Co-design workshops
Seven individuals attended the first co-design workshop, related to T2D, and eight individuals attended the second workshop, related to T1D. While it was not required for participants to offer these information, as the focus was on development of the data-sharing system, some did offer some personal information when asked introductory and ice-breaker questions. The available details are provided in Table 2. HCPs offered only basic information about themselves before offering their opinions of mHealth and data-sharing (Table 3).
Table 2. Demographics of T1D and T2D patient participants in both co-design workshops
Diabetes type _Patient#
|
Gender
|
Age range (yrs)
|
Duration of diabetes (yrs)
|
Reported technology used
|
Reported self-management foci
|
T1D_Patient#1
|
F
|
40-50
|
N/A
|
N/A
|
N/A
|
T1D_Patient#2
|
M
|
20-30
|
2
|
Apps, insulin pen
|
Physical activity, BG
|
T1D_Patient#3
|
M
|
50-60
|
N/A
|
Insulin pump, CGM, app
|
BG, physical activity, insulin, carbohydrates
|
T1D_Patient#4
|
M
|
40-50
|
38
|
Insulin pen, app
|
BG, insulin
|
T1D_Patient#5
|
|
|
N/A
|
App, insulin pen
|
BG, physical activity, sick days, insulin, diet
|
T1D_Patient#6
|
M
|
60-70
|
8
|
Smartwatch, insulin pen
|
Physical activity, insulin, BG
|
T2D_Patient#1
|
M
|
60+
|
N/A
|
Paper diary, app
|
BG, physical activity, diet
|
T2D_Patient#2
|
M
|
60+
|
N/A
|
BG meter, insulin pen, paper diary, app
|
Diet, medications (non-diabetes related), insulin, physical activity
|
T2D_Patient#3
|
M
|
60+
|
N/A
|
BG meter, app
|
Diet, physical activity, well-being
|
T2D_Patient#4
|
M
|
60+
|
N/A*
|
BG meter, paper diary
|
BG, input from doctor
|
*Participant stated that they were “in the introduction phase” of their diabetes
Table 3. Demographics of participating HCPs in both co-design workshops
Provider#
|
Gender
|
Age range (yrs)
|
Specialist#2
|
M
|
50-60
|
Specalist#1
|
F
|
60-70
|
Nurse
|
F
|
30-40
|
GP#1
|
M
|
50+
|
GP#2
|
F
|
30-40
|
GP#3
|
F
|
50+
|
While those with T1D and T2D differed in the specific answers they would like to gain from sharing their data, their motivations for using mHealth and the need for more frequent support from healthcare services were similar. Likewise, specialists and GPs expected different functionalities and specific information from PGD, yet both expressed frustration with not knowing how to relate to the different mHealth technologies and desired similar levels of engagement and responsibility from their patients.
Main themes identified
Across the workshops, the two main themes were: 1) mHealth technologies’ impact on patients and providers, with subthemes concerning both groups’ use of patient-gathered data and roles and responsibilities, and 2) data-sharing, with subthemes concerning Expectations of sharing and receiving PGD during consultations, when, what and how to share PGD, electronic health record (EHR) integration and concerns.
Theme 1: mHealth technologies’ impacts on patients and providers
As mHealth tools allow patients to monitor more than a single parameter, patients are more able to understand how internal and external factors together affect their health. “Diabetes doesn’t happen in a container. There are other things around it.” [Patient_T1Dpatient#3].
Subtheme 1A: Use of mHealth and patient-gathered data
Most participants with T1D stated that they used mHealth technologies to review historical self-registered data in order to find patterns in their data. One T1D participant stated, “Data collecting gives you experience…If you track the data and get a curve” [Patient_T1Dpatient#3]. Another added that “we can go back and look at similar situations or essential things. I rarely eat ice cream so I can go back and look at how much insulin I took then, and how [my BG] was afterwards.” [Patient_T1Dpatient#5]. Patients concluded that this helped them to “spend less time and energy on self-managing” [Patient_T1Dpatient#2].
Similarly, participants with T2D stated “I use [PGD] for control” [Patient_T2Dpatient#3] and they found benefits when “evaluating the measurements - the results for stress level, drinks, and such. I tried to find the causes for high blood glucose…to compare the days, sometimes a month sometimes three months between the evaluations” [Patient_T2Dpatient#2].
However, some were frustrated with the lack of support from the healthcare system, asking, “where is the course where I can learn as a patient? I take more responsibility for my own health when using mHealth tools…[and get] a better overview…But even though I know a lot…I want to know more and I want to do better” (Joint_T1Dpatient#1).
