it’s about the training and accreditation and ability to autonomously make decisions by whoever is looking after the case mix of patients where that service is being commissioned, and that could be primary care or secondary care.

P002, Optometrist, Focus Group 1

Clinicians discussed varied purposes and patient scenarios (varied clinical situations and parameters) where these technologies could be helpful, and it became apparent there was no consensus among clinicians as to the ideal clinical scenario or patient for glaucoma home monitoring. Suggested clinical scenarios included phasing, reducing overtreatment, monitoring for progression, risk-stratifying/referral strategies, promoting patient self-management, increasing localised care (reducing hospital visits) and increasing clinic testing capacity for higher-risk patients.

[T]he way I envisaged it is more as a trigger to find the patients that need intervention. So this isn’t going to be the be all [and] end all care of a patient’s pathway, it’ll be a service where you can gather whatever data you can regularly that gives you a trigger to say, ‘Well actually that patient’s doing absolutely fine, don’t need to get involved’, whereas another patient, ‘Oh, they need to come in hospital, we need to have a proper look at them’.

P015, Consultant, Focus Group 3

It could also be valuable across patient groups as some felt the increase in data collected would be relevant and beneficial for all groups.

[A]ny data points you can get will enable you to have a better picture and diagnosis of what’s going on with a patient, but I think the more points you can get, the more data you can get, the better. I think that could be taken to any patient group.

P015, Consultant, Focus Group 3

Clinicians generally agreed with the use of home monitoring technologies for patients considered at low risk of progressing to significant visual loss, but some could see value in considering these technologies for higher-risk patients in specific circumstances.

[C]hecking your pressure having changed treatment in somebody who’s relatively low risk and your next clinical decision probably isn’t going to be surgery. Then again, you’ve got your ocular hypertensive, in fact we discharge ours to community optometrists, but if you haven’t managed that service with your local optometry committee then you could argue very low risk glaucomas could be monitored quite long term with fields and IOP.

P005, Consultant, Focus Group 2

I think the patient… the people who we see in the hospital services are high risk that need to be seen, because I think the addition of… additional pathology, the very nature of the discussions we have to have with the patients that a home monitoring system isn’t ideal. However, for… even with our high-risk patients, I think again it comes down to how accurate the data is. This might help with anxiety for some patients who are always worried about what their pressures are doing and obviously you can’t see patients frequently, but if you get a baseline with this then patients being able to measure their own pressure at home a couple of… two or three times a week, it might ease that.

P004, Consultant, Focus Group 1

There were several statements suggesting that defining an ideal patient was perhaps not the right approach and that there should be flexibility afforded to individual services as to for whom and how home monitoring is used.

So I suppose each local… well it will have to be very individual to each place and how they work isn’t it as to who looks at the data and which group of patients, that kind of thing as well.

P016, Consultant, Focus Group 3

Some mentioned that age may be a deciding factor. Partly this was in relation to ability to use the devices, partly this was in relation to those considered at risk of the greatest QoL losses from losing visual function at younger age, and partly this was in relation to a subgroup of glaucoma patients who may still be working, and therefore for whom attending clinics for testing is perceived as being more burdensome.

I’m thinking of young normal tension glaucoma patients, so I mean 50. They’re the ones I worry about because they’ve got a lot of life to live and so they’ve got higher likelihood of going blind. And the normal tension glaucomas have trans-central scotomas, so they’re very sensitive. As soon as they lose one decibel that’s it, the whole vision is really bad, so those might be the ones.

P010, Consultant, Interview

Some clinicians felt that home monitoring may not be able to replace usual care for complex cases (e.g. multimorbidities) where information gathered from face-to-face observations is really important.

[T]he people who we see in the hospital services are high risk that need to be seen, because I think the addition of… additional pathology, the very nature of the discussions we have to have with the patients that a home monitoring system isn’t ideal.

P004, Consultant, Focus Group 1

Data relevance and integration

Distinct but complementary to concerns about data ‘ethicality’ was the intervention coherence concerns about the reliability and relevance of the data collected, and how these data will be integrated with electronic medical records. Those claiming to be less familiar with the evidence appeared sceptical that these technologies would produce reliable data to make meaningful decisions about patients’ care. There was disagreement about what aspects of visual health need to be measured; some agreed with both IOP and VF, others felt only IOP is required, others VF only and others worried about the lack of data in relation to other visual health domains, such as disc imaging and acuity, often measured in clinic. While it was agreed that data must be meaningful to clinicians (i.e. the measures being assessed are useful to make clinical judgements), there was little agreement regarding exactly what measures are relevant to make home monitoring meaningful, ‘[u]nless you can make the disc images cheaper and the pressures cheaper, then you can’t just rule out fields. I mean, fields itself are known as not very useful, if you catch my drift’ (P010, Consultant, Interview).

