Analysis of interviews with PWD and HCP identified several themes related to changing screening intervals: the acceptability of changing screening intervals in conjunction with conditions and safeguards attached to 6, 12 and 24 month screening intervals; the safety of the RCE; and the macro impact of changing screening intervals.
Acceptability of changing screening intervals
The majority of PWDs in both Phase 1 and 2 expressed the view that risk-based variable screening intervals were potentially acceptable. The views of PWD included the concepts of pragmatism and diverting any cost savings towards other PWD who may need to be seen more often.
You know if they don’t need them every year then yes, why do them every year. And so I would rely on the practitioner to make the best judgement. (David PWD, Phase 1)
Yes, you know because if your eyes are not going to go any worse it is saving money and time isn’t it, where, they can’t fit everybody in can they….it saves all that money and so it gives other people a chance of getting seen doesn’t it? So, I agree with that. (Susan PWD, Phase 2)
For HCP, the majority were also in favour of introducing risk-based variable screening intervals.
I think it will be fantastic because as you say there are lots of patients that you can quite happily review in 2 years, or 18 months 2 years, so I think that would be a definite benefit. (Sally HCP)
I don’t have any objections, as long as it is evidence based you know if there is evidence for it being 24 months, then I am fine. If there is evidence for a particular group being 10 years, you know if that what the evidence shows, I am very much things have to be supported by evidence. (Andrew HCP)
However, implementing such changes to eye screening was accompanied by a range of caveats which are discussed in relation to the 6, 12 and 24 month screening interval. The norm is currently 12 months, so we explore the other options in some detail below, as for some PWD this is more frequent than currently, for others no change, and for many, twice as long
Acceptability of 6 month screening interval
Within the variable screening model, allocation to a 6 month interval meant that a PWD was considered to be at high risk of developing STDR. However, for some PWD participants, this rationale was not fully understood, with the shorter screening interval interpreted as a security for checking eyes, with the longer screening interval of 24 months unwelcome as it was seen as too prolonged a time to be without eye screening. Additionally, some PWD viewed screening as a preventive measure in developing DR.
Well, early detection is better for the treatment you know what I mean, if you find something drastically wrong with your sight, and they can repair it, a lot, and if they find it earlier, they could repair it. (Arthur PWD, Phase 2)
Some PWDs, notably in the 55 plus age range, wanted to be screened every 6 months, or even every 3 months, and this related to a belief that they were more susceptible to developing eye disease related to diabetes and consequently needed to be monitored more often.
When you’re over a sort of certain age like say 60 that’s when things start going downhill - 60. I’m not saying like everybody - everyone’s different, aren’t they - but the way I look at it is that I think it should be every six months ... if you’re a type 1 diabetic you should have it every three months ... over a certain age every six months. (George PWD, Phase 2)
For HCP, the 6 month interval was welcomed unequivocally as a safeguard for high risk patients, and it was stressed that this would have to be clearly communicated to PWD.
I do like the idea of the medium risk, sorry the high risk ones coming back every 6 months… so I do think that that would be a good thing to have that as a standard if they were high risk, bring them back in the 6 months I think that would be good. (Sarah HCP)
…and there certainly are patients who need more closer care, which can’t always be determined just by looking at a picture of their eye ball essentially. You know they may be patients in certain ethnic groups, erm… and certain combinations might, I don’t know, but it might be for example Asian ladies from I am just talking off the top of my head, from a Pakistani origin for example, might it might be part of their societal erm… nature to not necessarily come to attention as much. (Andrew HCP)
As evidenced above, there are some tensions within PWDs’ understandings about the 6 month screening interval, with it being seen as clinical surveillance which was a reassurance, against the clinical reality that being allocated to this interval means that there is a high risk of developing STDR. Additionally, there was conflation about the purpose of eye screening, where it was commonly perceived to be a preventive measure against DR. For HCP, the shorter interval was welcomed to monitor high risk patients.
Acceptability of a 12 month screening interval with conditions
As the eye screening service has been in place for over 10 years in England and Wales, it was perhaps unsurprising that some PWD felt that annual screening was acceptable and should remain in place. This was often related to participants’ positive experiences of attending eye screening. Any changes made to their screening interval was felt by some PWD to be up to their HCP to decide upon, as shown below.
