Real-world lived experience of older adults with type 1 diabetes after an automated insulin delivery trial
Abstract
Aims
First-generation closed-loop automated insulin delivery improves glycaemia and psychosocial outcomes among older adults with type 1 diabetes in clinical trials. However, no study has previously assessed real-world lived experience of older adults using closed-loop therapy outside a trial environment.
Methods
Semi-structured interviews were conducted with older adults who were pre-existing insulin pump users and previously completed the OldeR Adult Closed-Loop (ORACL) randomised trial. Interviews focused on perceptions of diabetes technology use, and factors influencing decisions regarding continuation.
Results
Twenty-eight participants, mean age 70 years (SD 5), were interviewed at median 650 days (IQR 608–694) after their final ORACL trial visit. At interview, 23 participants (82%) were still using a commercial closed-loop system (requiring manual input for prandial insulin bolus doses). Themes discussed in interviews relating to closed-loop system use included sustained psychosocial benefits, cost and retirement considerations and usability frustrations relating to sensor accuracy and system alarms. Of the five participants who had discontinued, reasons included cost, continuous glucose monitoring-associated difficulties and usability frustrations. Cost was the largest consideration regarding continued use; most participants considered the increased ease of diabetes management to be worth the associated costs, though cost was prohibitive for some.
Conclusions
Almost 2 years after completing a closed-loop clinical trial, closed-loop automated insulin delivery remains the preferred type 1 diabetes therapy for the majority of older adult participants. Chronological age is not a barrier to real-world successful use of diabetes technology. Identifying age-related barriers, and solutions, to diabetes technology use among older adults is warranted.
Novelty statement
What is already known?
Closed-loop automated insulin delivery improves psychosocial outcomes and glycaemia among older adults with type 1 diabetes during clinical trials.
What this study has found?
Among older adults with long-duration type 1 diabetes, real-world automated insulin delivery use can be sustained long term and improves quality of life, feelings of independence and reassurance in glucose management. Affordability, retirement considerations, and usability issues were prominent, impacting automated insulin delivery usage outside trial environments.
What are the implications of the study?
Older adults' lived experience can be improved with long-term automated insulin delivery in the real world. Research and advocacy are required regarding automated insulin delivery cost, particularly for older adults approaching or in retirement.
1 INTRODUCTION
Older adults with long-duration type 1 diabetes are at high risk of hypoglycaemia and its sequelae.1 Within the past 2 years, two robust studies have provided the first randomised trial evidence for the glycaemic benefit of closed-loop partially automated insulin delivery among older adults with type 1 diabetes.2, 3 These trials addressed a notable age-specific gap in the evidence base for closed-loop glucose control, highlighting clinical benefit for older individuals. End-of-study interviews with participants in the OldeR Adult Closed-Loop (ORACL) trial found themes of improved quality of life that reflected the glycaemic benefits.4 To what extent these positive outcomes persist into long-term, real-world outcomes for older adults with type 1 diabetes is currently unknown. What is known, however, is that use of closed-loop systems does decrease over time among children and younger adults with type 1 diabetes.5, 6 Several reasons for discontinuation have been identified from these studies, such as sensor issues, accessibility of consumable components, fear of hypoglycaemia and incompatibility with exercise. While newer systems will address some, if not all, of these issues, there are no published data on the rates or reasons for discontinuation of closed-loop automated insulin delivery among older adults. Moreover, it cannot be assumed that older adults' experiences of closed-loop systems match those of younger cohorts.
The present study included participants of the ORACL randomised trial, completed in 2020, which showed closed-loop therapy improved key glucose metrics among older adults with type 1 diabetes when compared with sensor-augmented pump therapy.2 During the trial, which had intervention with Medtronic MiniMed 670G systems, participants were positive overall about closed-loop system use and were aware of and pleased with treatment effects, including consequential psychosocial benefits.4 By conducting follow-up interviews with these former ORACL participants, the aim of the present study was to explore perceptions of closed-loop system use after trial-end, reasons for discontinuation and overall attitudes and feelings among older adults towards using closed-loop therapy for type 1 diabetes management.