Some were so enthused by the evidence for glaucoma home monitoring that they were keen to see what future technology developments could achieve and made suggestions as to what else this home monitoring intervention could look to incorporate and overcome limitations from just assessing IOP and VF (data-related opportunities to improve for the future). For example, several clinicians discussed a contact lens device which can measure IOP or technologies permitting home disc imaging.

I was just wondering if when we’re talking about disc photographs, there are a few devices now that you can attach to smartphones to give you a disc photograph, I’m not sure would it be worth expanding to that you know, and you get your disc photograph.

P019, Consultant, Interview

Some participants reported being worried about how these data would integrate with existing medical records and any associated time required to input the data into existing systems. This links with perceived burden, discussed below. A solution put forward by many was to make the intervention software compatible with electronic patient records so that results can be immediately integrated without additional effort. A further solution proposed by some participants is the future use of AI for helping review the data as it comes in, making it quicker for clinicians to interpret the data collected remotely.

I’m pretty sure with artificial intelligence, we may be able to even define that what has changed has really change[d] or not, and then we can fast-track that… the results can be interpreted with the AI, I think I can see in five, ten years’ time, this is the way we do.

P001, Consultant, Interview

Support for patients

There was also significant importance placed upon the need to ensure the intervention is supportive for patients to address potential concerns. Participants felt many patients would require professional reassurance in understanding this intervention, particularly in terms of educating patients to understand that readings can fluctuate day to day and that it is the overall trend that the clinicians are concerned with rather than daily readings. Clinicians felt this would be important to prevent unnecessary worry.

You know, as long as they know that someone’s going to tell them I want a report of it or something, then all that hard work’s not for nothing, then I think that’s fine… then you could tell them that everyone’s pressure varies with… there’s a variation every day etc, and warn them not to worry about it, and the whole point is that we gather data over time so that we can make a judgement at the end of the day, so don’t worry about that. If you give them a clause, that will be fine….

P010, Consultant, Interview

Limiting access to results or reconsidering how results are presented to patients was one area of patient-related opportunities to improve in the future suggested by several participants. For example, more generic feedback, such as informing the patient that they had completed the tests correctly, was suggested to be a compromise to maintain engagement and prevent worry from results that may not be fully understood by the patient, causing unnecessary worry.

[W]e should just say in the app, ‘You have correctly answered 75% and you are among the top grade who have done the test very nicely’, or ‘You’ve done test reasonably, but it could be improved if you pay attention to these things’, or, ‘Your test needs to be repeated’. That is the only feedback going to the patient, and then we say, ‘Okay, now, you’ve done the test, it will be reviewed by the doctor… medical team, and we will come back to you.’ So that when we inform them that the situation is getting worse or the visual function is getting worse, we give them the solution there and then that, ‘Okay, we reviewed your results, your results shows deterioration, we have looked at your management plan, and we suggest you should be changing this, and after this you will be reviewed again by somebody.’

P001, Consultant, Interview

Another solution proposed is to offer a simpler intervention in the community for those struggling with the technology, for example facilities at shopping centres, etc.

This could even be done by a trained station in a supermarket, and we can just tell that person to go… you can’t do the technology for somebody, maybe there to help you guys, they don’t need to come to the hospital. They could go for their weekly shopping, and they go to a booth where there is one trained person who could have these appointments for these eighty, eight five… and those people who cannot cope with technology. And it is then outside the hospital, so we basically take it to the community in real sense.

P001, Consultant, Interview

In addition to clinician-provided support, participants felt this intervention would likely require utilisation of social support, making use of family and friends. For example, reminding patients to perform their assessments or physical support to use the devices.

[I]n the same way that we’re quite content to have, a lot of our elderly patients have their partner put their eye drops in, well it can be a team effort. Sometimes elderly people cope rather better when, ‘Oh, yes, my husband always remembers to do this’, who locks the door, there’s a certain way, people survive together don’t they and I just wonder whether there’s something in that for some of our patients. It goes beyond this technology really, but yes, you’d probably need to involve the younger generation.

P002, Optometrist, Focus Group 1

Beliefs about cost-effectiveness

Two clinicians reported that they believed the proposed intervention would be good value for money.

Well, it could potentially reduce the cost. If you’re not having to bring patients in, you don’t require transports, you won’t need technicians to gather the data of the patients as they can gather the data themselves. It would be an actual virtual data collection, and fewer staff costs.