I am quite happy with that. Some people may need close screening but I am quite happy with the 12 months. If they wanted to see me more frequent or less frequent, I would just go along with it. (Sheena PWD, Phase 1)
The reliance upon HCP to decide which interval to allocate a PWD was mentioned by several of the participants. For other PWD, the annual eye screening appointment was mistakenly perceived as a reassurance and safety net for any changes in the eyes during this time period, illustrating the misunderstanding of the relationship between diabetic health, DR and screening.
I’d rather be seen every 12 months to be honest. I think people should be seen even if they’re classed as very low risk. I think even within a year a lot can change. You can suddenly have a bout of you know, I don’t know if having a bout of problems with your sugars would affect your eyes but yes, I just think every 12 months it should be. (Joanne PWD Phase 1)
Whilst some reported feeling reassured by annual screening, there were misunderstandings about the impact of diabetic health upon eye health and the role of screening. As the current annual screening interval is established and embedded into practice, it was foreseeable that PWD felt that this was an appropriate length of time for their eye screening. However, discussions also highlighted misunderstandings about the purpose of eye screening, as a preventive measure against the development of DR, and related to diabetic control.
Acceptability of 24 month screening interval with conditions
Extending screening intervals to 24 months provoked the most reaction and responses amongst PWD. The range of views included an unequivocal rejection of a 2 year interval as illustrated below.
No way! (Melanie PWD Phase 2)
Other PWD were more nuanced in their responses with concerns about this interval being too long a time period to go without being seen within the eye screening programme.
It’s an awful long time, 24 months, isn’t it? (Jean PWD Phase 1)
There were allied concerns about this interval around the potential for changes to the eye over two years and not being screened.
I feel like leaving something for two years can be very risky, because someone could always, all of a sudden be in a low risk and then take a turn for the worst and have like their eyes get really bad, really quickly due to something else. I feel like six and 12 months is good but then I don’t think 24 months is good, I think it’s too long because you wouldn’t leave someone who had diabetes for two years and not check their HbA1c, so why would you do it for their eyes? (Polly PWD Phase 2)
For other PWDs, being assigned to the 2-year screening interval was a positive reflection of their diabetes control.
I thought well if I don’t need it doing every 12 months then good, send it to two years. And I didn’t think anything bad about it. I suppose I was quite positive about it really. I thought it was working in my favour if it was going to last two years ... not having another appointment to go to ... I always think, if they extend your visits it means you’re on a level playing field you know, things are going smoothly; that’s the way I look at it. (Mary PWD, Phase 2)
For HCP, the move to 2-year intervals was welcomed as they highlighted that with annual screening they have to screen very large numbers of negative patients in order to identify a screen positive patient and that this seems like an inefficient approach.
I think looking to change the intervals makes sense to me a lot because an awful lot of the screening we do is, there is nothing there. And, even the ones who have mild background retinopathy you see little or no progression over several years…And you are seeing an awful [lot] of patients who are either having no retinopathy whatsoever or very, very mild retinopathy. (Gerard HCP)
For HCP, extending the screening interval would enable the better targeting of resources and would benefit patients who, for example are difficult to engage and often do not attend for screening, or who are at higher risk of developing STDR.
Unless we have the resources to follow all these patients [non-attenders] up, which if we do go to two-years screening we probably would. (Frankie HCP)
Similarly to PWD, HCP were concerned about extending screening intervals for the potential negative impact upon patient behaviour, namely it would affect risk perception around eye screening attendance. HCP anticipated that some patients would interpret extended intervals to mean that eye screening is considered not essential and there would be a concomitant effect upon an increase in non-attendance.
If you give someone a two-year appointment, they are probably thinking, well it can’t be that important if I don’t have to come back for two years. (Sami & Suzanne HCP)
Some HCP drew upon their clinical experiences to support an argument about their unease on extending screening intervals being at odds with their embedded narrative to PWD of needing to be screened annually and related to the trust and relationship between PWD and HCP.
I just don’t agree with it. We have spent, well I have spent the last 11 years drumming it into patients how important it is to be screened every 12 months, and now this is just going against everything I have been saying. And 12 months is a long time, and serious, serious damage can happen in them 12 months, even if they have had nothing in the past, I have seen it so many times. So I just don’t think it is worth the risk of moving a patient to 24 months. (Judith HCP)
There was a concern that PWDs’ trust in the eye screening services would be undermined by any changes to screening intervals along with the potential development of DR. In light of previous comments about the embedded nature of the current annual screening programme for PWD, any changes were thought to require careful communication and management.