2 METHODS
2.1 Study design and participants
At the time of OldeR Adult Closed-Loop (ORACL) trial enrolment, eligible individuals were aged 60 years or older, with type 1 diabetes duration ≥10 years, and using an insulin pump without closed-loop automated delivery.2 During the trial, all participants undertook two 4-month stages, closed-loop therapy and sensor-augmented pump therapy, in a randomised crossover design. All 30 participants who completed the ORACL trial were then invited via email to take part in post-trial semi-structured interviews. Recruitment and data collection were undertaken by EK as research for her clinical psychology doctorate (supervised by ST and SAM). This research was approved by the St Vincent's Hospital Melbourne Human Research Ethics Committee (HREC 275/18), conducted in accordance with Australia's National Statement on Ethical Conduct in Human Research and recorded in line with COREQ guidelines.7
2.2 Post-trial semi-structured interviews
Post-trial semi-structured interviews were conducted via videoconference with all participants who accepted the invitation. Informed consent was obtained prior to each interview. Each participant's current diabetes therapy was recorded, alongside their retirement/employment status and residential postcode. The postcode data allowed for socio-economic status and remoteness of their location to be estimated from the Australian Bureau of Statistics database; specifically, the Index of Relative Socio-economic Advantage/Disadvantage (IRSAD) and the Remoteness Index. The IRSAD compiles census data for a specific postal code region, encompassing factors that relate to both economic advantage and disadvantage including income, education and unemployment.8 A higher score on this index represents a greater level of socio-economic advantage and was represented as tertiles. Postcodes were also used to identify which of the five classes of remoteness the participant resided in (Major Cities, Inner Regional, Outer Regional, Remote and Very Remote).9
The interview topic guide was informed by a literature review and consultation with trial investigators (Table 1). The purpose of the interviews was to examine participants' current diabetes management and explore the factors that influenced their decision making regarding continued use of closed-loop insulin delivery. Lived experience of using closed-loop technology outside of a closely monitored trial environment was elicited, offering participants the opportunity to share their experience in their own words. Interviews were conducted by a psychologist with training in qualitative research (EK).
| Have you continued using closed-loop therapy since completion of the trial? |
| What system are you currently using? |
| What factors led to that decision? |
| What have you learnt works well for you when using a closed-loop system? |
| How could the closed-loop system be improved? |
| Do you recall how you felt emotionally during the trial using the closed-loop system (670G in automatic mode)? |
| What would you say to other people considering using closed-loop insulin delivery? |
2.3 Data analysis
An inductive thematic analysis approach was employed to analyse the semi-structured interview data, using procedures outlined by Braun and Clarke.10 The de-identified interview transcripts were imported and coded in Quirkos qualitative analysis software (version 1.4.1; Quirkos Limited, Edinburgh, UK). Repeated reading and familiarisation of the content was performed before interview transcriptions were thematically coded and counted. Researchers EK and ST consulted regularly to review themes and examples, using the live collaboration functionality of the Quirkos Cloud platform. Adopting an iterative process, the researchers discussed and compared codes during the analysis period, resulting in the coding framework being regularly updated. During this phase, weekly meetings were held between EK, ST and SAM to facilitate theme development and interpretation. Consolidation of themes occurred once updated codes were applied.
3 RESULTS
3.1 Participants and interviews
Interviews were undertaken from February to September 2022, inclusive. Of the 30 participants invited to be interviewed, one declined due to being on holiday, and another did not respond to the invitation. There were 28 participants who accepted the invitation and engaged in interviews—18 women and 10 men, with mean age 70 years (SD 5) and type 1 diabetes duration 40 years (15; Table 2). Over 80% of the participants resided in a socio-economically advantaged location and lived in a metropolitan area. These post-trial interviews were conducted at median 650 days (IQR 608–694) after trial completion; median run-time was 23 min (18–30). At the post-trial interviews, all participants reported current insulin pump therapy use, with 26 (93%) reporting current continuous glucose monitoring (CGM) use. Their insulin pump types were MiniMed 670G/770G (n = 22); MiniMed 780G (n = 1); and t:slim X2 (n = 5). CGM types were Medtronic Guardian Sensor 3 (n = 20); Dexcom G6 (n = 4); and Abbott Freestyle Libre 2 (n = 2). 23 participants (82%) reported current closed-loop therapy use at interview: 18 were using systems incorporating the first-generation Medtronic commercial algorithm (MiniMed 670G/770G); one was using a system with the second-generation commercial Medtronic algorithm (MiniMed 780G); four were using the Tandem Control-IQ automated system.