P018, Consultant, Interview

Several clinicians specifically stated that this approach is not good value for money, and many were concerned about the cost of the equipment, particularly the home tonometer. Additional costs raised were maintenance of this equipment, having spares in case of damages, and the staffing to train patients and review data.

[W]e could be looking at wasting our resources looking after too much data points from patients who are well and were not really at risk of going blind with their glaucoma and are we diverting our resources be it instrumentation, time looking at the data and collecting the data which could be used more effectively for other patients.

P017, Consultant, Focus Group 3

I think I prefer something which would be a bit even easier than Home iCare tonometer at the moment, and perhaps less expensive as well, because we give it to such a big population and if you can just give it to one person they will take it for a week, so just calculate for one person the population how many Home Care eye tonometer you need, and then if you extrapolate that cost and the overall benefit you are going to get from that, I’m not convinced that it is for everyone.

P001, Consultant, Interview

Some felt this would make it difficult to obtain management or clinical commissioners’ support for such a service.

I think the distribution and the (inaudible) the devices and persuading the health and social care service that we want to spend thousands of pounds on purchasing these things, I think that could potentially be an issue.

P019, Consultant, Interview

Adjustments to address affordability

However, several clinicians perceived that it could be made more affordable, firstly by looking at app services and dropping the iCare tonometer or asking patients to purchase their own equipment. One clinician felt that the costs would reduce as the technology improved and became more mainstream.

Yeah, that’s where your visual field app may be more cost effective in terms of economics and access to the piece of equipment that the patient has to take home, perhaps an app or something that they can download on to a tablet with a licence, it may be much more generalisable than to take home an iCare HOME.

P009, Consultant, Focus Group 2

Clinician burden

Burden upon clinicians was perceived to stem from several aspects, most commonly from concerns about finding the time to manage services and review and action patient data.

[S]o you’ve got all of these other notes and scans and things to look at from virtual clinics and I think this will add to it, even though these patients aren’t clogging up your clinic they are sending you lots of data which you need to be on top of, and if you don’t action it then they’re going to miss out anyway. So for example, if someone’s in trouble you need to be able to action that and find a space in your clinic for them.

P011, Consultant, Interview

There were also concerns about the potential to increase patient caseload and unscheduled care. This was mostly in relation to spurious results and anxious patients making contact concerned about their results. One clinician raised the burden of adapting to change and learning new skills, likening it to when electronic medical records were introduced, and discussing the time and effort it took to learn a new way of working.

But I suspect what will happen, potentially, is that as the numbers go up, we’re able to deal with more patients and you’ll just end up with a bigger workload, more patients under your care, I guess.

P018, Consultant, Interview

Conversely to these many concerns, one clinician felt that there would not be additional burden from this service, but instead a change in burden, away from the footfall of patients towards an increase in administrative burden, ‘It’ll lighten the burden in terms of footfall perhaps, but it increases the burden in terms of additional or admin work that you have to do’ (P011, Consultant, Interview).

Several clinicians discussed burden more in terms of its impact upon services rather than themselves as individuals. This was often qualified with statements such as ‘I’d have to delegate’ as clinicians felt they did not have time to take on these roles. Tasks to be delegated included maintaining and distributing equipment, training patients, and reviewing data.

Then you have to find the people that are suitable and then you have to train someone who’s able to train the patients on how to use it and someone to distribute the devices, to maintain them, to chase them up when they’re not brought back, to clean them, and then someone to look at the information. So probably quite a lot of work actually.

P019, Consultant, Interview

Patient burden

Several references were made in relation to burden upon patients. While many felt this service could reduce patient burden (linked to and discussed in Intervention coherence), there was some concern about the psychological burden patients may experience, particularly in relation to those who may worry about their results (see Perceived effectiveness). There was also a suggestion about the burden of additional tasks this would involve for the patient, such as collecting equipment from the clinic. One clinician wondered whether, by the time patients collect equipment and learn how to use it, they could have just had their tests done in the clinic, ‘So you’re kind of thinking if they have to come and pick up an iCare [tonometer] you might as well do it’ (P019, Consultant, Interview).

Anticipated outcomes

In line with different perceptions of the purpose of home monitoring, several expected outcomes were reported: improved management from increased data collected, detecting progression quicker and preventing sight loss, either directly from the monitoring patients or indirectly as home monitoring allowed clinicians to spend more time with high-risk patients in clinic. Several discussed how this intervention would be more convenient and less stressful and burdensome for patients. However, one clinician felt that some patients would demand in-clinic services regardless.