Additionally, with a potential increase in non-attendance, there were concerns about the length of time a patient would go without being seen in the screening service and the possible impact upon a patient developing DR and the related costs to the NHS.
What would happen if they DNA if we went onto the two-yearly intervals and they DNAd that two-yearly one? It would be four years then. And that would be more expense wouldn’t it towards the NHS, I think that would cost more because we would have more things going wrong. (Janine & Hannah HCP)
For PWD and HCP, there were a range of responses to extending screening intervals to 2 years. For some PWD, an extension was welcome as it reflected good diabetes self-care, contrasted with outright rejection for others over concerns about developing eye disease in the extended time period. For HCP, 2-year intervals were acceptable in the context of many patients having minimal or no disease. However, there was some apprehension about the perceptual impact upon patients of changing screening intervals, with PWD feeling that screening was not as important if changed to a 2-year interval.
Safety of the Risk Calculation Engine (RCE):
Many of our participants (both PWD and HCP) indicated that they would be supportive of the introduction of risk-based allocation to variable screening intervals, on the condition or expectation of particular safeguards or service enhancements being introduced. For HCP, their concerns were focused on the safety of the RCE, specifically, around the quality and availability of data from different areas of health care services (primary and secondary care in the UK) into the RCE and any subsequent
One of my concerns would be getting the right information about the patient from the GPs, from hospitals. (Sandy & Liz HCP)
This issue of information quality and access was mentioned only briefly by a small number of PWD participants, as might be anticipated given their limited exposure to NHS information systems.
My only slight concern would be how up to date would the information in the (risk) engine 1 if you like would be, in terms of making that decision. And would it be an annual decision that the software made? (Kevin PWD, Phase 2)
As illustrated above, understanding when the RCE would calculate a PWD screening interval was seen as important. The ISDR RCE calculates the screening interval every time a PWD attends an eye screening appointment; PWD in low and medium risk groups who do not attend are assigned to 12 months for their next invitation and those in the high risk group to 6 months. Some HCP were concerned that the increased complexity of the RCE and subsequent screening allocation could create increased RISK for patients, implying that the mix of data and systems could result in incorrect calculations of risk and allocation to the wrong screening interval.
It is a more complex system, more complex recipe so there may be more opportunities for it to go wrong.” (John HCP)
The risk engine uses five unconnected data record systems extracted from primary and secondary care, and the screening programme, all with different administrative teams and access/governance arrangements. Data are screened and cleaned through bespoke processing. Risk is then calculated by a chain Markov model using 6 covariates.15
PWD had similar questions to HCP around the decision-making involved in allocation to screening intervals. They specifically asked about the process of the RCE and how it is constructed.
Who decides your risks? That’s what I’d like to know. (Arthur PWD, Phase 2)
In addition, PWD and HCP wanted to be able to self-refer, or refer patients back into annual screening if their ‘risk-factors’ changed between extended screening intervals, as explained below
I would feel more confident if there were safeguards where I can say, well the nurse can say, oh this is a bit erratic, we will need recourse to the testing place and see if we can get you a quicker appointment… ... in theory if everything stays the same there is no problem with me having the test every three years… so long as there are contingencies in place. If I was confident about safeguards I would be quite happy. (Derek PWD, Phase 1)
There were calls from PWD for assurances that the recall system would need to mitigate against any diabetic changes which would warrant an earlier recall to eye screening.
In theory, I would be all right as long as my reading stayed the same. So I suppose if my readings went high, and my sugar levels went high, I could say to the nurse well ok, I've got to have my eyes done now ... If my sugar levels go up for some reason, and I can’t control them you know, I could have my eyes done. (Jane PWD, Phase 2)
Similarly, HCP wanted assurances that the risk engine would be robust in identifying and inviting all PWD to be screened. Whilst recognising that the RCE was complex and sophisticated, HCP stressed there had to be obvious checks and balances of the system, rather than relying on computers and software.