| Participants (n = 28) | |
|---|---|
| Age (years) | 70 (5) |
| Sex (female) | 18/28 (64%) |
| Retired from the workforce | 21/28 (75%) |
| Type 1 diabetes duration (years) | 40 (15) |
| HbA1ca | |
| mmol/mol | 59 (9) |
| % | 7.6 (0.9) |
| Currently using closed-loop therapy | 23/28 (82%) |
| History of microvascular complicationsa | 15 (54%) |
| History of macrovascular complicationsa | 5 (18%) |
| IRSAD | |
| Tertile 1 (most disadvantaged: score ≤953) | 22/28 (79%) |
| Tertile 2 (score ≥954 and ≥1016) | 6/28 (21%) |
| Tertile 3 (most advantaged: score ≥1017) | 0/28 (0%) |
| Remoteness Index | |
| Major cities | 21/28 (75%) |
| Inner regional | 7/28 (25%) |
| Outer regional, remote or very remote | 0/28 (0%) |
| Highest education level completed | |
| University | 14 (50%) |
| Diploma | 6 (21%) |
| Secondary school | 4 (14%) |
| Year 10 or equivalent | 2 (7%) |
| Primary school | 2 (7%) |
- Note: Data presented as n (%) or mean (SD). Data were collected at the time of interview unless otherwise stated.
- Abbreviation: IRSAD, Index of Relative Socio-economic Advantage and Disadvantage: a higher score indicates greater level of socio-economic advantage encompassing factors relating to income, education and unemployment.
- a Data collected at the first ORACL trial visit.
It is important to note that when most of the interviews were conducted, Australia's National Diabetes Services Scheme (NDSS) would only provide subsidised consumables for people under the age of 21 years or people eligible for a means-tested health care card concession. This meant that for the majority of our participants, who were not eligible for concessions, consumables such as glucose sensors were not subsidised. In July 2022, the Australian Government expanded eligibility criteria for CGM product subsidies to all people with type 1 diabetes.11 As 22 out of the 28 interviews were conducted before July 2022, the cost concerns and resentment of reimbursement only to young people were shared by participants in the context of not yet being eligible, evoking feelings of discrimination and highlighting the importance of equitable access to diabetes technology for people of all ages and socio-economic backgrounds.
3.2 Factors influencing therapy choice
There were five participants who had discontinued closed-loop therapy. Reasons recorded for discontinuation were dissatisfaction in glucose control (n = 1), frustration with alarms and lag in glucose correction (n = 1), difficulties using CGM due to osteoarthritis (n = 1) and costs not deemed worth the benefit (n = 2). There were four participants who had changed from MiniMed systems to t:slim X2 systems since the trial. Reasons for changing systems included less finger-pricking required with Dexcom G6 sensors compared with Medtronic Guardian Sensor 3, the perception that the t:slim system allowed for tighter glucose control and had fewer alarms, and the t:slim X2 being physically more compact.
3.3 Themes
To aid interpretation and understanding of the interview results, the authors developed four over-arching themes, each with sub-themes, drawing out the main response points from the interviews. These four themes are: impact of continued closed-loop therapy; life beyond diabetes with closed-loop therapy; financial considerations when ageing with type 1 diabetes; frustrations with closed-loop use in older age. Themes and subthemes are presented below, with illustrative quotes presented in Table 3.
| Theme | Example quotes |
|---|---|
| Theme 1: Impact of continued closed-loop therapy use on living with type 1 diabetes | |
| Sustained glucose benefits post-trial |
‘I think it just seems to keep me better controlled. I don't know greatly the ins and outs of it or how it works, but it just seems to work for me. It worked on the trial, and it's worked since’. ID26, current automated insulin delivery (AID) user (MiniMed 670G) ‘It's actually about control, it is about having the level low, but not too low and the automatic correction … you don't feel good when your sugar's high. It's sub-optimal. It doesn't have to be very high for you not to feel good about it. It is about being woolly. Like not being able to make decisions or not be able to see things clearly and act on them and well-being’. ID15, current AID user (Control-IQ) |
| Reduced psychological burden post-trial | ‘It does the adjusting in other words. There was one day that I had a shower and I had forgot to put it back on and I'd gone out somewhere and I was having a meal and I had realised I didn't have it. Then it went back to I had to give myself the injection because I always had my pen with me. It took me back years. I think, “Oh God, where's the hazard badge?” Having to calculate and all of that’. ID4, current AID user (MiniMed 770G) |
| Confidence in diabetes management post-trial |
‘Confidence in my disease… much more confident. I don't feel like I'm at risk of having a hypo. I was having far too many hypos before. I'd be in the ambulance before I knew what I was doing… having this control has just meant that I'm safer. I also live on my own now, it's really a security thing’. ID5, current AID user (MiniMed 770G) ‘At first, I was feeling that, here I am putting my life in the hands of a square machine and I'm wondering whether if it really knows what it's doing and stopped the data or whatever, so I basically double-checked all the time. Then the longer I was on it, the more confidence I gained … The pump isn't 100% accurate but at least it is fairly consistent and it's reliable so my confidence in the pump increased with time’. ID29, current AID user (MiniMed 780G) ‘I carry a spare syringe and an inserter in the car, but then I may not have the insulin that I need to fill it up with. Then you're out at lunch and you haven't got any insulin, and you think, because you carry less and less with confidence, I used to carry a pouch, but now, I don't carry that pouch anymore. In that pouch is an insulin pen and insulin and a spare vile. I carry less and less’. ID6, current AID user (MiniMed 670G) |