Balancing benefits and harms

There was mixed certainty across the clinicians’ accounts as to the degree of clinical effectiveness that could be achieved. Some clinicians were confident that patients would benefit from this intervention. However, a number of clinicians focused upon the potential unintended consequence of increased patient anxiety arising from monitoring. The impact of this was unclear, but clinicians were concerned that rather than empowering patients with responsibility and knowledge, it could add psychological burden. Some were concerned that any benefits in terms of clinic capacity could be eliminated by increase in patient contacts from the ‘worried well’ or ‘unreliable data’. Statements about expected outcomes were often tempered with ‘potentially’ reflecting uncertainty.

This might help with anxiety for some patients who are always worried about what their pressures are doing and obviously you can’t see patients frequently, but if you get a baseline with this then patients being able to measure their own pressure at home a couple of… two or three times a week it might ease that. But at the same time sometimes it can make it worse (over speaking) get phone calls unnecessarily coming back to you so it’s going to be a double-edged sword, I think.

P004, Consultant, Focus Group 1

A number of additional factors which could enhance or limit effectiveness were raised, falling under categories of data and technology limitations, patient demographics and health, and service support and culture.

Impact of perceived data and technology limitations upon intervention effectiveness

An overarching theme was doubt as to whether current technologies are reliable enough to home monitor safely and effectively. Specifically, many clinicians were not convinced that they or their patients would find these technologies acceptable, adversely impacting effectiveness through low engagement.

I still don’t know how reliable these technologies are at the moment so I can’t really comment whether we are going to have meaningful data from this at the moment, certainly with the visual field at home I don’t know personally, yeah… So I think yes, certainly there is a role for this in future but I think there is still a way to go before it comes common practice based mainly on the technology.

P017, Consultant, Focus Group 3

We have to be clear about how meaningful it is, they’re measuring visual function as has been suggested in patients who’ve got a range of other conditions, and clinicians as I say can disagree on routine perimetry. So I’m going to take some persuading, I feel, that the visual function measures are meaningful and valid for glaucoma care personally. That doesn’t mean I’m saying that… I just feel that the jury’s out and we need that data to be confident.

P002, Optometrist, Focus Group 1

Patient characteristics impacting intervention effectiveness

Several patient demographic and health characteristics were stated to have potential impact upon how effective these home monitoring interventions could be. Clinicians were concerned about the physical and cognitive abilities required for this intervention, fearing that a substantial proportion of the glaucoma population would either be excluded from this intervention or would find it unacceptable and would not engage in the first place. This would limit its impact.

Yeah, they have to have the manual dexterity for it as well. So I suppose that will exclude a few people from the clinic able to do that, particularly for the elderly patients.

P019, Consultant, Interview

There are groups of people who are frail, vulnerable, who are serial DNA’ers [Do Not Attend] or whatever and they might be tricky to persuade to do these home-based tests….

P002, Optometrist, Focus Group 1

Impact of service levels factors on effectiveness

There were also several references to effectiveness being related to wider service issues, such as support from NHS Trusts to staff such a service appropriately and the culture of the work environment. One clinician discussed how adoption of this intervention would likely be gradual, requiring time for people to learn and adapt.

So certainly there’ll be a learning curve and I’m guessing after the learning curve people might abandon in it or… like in anything that’s implemented. Certainly like electronic patient records, people hated them to start with. I think now we’re in a situation I really can’t bear to look at paper notes, but it did take us a while.

P016, Consultant, Focus Group 3

Low self-confidence and confidence in patients to use technology

There were fewer statements reported in relation to self-efficacy, likely due to the hypothetical nature of the interviews in relation to the use of these interventions. Clinicians were generally not confident about delivering these home monitoring interventions. Confidence concerns were often related to low confidence in the technologies and doubt regarding service support.

. . . have to say our Trust is very… anything would need a lot of planning and a lot of business plans and those kind of things. I don’t know whether it’s just my Trust, but I’ve certainly seen making anything happen in this Trust is really hard.

P016, Consultant, Focus Group 3

I’d have to say without having all the data, the validity data to hand I’d have to say not very at the moment.

P004, Consultant, Focus Group 1

Several clinicians discussed concerns about the self-efficacy of patients to participate in this intervention. Several aspects were perceived to be quite challenging for patients’ confidence, including worries about using the equipment, risks of damaging equipment, and some patients really struggling with other aspects of eye care.

I don’t know how much an iCare HOME costs, but I can imagine that there’d be some patients who’d be frightened of breaking it or damaging it and then having to kind of pay for it.

P018, Consultant, Interview

I’m really quite select(?) who I give it to because it’s amazing actually how many people just don’t like anything to do with their eyes and it’s all they can do to put their eye drop in.

P019, Consultant, Interview