There should be some kind of backup where at least there is somebody in the real world who is actually ensuring that things have not gone really haywire. (Sami & Suzanne HCP)
A further safety concern on extending screening intervals was the potential of missing STDR and the subsequent risk of patients developing visual impairment, as expressed below by a HCP.
The danger is the longer you leave a recall of course, the more chance you have of missing that occasional patient, so it is cost isn’t, it versus benefits really. And also once you have missed that patient, then trying to deal with them is more expensive. (Mark HCP)
These concerns were echoed among some PWD who considered that extending eye screening intervals was considered risky as illustrated below.
How do I know nothing is going wrong in all that time? (David PWD, Phase 1)
PWD willingness to accept longer intervals between screening episodes was often linked to their general diabetes care, such as regular monitoring of blood sugar levels, and liaising with primary care HCP and eye screening services. So the greater the perceived risk, the less willing PWD were to support an extended screening interval.
Macro impact of changing screening intervals
Whilst there were many comments about the safety of the ISDR model and its three screening intervals, there were other more wide-ranging comments about the macro effect of changing screening intervals. For example, there was recognition by HCP that the current eye screening system would not be able to manage demands in light of the ever increasing numbers of PWD and the related future cost of screening
From a burden of health and competing priorities, the NHS finance, we probably would say, there is a recognition that this [risk-based variable-interval screening] is probably for the increasing diabetic population on an annual screen. (Alice HCP)
Whilst recognising the impact of increasing rates of diabetes on screening, there were some concerns voiced by HCP on their job security with the introduction of variable-interval screening.
…I think the primary thing everyone is worried about is their jobs. That is, because again we don’t know how many people are going to go to 24 months and how many people are going to go to 6 months, potentially it could you know, cut a lot of people off our list…we are quite concerned about our jobs. (Janine and Hannah HCP)
There were suggestions that the complexity of the variable screening may serve to disadvantage particular groups of patients. In particular, those groups who do not engage well with services and as a result are at higher risk of developing STDR.
It will certainly disadvantage this group that we don’t get. We have got to find some way of getting these young, you know, sort of 20s to 40s probably, and a little bit beyond. Because I think you give them an inch that you don’t need them to come for two years – we will never see them for longer. (Sandy & Liz HCP)
PWD also voiced similar views that risk-based variable-interval screening should enable better targeting of resources and would benefit patients who, for example are difficult to engage and often do not attend for screening, or who are at higher risk of developing STDR.
In the scenario suggested below, there is a recognition that the NHS has finite resources and as such they need to be allocated in a more effective manner.
We’ll have less demand on the service, therefore we’ll be able to do a better service for other people who need it. That’s my logic. I think it’s sensible to do ... If the evidence shows you that it’s feasible and worthwhile well it just makes sense to refine what you’re doing in a way which is more productive. It doesn’t jeopardise the patient, and it’s a better use of resources which are limited. Makes sense, ticks the boxes, doesn’t it? ... I’m glad it’s happening as a process; it needs to be done. (Sid PWD, Phase 1)
Other PWD voiced a suspicion that extending intervals between screening episodes for the majority of people with diabetes was financially driven. But instead of being redistributed to be more productive, the cost-savings were aimed at restricting patients’ access to health services.
I don’t know, if it’s like cost-effective you know, they’re saving money. You feel like they’re saving money to say we don’t want to see you for two years. In your mind you think it’s about the money, otherwise you’d be screening people every six months anyway. (Ray PWD, Phase 1)
Of note, was that some PWD participants imagined ways in which they could continue to have their eyes screened on a yearly basis, such as staggering other eye appointments, as expressed below.
If they said you only need it 12 months that will do because I have a second one in the optician anyway. (David PWD, Phase 2)
Such comments demonstrate misunderstanding about the rationale for different eye appointments and their purpose. Whilst the ISDR model re-calculates a PWD screening interval at every visit, the gaps in understanding of eye screening appointments are a significant issue in supporting PWD to manage all aspects of their diabetes and related care.
If I go to the optician and I can stagger those visits so one year it’s the diabetes test and the next year it’s the optician’s test, because the optician does look at the back of your eye, then that’ll be ok for me. (Becky PWD, Phase 2)
As mentioned elsewhere in this paper, there were many examples of confusion amongst PWDs about diabetes, eye disease and eye screening, and, as illustrated above, conflated health beliefs are unhelpful in managing any changes within the eye screening service.