| Reflecting on previous diabetes management eras | ‘I've had [diabetes] since I was 12, so I grew up with it … Mum was very strict, so everything was measured. We had the little, tiny bit of cardboard and that was how much bread I could eat for a portion … testing the urine and boiling up your glass needle and using your steel needles and stuff. It was really old-fashioned stuff, but it's amazing how it's transgressed up and it's come to what it is now. It really is terrific. I've had it 62 years now, so I know what it's like’. ID13, NOT using AID at time of interview (MiniMed 670G pump and FreeStyle Libre 2 CGM) |
| Theme 2: Life beyond diabetes with closed-loop therapy | |
| Adjustments in daily life |
‘The best thing about it is that you can look at your pump and you can see the number, how you are, you almost become addicted to it. I remember at first, I was all the time checking my pump with the number or beeps and that, but you get used to them. You get used to what it's trying to tell you’. ID20, current AID user (MiniMed 670G) ‘When you were on injections daily, you didn't have any evidence of what the changes were, if you can see what I'm saying. It wasn't constantly in your thoughts all the time, you just had an injection and assumed things were going okay. Then you have your next one and so on, but with the closed loop, it's all there all the time. You can become a little bit obsessive about it. It's finding that balance between the benefits and not allowing it to take over your life’. ID12, current AID user (MiniMed 770G) ‘All those new things, young people just take it like that, but the older you get, the more you hate change. It's part nature, it just is. I'm also aware in my mind, “you're hating change, don't be old, just do it”. I force myself along the tracks and turn up’. ID6, current AID user (MiniMed 670G) |
| Improved relationships | ‘It [diabetes] has affected my relationship with my grandchildren because my daughter-in-law didn't really feel secure with me. Because I had so many hypos, she wouldn't let me mind the children… they've become more confident in me, which has been a bonus’. ID5, current AID user (MiniMed 770G) |
| Independence |
‘I live on my own and it is a worry if I was really bad here on my own. That's why the pump's good because it basically keeps me at a level where I can cope, whereas before I was going low or high or whatever. There's no doubt it's the way to go. There's no other way to go. It's the closest thing we're going to get to a pancreas, isn't it’. ID24, current AID user (Control-IQ) ‘Having the alerts on and what have you, and it telling you when you dropped low, that [is] a great bonus to me. Yes, because I've lost my husband … he would notice before I would notice. Now that obviously he's not here, because he died, it's a great benefit to me because it alerts me before I get to that point. When you're dropping low, it alerts, and then you now have to do something about it, which is a huge benefit for many because, if I don't have that on, then I'm not aware until I'm quite not good at all’. ID26, current AID user (MiniMed 670G) |
| Theme 3: Financial considerations when ageing with type 1 diabetes | |
| Benefit not worth the cost |
‘I was doing okay just by using the pump and I test myself quite a lot. I thought for the extra cost, it just wasn't kind of financially worth it… that's the main reason I think was the actual cost of the whole thing’. ID7, NOT using AID at time of interview (MiniMed 670G pump without CGM) ‘The prohibitive cost… I just don't think that there's value for money in it for me… If I can achieve better than seven with just doing finger pricks, then… I don't see that a closed-loop system would be of benefit to me in that regard’. ID14, NOT using AID at time of interview (Tandem t:slim X2 pump without CGM) |
| Impossible without monetary subsidies |
‘Oh, I couldn't do it without. I couldn't have afforded to have it if I didn't get it free from the NDSS. I couldn't do it’. ID24, current AID user (Control-IQ) ‘Otherwise, I couldn't do it. Not on a pension. There's no added cost to it. It's just the normal things like if you have any pump, it's the reservoir and the lead, the line. They're all supported by the government under the NDSS’ ID4, current AID user (MiniMed 770G) |
| Retirement considerations |
‘I think that it gets worse when you retire, because you have access to less money. If you are receiving superannuation, then you're not eligible for any health card benefits. That's an additional issue’. ID14, NOT using AID at time of interview (Tandem t:slim X2 pump without CGM) ‘Because you know that $350 bucks a month is a fair amount of money, and I'm not one of these people that would say, “Oh yes, I'm working now, so I'll spend the $350 per month. That's not a problem.” Then when I retire, I'll stop because I may not be able to afford it. No, I'm not that sort of person. Once you start something, you're not obliged, but internally obliged to keep going because that becomes your norm’. ID23, current AID user (MiniMed 670G) ‘One would be that I wanted to work less days or less hours to take care of my health and I haven't. I basically only just swapped to four days a week recently. That's a compromise and a balancing act because we don't know our date of death’. ID2, current AID user (MiniMed 670G) ‘We had to do it very tough … We have to just find the money and put it towards that. We might not have had a holiday, but the money has been there to pay for the sensors’. ID20, current AID user (MiniMed 670G) |