Providers agreed that mHealth technology should aid the patient first in their own self-management, that the “use of technology needs to create patient action.” [HCP_Specialist#1]. Especially given that healthcare personnel struggle to provide detailed recommendations in such a short period of time, all seemed to agree that, “We want these sort of [patient-gathered] data to be self learning technology” (HCP_Specialist#1).
Subtheme 1B: Roles and responsibilities
Specialists asserted that even before they could consider discussing patient responsibilities they are making judgements about patients’ ability to use the technology, “we have to start differently and expect differently from our patients. So this is about individualization of treatment” (HCP_Specialist#1). Especially regarding the use of mobile health devices, one specialist stated that, “For some individuals, I put a [continuous glucose monitor] on them and say “you are supposed to use it” while others, I keep it away from them” (HCP_Specialist#2), because not all technology is useful or appropriate for everyone.
After establishing that they would be focusing their discussion on the specific subset of patients who use mHealth technologies, HCPs stated that they expect patients “to make the appointments, and to bring some own generated data” (HCP_Specialist#2) and then to “be prepared for the consultation. Because we have so little time” (HCP_Specialist#1). However, if the patient has not reviewed or attempted to understand their own-gathered data before sharing it with the doctor, “then it’s useless” (HCP_Specialsit#2). The expectation of the HCPs was that patients are to be active in using and understanding the data they generate to foster the discussion because of the limits of their time to interpret the volume of individual data that can be generated.
Both GPs and specialists commonly emphasized, “We can’t change anything, we can just give advice. The data by itself needs to help the patients to do the best thing…because our task is to be advisors” (HCP_Specialist#1). Another expressed that, “they are our pupils, and we’re their teachers and when they do homework, of course I want to see what they’ve done. And then…I can begin to give some advice” (HCP_GP#3).
Patients saw that their “role [in sharing data] could be to be more exact in documenting information, such as diet, physical activity and everything that can help the GP confirm where I am in the process” (Patient_T2Dpatient#2). Others acknowledged, “you have to take responsibility for the things not being done by healthcare…you have to follow up yourself” (Patient_T1Dpatient#3). This difference in role from passive data-collectors to active interpreters similarly reflects the concerns of the HCP above in the different understandings and roles that patients may adopt which may limit or support the use of technology in consultations.
Regarding their expectations of HCPs, all patients wished for more detailed, individualized feedback. For example, “more specific answers on situations and questions when I am in meeting with the nurse. I have sometimes questions about different situations… two similar situations can become two completely different ones. [And the nurses] never has any good answers” (Patient_T1Dpatient #5). More importantly, was the immediacy of their concerns, urging that they “wish for answers right then, not three months later” (Patient_T1Dpatient #3).
However, participants with T2D believed that GP’s “really lack the knowledge in which we diabetics struggle with [because they] do not have enough education to cope with those specific health issues” (Patient_T2Dpatient#2). One patient considered alternatively that, “there are also diabetes nurses…they can maybe give more input about what you should do and not do…Let the doctor take the more serious, while nurses help along the way” (Patient_T2Dpatient #5). Here, by demonstrating their different expectations toward various HCPs, patient participants also highlighted that they value the HCPs as supporters and educators in diabetes care. However, they also note that this value is based upon HCPs’ understanding of the patients’ everyday reality of living with diabetes.
Theme 2: Data-sharing system
Subtheme 2A: Expectations of sharing and receiving PGD during consultations
During the discussion of participants’ ideal data-sharing systems, one patient noted that, “maybe [HCPs] can help me more if they see that there’s a reoccurring problem…if I’m high during the evening...we can try to talk more specifically” (Patient_T1Dpatient #4). Based on their own experience, another patient noted that, “the [HCPs] could interpret data with the knowledge they have and then give specific tips and feedback on the data….I do feel I get specific tips with things [the HCPs] extract through the data which I don’t feel like I see myself. I’ve gotten advice that works” (Patient_T1Dpatient #2).
Two GPs echoed one another with regards to the fact that data would be easier to understand and interpret if the data were presented in a summative form that provided specific and relevant information first. “If the patient comes with [PGD] and it is easy to understand” (HCP_GP#2), because “specific information will save us time. Instead of trying and trying to make people tell us” (HCP_GP#1).
Subtheme 2B: When to share data
Patients and providers disagreed about when was an appropriate time to share PGD. Patients believed, for example, that “If I could say that now I’m struggling with something, and question if you [the HCP] could connect up and see the data…easier than booking an appointment” (Patient_T1Dpatient#2). The alternative was to have “data already be shown for the doctor before a consultation, so we have a baseline….To not waste time” (Joint_T2Dpatient#1).