| Frustration and resentment regarding older adult population being disregarded |
‘It is expensive, there's no two ways about it … both my husband and I are retired. I used to get angry or upset that people up to 21, I think they get everything for free … I have been a diabetic since I was 14 years old. I've been through the period of five, six injections a day and all of that. They get it for free and then nothing for us, or if you're older than 21 or you're pregnant or planning a pregnancy. I used to get very angry; hey, diabetes doesn't go away at 21, it doesn't go away after you've had a baby’. ID20, current AID user (MiniMed 670G) ‘As is so often the case with diabetics, especially as you get older, and again, it depends on your medical issues, but you may need to require statins, you may need to require blood pressure tablets, you may need to require all sorts of other medications … All those costs add up to a lot of money … that is huge, that is a huge cost’. ID14, NOT using AID at time of interview (Tandem t:slim X2 pump without CGM) ‘I'm 72 now, how much longer have I got? If I'm lucky, 20 years, and if those 20 years are going to be much easier to manage because of the closed loop, then the government should be helping us out and preventing us going into hospitals to have our feet chopped off and our eyes fixed up’. ID10, current AID user (MiniMed 670G) |
| Theme 4: Frustrations with closed-loop use in older age | |
| Giving up control | ‘I think because I'm an older diabetic, and I know this trial was for senior citizens, but I think that all of us who have grown up with diabetes, that we were in control and handing over to an AI and letting it take control is disconcerting, because I think AI is still early days with so many things’. ID14, not using AID at time of interview (Tandem t:slim X2 pump without CGM) |
| Lag in corrections | ‘I find it's too slow these days… the closed-loop system is good for dropping low, but when you are high, it just takes forever to bring you back down’. ID9, NOT using AID at time of interview (MiniMed 670G pump and compatible CGM, used in manual delivery mode) |
| Alarms and increased finger pricking |
‘…honestly, the alarms just drive me mad, and I was pricking my finger more times a day than what I was when I first got diabetes at 56 years of age’ ID16, NOT using AID at time of interview (MiniMed 670G pump and FreeStyle Libre 2 CGM) ‘All the finger-pricking and that … my fingers are really sore at the moment. I've got osteoarthritis in them and I'm having a bit of trouble with that. I know you can use the palm of your hand and that sort of thing … but I'm not doing that’. ID13, NOT using AID at time of interview (MiniMed 670G pump and FreeStyle Libre 2 CGM) |
| Physical aspects |
‘There was a certain amount of fiddliness to it, and the fact you had to have two things attached to your body instead of one. I'm not using it. There's a lot of things to factor’. ID7, NOT using AID at time of interview (MiniMed 670G pump without CGM) ‘Well, it definitely puts that on your sex drive … having these two things hanging off you’. ID7, NOT using AID at time of interview (MiniMed 670G pump without CGM) |
3.3.1 Impact of continued closed-loop therapy use on living with type 1 diabetes
Maintaining glucose levels in target range that was achieved more easily with closed-loop therapy than with manual insulin delivery, both during the trial and post-trial period, was reported as the main benefit by the 23 participants still using closed-loop therapy. Participants reported feeling better knowing their glucose levels were ‘more normal’, with closed-loop therapy perceived to minimise their risk of long-term complications related to type 1 diabetes. Time spent worrying about glucose levels and how to manage them was reduced, with participants commenting on the capability to monitor glucose trends during closed loop being an important aspect. Many participants reported that their trust in the system continued to develop over time. Reduced concern about hypoglycaemic events during exercise was reported, as was greater flexibility in lifestyle due to less planning required. Feelings of support were described, knowing that company representatives from Medtronic and Tandem were contactable for troubleshooting assistance, adding to the support received from treating clinicians. Experience with the technology was reported as a factor contributing to satisfaction using closed-loop therapy, with longer time spent using closed-loop systems reported to increase comfort and confidence, particularly in relation to alarms. Comparisons between recent advancements in diabetes management technology and past diabetes eras (e.g. testing urine glucose, measuring food portions with pieces of cardboard) were raised; and there was deep appreciation for the ease of closed-loop therapy use, once participants had become accustomed to using the system. This theme was shared both by participants who were using closed-loop therapy, and by those who had discontinued use.