GPs, on the other hand, were the most vocal about patients being prepared and presenting their data in-person during the consultation, otherwise, “without the patients, it is not useful. The patients should be there to use it for discussion and planning…knowing what’s going on” (HCP_GP#2). One GP asserted that they did “NOT want [PGD] in advance [of the consultation]” (Joint_GP#3). For the GP participants in general, data on its own was of limited meaning without the patient being there to provide context to aid interpretation.
However, during the T2D joint session, all agreed that data was important to share, for example, “if [the situation is] suddenly changing, the values go up or down, their health situation is getting worse or something, it could be okay to have more values, to see what is actually happening” (HCP_GP#1).
Subtheme 2C: What and how to display data
Patients were aware that not all data was equally as relevant: “I can collect irrelevant data - I can gather data about my own situation that may not be relevant for doctors” (Joint_T2Dpatient#4). In terms of specific data types, specialists wanted PGD that reflected “an intensive period [worth of data], maybe some days or weeks before they come to me because I want to see variation. And document pretty carefully…Then we can see the context…So these very like, these short, tiny, detailed periods is very valuable for me even if it’s not representative for the long life” (Joint_Specialists#1).
Suggestions from both patients and providers were similar in that they would like a system that summarized the PGD, with the option of choose which data to explore further when trends or outlier points were identified. One specialist “would first like to see the fluctuation over 24hours - it’s the most important for me. Then have a look at some data because there was something special going on” (HCP_Specialist#1).
The other specialist presented a system that showed trends for “the last week or 14 days…where you can see meals, calibrations - to see that you calibrate correctly - physical activity and illness…to explain why you are high the whole night, and of course insulin doses. Additionally, if the algorithm can pull statistics and say “ok, you are always low after correcting extra” or such things” (Joint_Specialist#2).
GP’s were more varied in their paper-prototypes. Some illustrated a “type of compressed summary...Instead of having to look at a thousand measurements” (Joint_GP#3). Others “tried to get in everything at once [to] see a correlation if you have [different PGD] together…You won’t bother to plot it every day, but rather if it was something special, [you’d have] a marker of some sort” (Joint_GP#1).
The paper-prototypes designed by both groups of patients, with T1D and T2D, illustrated their wish for feedback about their progress, for example, suggesting “we could get a sign on the graphs…maybe statistics on how the blood glucose is…in the evenings or afternoons” (Joint_T1Dpatient#5). Patients also wished to be able to explore and understand their own data first by “just having summaries of the data, and then click on blood to get [more details]…what you’ve done that day and time and all. Everything in a submenu of the main” (Joint, T2Dpatient#2).
Subtheme 2D: Electronic health record integration
Different types of HCPs preferred different ways of accessing and integrating the data into their everyday practice. Specialists were open to “automatically getting the continuous glucose values for the last week, into my electronic diabetes journal system... and the use of insulin or automated data easily, visually presented” (HCP_Specialist#1). However, GPs argued that, “we don’t need to load into the EHRs, because there are many problems and overload of information. And, why should we keep it? (HCP_GP#3). Instead, GPs preferred “to type [notes about PGD] myself…write it short. Reminder [to focus on this] for next time” (HCP_GP#3).
Subtheme 2E: Concerns
Despite participants’ optimism and the potential that they saw with sharing PGD, HCPs consistently noted their concerns that “it’s always a chance of overload…a whole lot of data. We can’t relate to it” (HCP_GP#1). One specialist's thoughts of automatic data-transfer into the EHRs was “please stop it! Because if you are the patient and I get your data continuously for your whole life on my screen, then if you have a traffic accident with hypoglycaemia it’s my responsibility because I should have seen that last week you had several hypoglycaemias” (HCP_Specialist#1).
Of particular concern for specialists was resource management. Continuous data transfer potentially meant the need “to have a diabetes nurse continuously, 24 hours-a-day, checking on continuous glucose monitors (CGMs) like we do with hospital patients. We don’t have resources for this” (HCP_Specialist#1). And for patients with CGMs “10% use the CGM. Then those patients get much more consultations with the nurses because they need to be taught the CGM, they need follow-ups. So this small group…maybe they use 80% of the nurses’ time” (HCP_Specialist#2).
Even patients were concerned with data-overload because when they have “a lot of data…I get little use out of meeting diabetes nurse. Last time she said it wasn’t much she could help me with” (Patient_T1Dpatient #3).