3.3.2 Life beyond diabetes with closed-loop therapy
Adjustments in daily life
The capability to visualise glucose levels in real-time was reported to be helpful; however, this often led to participants feeling ‘addicted to’ or ‘obsessive about’ checking glucose levels when adjusting to using CGM as part of their closed-loop system. Participants reported adapting to this over time as they became familiar with communication of the system, and as trust in alerts was built. Reports of changes in mindset when adjusting to the automation of insulin delivery in closed-loop use were described as challenging, but the benefits are now recognised as worth persisting through the often-frustrating initiation period.
Improved relationships
Family members were reported to feel more comfortable with these older adults living alone and looking after grandchildren, due to minimised risk of adverse glycaemic events when using closed-loop therapy. Participants reported being more inclined to dine out with friends due to the ease of glucose management offered by closed-loop therapy. However, some participants expressed feeling annoyed with pump alarms when they are around other people and described their discomfort calibrating CGM when in social settings.
Independence
Increased perception of independence was conveyed and considered to result from feeling confident in closed-loop system use. Participants expressed feeling that they could cope better with their day-to-day diabetes management whilst using closed-loop therapy, with some reporting they did not need to see their diabetes educator or dietician as frequently as before. Not needing to rely on third-party assistance to recognise symptoms of hypoglycaemia also improved feelings of independence.
3.3.3 Financial considerations when ageing with type 1 diabetes
Benefit not worth the cost
The cost of closed-loop therapy was a prominent concern for all participants, with insulin pumps and CGM not being subsidised in Australia at the time of most of these interviews. Some individuals determined the perceived benefits from the technology were not worth the cost. Nevertheless, these participants recognised they were ‘not doing as well’ as when they had been using the system. Cost was referred to not only as financial, but also as time, effort and potentially attending an increased number of clinical appointments during the initiation process of their closed-loop use. The chronic nature of type 1 diabetes was discussed and contextualised with how their long-term expenditure requires careful consideration. Notably, after Australian government CGM subsidies were announced in early 2022, effective from July 2022, participants expressed a desire to reconsider closed-loop therapy, once the costs of CGM were substantially reduced and became ‘one major thing out of the equation’.
Impossible without monetary subsidies
Participants reported not being able to afford closed-loop therapy if not for government subsidies. Costs associated with using closed-loop therapy in Australia were high for many individuals outside of clinical trials, and while waiting to become eligible for government subsidies, closed-loop therapy use was not possible due to the cost. Participants who were not using closed-loop therapy due to cost reported plans to resume when the subsidies were implemented.
Retirement considerations
The Australian retirement system is based on a combination of mandatory superannuation savings, a means-tested age pension and personal savings. People aged 55 years and over can access their superannuation subject to specific eligibility criteria. Age pension payments are available to people who are 67 years and older, meet residency requirements and are deemed eligible through income and assets testing. For individuals of this study who were currently working, retirement timing was an important consideration in their decision whether to utilise closed-loop therapy in the real world. Financial restrictions during retirement were a common concern, as was means-tested eligibility for treatment subsidies after selling the family home or beginning superannuation payments. Some individuals reported delaying retirement to afford closed-loop therapy. For all interviewees, uptake initially required careful consideration post-trial. Participants reported wanting to avoid starting closed-loop therapy before retirement and then needing to cease use due to monetary restrictions once entering retirement, before government subsidies were announced. Some participants reported working additional days to pay for CGM and delaying retirement due to associated costs.
Frustration and resentment regarding older adult population being disregarded
Comparing their financial situation to younger cohorts, the older adult participants voiced frustration and annoyance at the perceived lack of support regarding financial considerations for their age group. Recognition of the financial obligations associated with maintaining optimal health as they age with diabetes was frequently discussed in relation to the cost of closed-loop therapy; differences in the support they are given, relative to younger people with type 1 diabetes, were also highlighted.
3.3.4 Frustrations with closed-loop use in older age
Giving up control to the automated system was an adjustment that took time and development of trust to build for these older adults with long-duration type 1 diabetes who have lived through many eras of diabetes management. This adjustment period was described as frustrating and scary in some instances. Depending on the system being used, reports of lag in hyperglycaemia corrections and increased finger-pricking were also shared, as were frustrations regarding the sound of alarms, which led to cessation of closed-loop therapy for one participant. Physical aspects of closed-loop therapy, such as CGM attachments and operability of the system, were also reported to be frustrating.
4 DISCUSSION
This study explored the lived experience and perceptions of older adults regarding closed-loop automated insulin delivery systems more than 18 months after completing a closed-loop clinical trial, and the factors influencing their type 1 diabetes treatment decisions. Their accounts provide insider perspectives on using closed-loop systems under real-world conditions. The interviews highlighted sustained benefits for glucose management and psychosocial benefits; monetary and retirement considerations; closed-loop system usability concerns of older adults. Although similar themes have been elicited in previous clinical studies,4, 12-14 to our knowledge, this is the first to specifically explore the perceptions of older adults' use of closed-loop automated insulin delivery in the real world.
Previous studies have shown closed-loop therapy can have psychosocial benefits for older adults in a monitored trial environment.4, 15 Our research demonstrates that these improvements endure beyond the trial setting, with benefits in the real world with usual clinical care. This corroborates previously reported real-world data which found glycaemic control improved, and time spent in hypoglycaemia decreased, over a 3-month period.16 We also presented evidence of sustained long-term psychosocial benefits, previously being reported only among younger long-term users.17 Improved perceptions of independence, trust in the system and improved relationships persisted beyond the trial; participants reported improved relationships with family members who recognised the effectiveness of closed-loop therapy for reducing adverse glycaemic events and, in turn, trusted them to look after children and live alone. This high level of trust is a novel finding for this population, similar to findings among parents with children using closed-loop therapy and younger participants.18, 19 Participants, both during the clinical trial and in routine clinical care afterwards, reported thinking less about their diabetes. Similar findings have been reported among younger closed-loop system users and their caregivers.20 Despite concerns that older adults might struggle with closed-loop therapy,21 our findings show this cannot be assumed. An over-arching theme of persistence was noted; although frustrations and discomfort during familiarisation were described,4 those who persisted felt comfortable and confident with closed-loop therapy use nearly 2 years after the ORACL trial had ended. This suggests that long-term use of closed-loop therapy use can yield satisfactory glycaemic and psychosocial outcomes in older adults.
Only a minority of interviewed participants discontinued closed-loop insulin delivery after the trial. Reasons included prohibitive cost, dislike of alarms, lag in glucose corrections and health limitations, consistent with reasons recently reported in real-world evidence of closed-loop use among adults of all ages.22 Previous real-world data have shown use of closed-loop therapy substantially declines over time, with prospective observational studies of children and youth finding around 30% of participants discontinued closed-loop therapy (MiniMed 670G) over 3- and 6-month periods.5, 6 Messer et al. found a smaller attrition rate with 3.5% of youth discontinuing Tandem t:slim Control-IQ technology use over a 6-month period.23 One possible explanation for this could be reported communication difficulties between the CGM and MiniMed 670G pump24 or the frequency of finger-pricks for sensor calibrations required for this system, whereas finger-pricks are almost eliminated from routine use of Control-IQ. The number of routine finger-pricks with MiniMed 670G was reported by some participants who had ceased closed-loop as their reason for discontinuation. Other participants reported health concerns as reasons for discontinuation; interestingly, findings have shown that clinicians are reluctant to keep older adults on closed-loop systems if not confident in their ability to manage, resulting in advising older adult patients to go back to multiple daily injections.25 Understanding clinicians' perspectives on managing diabetes in older adults and their use of closed-loop systems is an area in need of further investigation. This research could inform usability considerations for future devices to accommodate the changing functional status of aging adults.26
Many participants expressed concerns about the increased financial burden of continuing closed-loop therapy after the trial, particularly in retirement. This aligns with the broader issue of cost being a major barrier to closed-loop therapy across age groups,27 despite evidence that it is more cost-effective than multiple daily injections or open-loop therapy.28 What is striking from these interviews is the intersection between increased treatment costs and retirement considerations. Being retired or transitioning to retirement is an important context for understanding their frustration regarding perceived lack of government support for closed-loop therapy costs. This economic barrier fuelled a sense of disappointment and discrimination that a superior diabetes management system was potentially out of their reach. Our interpretation of these interviews is that increased cost was the main deciding factor for the older adults who were not currently using closed-loop therapy. This aligns with clinician advice that individuals should only consider closed-loop therapy if they can cover the ongoing costs.29 It is notable that while transition to adult health care for youth with type 1 diabetes is well-researched,30 retirement transition for older adults with type 1 diabetes has not been explored.
In mid-2022, after more than three-quarters of this study's interviews were conducted, the Australian Government eligibility criteria for CGM product subsidies expanded to include people of all ages with type 1 diabetes.11 A previous Government CGM subsidy was introduced in 2017 for people aged under 21 years, resulting in increased CGM uptake and improved glycaemic control among young people with type 1 diabetes.31 However, unlike in the United Kingdom, subsidised access to insulin pumps is not available for Australians without private health insurance, thereby resulting in out-of-pocket costs above AU$2000 per year for pump therapy.32 Participants in this study were pre-existing insulin pump users and were interested in assessing the impact of CGM and automated basal insulin delivery.
Time was crucial for many participants when evaluating their experience with closed-loop systems. Participants acknowledged the utility of persistence and trust for long-term use, which aligns with findings among younger adults,33 and clinicians working with people using closed-loop systems.29 Alarms and finger-pricking required for closed-loop use were reported as annoying; however, attitudes towards pump alarms were more positive during interviews of the present study than those held during the trial.4 In the post-trial interviews, participants who were still using closed-loop therapy reported feeling comfortable when the alarms sounded due to growing recognition of alarm meanings and how to rectify issues the system was alerting to. Physical aspects of the pump were commented on, with participants reporting a need to get used to the attachments and ‘fiddliness’ of closed-loop systems. Although available now in Australia, G4 sensors were not available in Australia at time of interviews; this was significant as those using Medtronic CGM were using G3 sensors, which require routine calibration via finger-pricking. The G4 sensors do not require routine calibration, and it is expected such advancements in technology may address negative perceptions of older adults using closed-loop insulin delivery, making life easier for those with health complications such as arthritis.
This study has several strengths: To our knowledge, this detailed evaluation of older adults' perceptions of closed-loop use is the first to explore the lived experiences of older people using closed-loop therapy in the real world over the medium- to long-term. The timing of these interviews allowed for a more extended reflection period compared to previous studies. However, we acknowledge that this group is highly selected, all using insulin pumps and managing their diabetes effectively before joining the ORACL trial. Due to a lack of frailty status across the spectrum among the group at enrolment, the generalisability of our findings is limited and cannot be applied to the broader population of all older adults with type 1 diabetes. Additionally, the findings regarding cost and retirement considerations may not be relevant to individuals residing in countries with varying healthcare support structures. Lastly, the usability concerns relate to lived experience with first-generation closed-loop systems, and these issues are likely to be addressed in subsequent generations including models already in use.
Our real-world findings indicate that the lived experience of older adults with type 1 diabetes can be improved with long-term use of closed-loop automated insulin delivery, consistent with outcomes seen among younger people using such technology. Affordability, retirement considerations and usability issues remain potential barriers to technology access and use by older individuals. Further research into age-related barriers to diabetes technology, and focused solutions relating to costs, accessibility and usability is warranted.
AUTHOR CONTRIBUTIONS
Erin Kubilay, Steven Trawley and Sybil A. McAuley co-designed the study. Erin Kubilay conducted the interviews. Erin Kubilay and Steven Trawley analysed and interpreted interview data and wrote the report, with input from Sybil A. McAuley. All authors critically reviewed the report and made the decision to submit for publication. Sybil A. McAuley and Steven Trawley are responsible for the integrity of the work as a whole.
ACKNOWLEDGEMENTS
The authors thank the study participants and their families for their involvement in this project. The authors acknowledge support during the trial from investigators, research nurses, diabetes educators and dietitians at St Vincent's Hospital Melbourne and the Royal Melbourne Hospital, Melbourne, Australia. Open access publishing facilitated by The University of Melbourne, as part of the Wiley - The University of Melbourne agreement via the Council of Australian University Librarians.
FUNDING INFORMATION
This project was funded by the St Vincent's Hospital (Melbourne) Research Endowment Fund, and the ORACL trial was funded by JDRF (3-SRA-2018-667-M-R) and the Diabetes Australia Research Program. Medtronic supplied discounted insulin pumps and glucose monitoring devices for the ORACL trial. S.A.M is supported by a JDRF Research Award and University of Melbourne Paul Desmond Clinical Research Fellowship. The funders of this study had no role in trial design, data collection, data analysis, data interpretation or writing of the report.
CONFLICT OF INTEREST STATEMENT
S.T. reports non-financial support from Abbott Diabetes. S.F. reports receiving honoraria for lectures from AstraZeneca, Boehringer Ingelheim/ Eli Lilly, and Novo Nordisk; serving on advisory boards for Medtronic, Mylan, Pfizer, Sanofi and Viatris. S.A.M. reports support for research from Medtronic to her institution, speaker honoraria from Eli Lilly, Roche and Sanofi and serving on advisory boards for Medtronic and Ypsomed. All other authors declare no competing interests.
Open Research
DATA AVAILABILITY STATEMENT
De-identified data relating to this manuscript will be made available for research purposes for up to 12 months after publication on reasonable request by email to the corresponding author.
